<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE article PUBLIC "-//NLM//DTD Journal Publishing DTD v2.0 20040830//EN" "http://dtd.nlm.nih.gov/publishing/2.0/journalpublishing.dtd">
<article article-type="review-article" dtd-version="2.0" xmlns:xlink="http://www.w3.org/1999/xlink">
  <front>
    <journal-meta>
      <journal-id journal-id-type="publisher-id">JA</journal-id>
      <journal-id journal-id-type="nlm-ta">JMIR Aging</journal-id>
      <journal-title>JMIR Aging</journal-title>
      <issn pub-type="epub">2561-7605</issn>
      <publisher>
        <publisher-name>JMIR Publications</publisher-name>
        <publisher-loc>Toronto, Canada</publisher-loc>
      </publisher>
    </journal-meta>
    <article-meta>
      <article-id pub-id-type="publisher-id">v9i1e83031</article-id>
      <article-id pub-id-type="pmid"/>
      <article-id pub-id-type="doi">10.2196/83031</article-id>
      <article-categories>
        <subj-group subj-group-type="heading">
          <subject>Review</subject>
        </subj-group>
        <subj-group subj-group-type="article-type">
          <subject>Review</subject>
        </subj-group>
      </article-categories>
      <title-group>
        <article-title>AI-Based Monitoring Devices for the Identification of Circadian Parameters as Digital Predictive Biomarkers of Neurodegenerative Diseases in Older Adults: Systematic Review</article-title>
      </title-group>
      <contrib-group>
        <contrib contrib-type="editor">
          <name>
            <surname>Schütz</surname>
            <given-names>Narayan</given-names>
          </name>
        </contrib>
      </contrib-group>
      <contrib-group>
        <contrib contrib-type="reviewer">
          <name>
            <surname>Mitra</surname>
            <given-names>Sayan</given-names>
          </name>
        </contrib>
        <contrib contrib-type="reviewer">
          <name>
            <surname>Dhahbi</surname>
            <given-names>Wissem</given-names>
          </name>
        </contrib>
      </contrib-group>
      <contrib-group>
        <contrib id="contrib1" contrib-type="author" equal-contrib="yes">
          <name name-style="western">
            <surname>Herrera-Zamora</surname>
            <given-names>J Manuel</given-names>
          </name>
          <degrees>PhD</degrees>
          <xref rid="aff1" ref-type="aff">1</xref>
          <ext-link ext-link-type="orcid">https://orcid.org/0000-0002-6694-6902</ext-link>
        </contrib>
        <contrib id="contrib2" contrib-type="author" equal-contrib="yes">
          <name name-style="western">
            <surname>Monroy-Rodriguez</surname>
            <given-names>Jesus</given-names>
          </name>
          <degrees>PhD</degrees>
          <xref rid="aff2" ref-type="aff">2</xref>
          <ext-link ext-link-type="orcid">https://orcid.org/0009-0004-2334-3169</ext-link>
        </contrib>
        <contrib id="contrib3" contrib-type="author" equal-contrib="yes">
          <name name-style="western">
            <surname>Garcia-Mancilla</surname>
            <given-names>Jesus</given-names>
          </name>
          <degrees>MSc</degrees>
          <xref rid="aff3" ref-type="aff">3</xref>
          <ext-link ext-link-type="orcid">https://orcid.org/0000-0002-2104-8033</ext-link>
        </contrib>
        <contrib id="contrib4" contrib-type="author" corresp="yes" equal-contrib="yes">
          <name name-style="western">
            <surname>Santana-Mancilla</surname>
            <given-names>Pedro C</given-names>
          </name>
          <degrees>Prof Dr</degrees>
          <xref rid="aff4" ref-type="aff">4</xref>
          <address>
            <institution>School of Telematics</institution>
            <institution>Universidad de Colima</institution>
            <addr-line>Av. Universidad 333</addr-line>
            <addr-line>Colima, Colima, 28040</addr-line>
            <country>Mexico</country>
            <phone>52 3123161075</phone>
            <email>psantana@ucol.mx</email>
          </address>
          <ext-link ext-link-type="orcid">https://orcid.org/0000-0002-4184-0116</ext-link>
        </contrib>
      </contrib-group>
      <aff id="aff1">
        <label>1</label>
        <institution>University Center for Biomedical Research</institution>
        <institution>Facultad de Medicina</institution>
        <institution>Universidad de Colima</institution>
        <addr-line>Colima, Colima</addr-line>
        <country>Mexico</country>
      </aff>
      <aff id="aff2">
        <label>2</label>
        <institution>Department of Basic Health Sciences</institution>
        <institution>Centro Universitario del Sur</institution>
        <institution>Universidad de Guadalajara</institution>
        <addr-line>Ciudad Guzmán, Jalisco</addr-line>
        <country>Mexico</country>
      </aff>
      <aff id="aff3">
        <label>3</label>
        <institution>Engineering</institution>
        <institution>Sports Illustrated Tickets</institution>
        <addr-line>Houston, TX</addr-line>
        <country>United States</country>
      </aff>
      <aff id="aff4">
        <label>4</label>
        <institution>School of Telematics</institution>
        <institution>Universidad de Colima</institution>
        <addr-line>Colima, Colima</addr-line>
        <country>Mexico</country>
      </aff>
      <author-notes>
        <corresp>Corresponding Author: Pedro C Santana-Mancilla <email>psantana@ucol.mx</email></corresp>
      </author-notes>
      <pub-date pub-type="collection">
        <year>2026</year>
      </pub-date>
      <pub-date pub-type="epub">
        <day>20</day>
        <month>11</month>
        <year>2026</year>
      </pub-date>
      <volume>9</volume>
      <elocation-id>e83031</elocation-id>
      <history>
        <date date-type="received">
          <day>26</day>
          <month>8</month>
          <year>2025</year>
        </date>
        <date date-type="rev-request">
          <day>20</day>
          <month>11</month>
          <year>2025</year>
        </date>
        <date date-type="rev-recd">
          <day>21</day>
          <month>9</month>
          <year>2026</year>
        </date>
        <date date-type="accepted">
          <day>19</day>
          <month>10</month>
          <year>2026</year>
        </date>
      </history>
      <copyright-statement>©J Manuel Herrera-Zamora, Jesus Monroy-Rodriguez, Jesus Garcia-Mancilla, Pedro C Santana-Mancilla. Originally published in JMIR Aging (https://aging.jmir.org), 20.11.2026.</copyright-statement>
      <copyright-year>2026</copyright-year>
      <license license-type="open-access" xlink:href="https://creativecommons.org/licenses/by/4.0/">
        <p>This is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in JMIR Aging, is properly cited. The complete bibliographic information, a link to the original publication on https://aging.jmir.org, as well as this copyright and license information must be included.</p>
      </license>
      <self-uri xlink:href="https://aging.jmir.org/2026/1/e83031" xlink:type="simple"/>
      <abstract>
        <sec sec-type="background">
          <title>Background</title>
          <p>Neurodegenerative diseases such as Alzheimer disease, Parkinson disease, and related dementias are a growing health burden among older adults. Circadian disruption is increasingly recognized as both a consequence and a potential early marker of neurodegenerative progression. Wearable sensing combined with AI enables continuous, noninvasive, and ecologically valid monitoring of circadian parameters as potential digital biomarkers.</p>
        </sec>
        <sec sec-type="objective">
          <title>Objective</title>
          <p>This systematic review synthesizes evidence on AI-enhanced wearable and objective sensing devices used to monitor circadian parameters in older and cognitively vulnerable adults and evaluates their potential as digital predictive biomarkers for neurodegenerative diseases.</p>
        </sec>
        <sec sec-type="methods">
          <title>Methods</title>
          <p>Following PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) 2020 guidelines, an original search of PubMed, Scopus, and JMIR journals (June 2025) yielded 12 studies, 7 of which met updated eligibility criteria on rescreening. A complementary search of PubMed, IEEE Xplore, and the ACM Digital Library (March 24, 2026) added 3 eligible studies and 3 sensitivity references (excluded records retained to delimit the corpus boundary). Two reviewers (JMH-Z and JM-R) independently assessed risk of bias with the Mixed Methods Appraisal Tool. Heterogeneous end points precluded meta-analysis; the narrative synthesis covered 4 prespecified domains: device class, circadian construct, AI and machine learning (ML) method family, and population and neurocognitive context.</p>
        </sec>
        <sec sec-type="results">
          <title>Results</title>
          <p>The final corpus comprised 13 records (10 eligible studies and 3 sensitivity references) published from 2015 to 2025. Wrist actigraphy predominated, and samples ranged from community-dwelling older adults to nursing home residents with advanced dementia (n=16-7297). Circadian parameters included interdaily stability, intradaily variability, relative amplitude, midline estimating statistic of rhythm, acrophase, and circadian stimulus. AI/ML methods included supervised learning (random forests, gradient boosting, and regression models), unsupervised clustering, and algorithm development or validation; 7 studies applied deterministic algorithms without fitting a predictive model. Circadian fragmentation was associated with functional decline, dementia risk, and cognitive performance. Most studies were descriptive or observational with small, nondiverse samples; no AI model was externally validated, and no study followed an AI-specific reporting guideline.</p>
        </sec>
        <sec sec-type="conclusions">
          <title>Conclusions</title>
          <p>AI-enhanced wearable devices show meaningful potential for noninvasive measurement of circadian disruption in older and cognitively vulnerable populations. Evidence certainty is moderate for actigraphy-derived fragmentation-function associations in institutional settings and low for predictive biomarker claims in community populations. From 6 recurring challenges along the sensing-to-biomarker pipeline, we propose a 4-layer reference architecture (sensing and acquisition, circadian parameter extraction, AI/ML inference, and a clinical and validation interface) with a layered validation pathway. Longitudinal, multisite validation studies using standardized circadian metrics, diverse populations, and explainable AI remain necessary before clinical deployment as neurodegenerative biomarkers.</p>
        </sec>
      </abstract>
      <kwd-group>
        <kwd>circadian rhythms</kwd>
        <kwd>wearable devices</kwd>
        <kwd>AI</kwd>
        <kwd>neurodegenerative diseases</kwd>
        <kwd>older adults</kwd>
      </kwd-group>
    </article-meta>
  </front>
  <body>
    <sec sec-type="introduction">
      <title>Introduction</title>
      <sec>
        <title>Background</title>
        <p>Neurodegenerative diseases are a heterogeneous group of disorders characterized by the progressive loss of neuronal structure and function, significantly affecting memory, cognition, behavior, and sensorimotor function. These conditions, which include Alzheimer disease (AD), Parkinson disease (PD), and Lewy body dementia, represent a growing global public health challenge. The global number of people living with dementia is projected to reach 153 million by 2050 [<xref ref-type="bibr" rid="ref1">1</xref>], underscoring the urgency of developing tools for earlier detection and monitoring.</p>
        <p>Disruptions in circadian rhythms are increasingly recognized as a hallmark feature of neurodegeneration, appearing not only as a functional consequence of advancing disease but also as a potential early biological signal that may precede clinical onset [<xref ref-type="bibr" rid="ref2">2</xref>,<xref ref-type="bibr" rid="ref3">3</xref>]. The circadian system functions as the body’s intrinsic biological clock, coordinating cellular processes, sleep-wake cycles, hormonal secretion, and behavioral rhythms with the 24-hour light-dark environment [<xref ref-type="bibr" rid="ref4">4</xref>]. In older adults and those with neurodegenerative conditions, this system progressively deteriorates, resulting in measurable fragmentation of rest-activity rhythms, altered sleep architecture, and disrupted melatonin secretion [<xref ref-type="bibr" rid="ref5">5</xref>]. Recent conceptual frameworks, developed in populations other than older adults, treat sleep and circadian disruption as the joint product of circadian, neuroendocrine, and behavioral stressors rather than as a single timing signal [<xref ref-type="bibr" rid="ref6">6</xref>]; this multidimensional view is relevant to aging, where circadian decline coexists with endocrine and behavioral change.</p>
        <p>Recent advances in wearable technologies and AI offer new approaches to continuously and ecologically monitoring circadian parameters outside clinical or laboratory settings. Wrist-worn actigraphs, electroencephalography (EEG) headbands, multisensor wearable patches, wearable light sensors, and ambient smart-home monitoring systems can collect physiological and behavioral data from individuals in their natural environments. When combined with AI and machine learning (ML) algorithms, these data streams can be processed to extract circadian parameters, such as interdaily stability (IS), intradaily variability (IV), relative amplitude (RA), midline estimating statistic of rhythm (MESOR), and acrophase, enabling continuous characterization of individual circadian profiles [<xref ref-type="bibr" rid="ref7">7</xref>].</p>
        <p>Despite the promise of this technology, the evidence base remains fragmented. Existing reviews have addressed wearable sensing and AI in aging broadly, without systematically evaluating the extent to which included studies operationalize genuine circadian constructs, apply AI methods rigorously, and target populations with neurodegenerative relevance. This review addresses that gap by synthesizing evidence from studies that explicitly combine objective sensing, AI, and circadian monitoring in older or cognitively vulnerable populations, with structured attention to the specificity of the circadian construct, the AI method family used, and the clinical relevance of reported outcomes. Drawing on this synthesis, we identify 6 structural challenges at the sensing-to-biomarker interface and propose a 4-layer reference architecture with a layered validation pathway intended to guide future system design and empirical evaluation in this field.</p>
      </sec>
      <sec>
        <title>Contributions of This Review</title>
        <p>This systematic review makes the following scholarly contributions:</p>
        <list list-type="bullet">
          <list-item>
            <p>We synthesize evidence combining AI-enhanced wearable and objective sensing devices with circadian monitoring in older and cognitively vulnerable adults, organized across 4 prespecified analytical domains: device class, circadian construct, AI/ML method family, and population/neurocognitive context.</p>
          </list-item>
          <list-item>
            <p>We identify and characterize 6 structural challenges that currently limit the translation of circadian monitoring into validated predictive biomarkers for neurodegenerative disease, situating each challenge within the pipeline stage it most directly constrains.</p>
          </list-item>
          <list-item>
            <p>Drawing on the recurrence of these 4 analytical domains across the reviewed corpus, we propose a 4-layer reference pipeline architecture together with a layered validation pathway that maps each challenge to architectural responses and implementation propositions. This proposal represents a conceptual framework derived from the reviewed evidence; empirical validation through prospective field studies or multisite testbeds is outside the scope of this review but constitutes an explicit priority for future work.</p>
          </list-item>
          <list-item>
            <p>We articulate 4 implementation propositions grounded in the evidence base, including harmonized circadian end-point reporting, explainable AI (XAI) integration at model design time, population-specific device validation, and demographic representativeness as an operational validation criterion.</p>
          </list-item>
        </list>
      </sec>
    </sec>
    <sec sec-type="methods">
      <title>Methods</title>
      <sec>
        <title>Overview</title>
        <p>This systematic review was conducted in accordance with the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) 2020 guidelines [<xref ref-type="bibr" rid="ref8">8</xref>] to ensure methodological transparency and rigor. The completed PRISMA 2020 checklist is available in <xref ref-type="supplementary-material" rid="app1">Multimedia Appendix 1</xref>. The review protocol was prospectively deposited in Zenodo prior to data collection, in accordance with Findable, Accessible, Interoperable, and Reusable data principles and open science standards recommended by PRISMA 2020 [<xref ref-type="bibr" rid="ref1">1</xref>]. The Zenodo record is available at [<xref ref-type="bibr" rid="ref9">9</xref>].</p>
        <p>During peer review, reviewers noted that the original search (PubMed, Scopus, and JMIR journals) underrepresented the engineering, biomedical informatics, and human-computer interaction literature, where AI-based sensing work is often published. To address this, we conducted a complementary systematic search across IEEE Xplore, the ACM Digital Library, and an updated PubMed query, applying the same prespecified eligibility criteria without modification. Because this search postdated the protocol deposited in Zenodo, we treat it as a documented amendment to the protocol rather than a change to the review question or eligibility rules; the deposited record reflects only the initial search.</p>
        <p>The aim of this systematic review was to evaluate the current use of wearable devices that monitor circadian parameters using AI and to explore their potential as digital biomarkers for predicting the onset of neurodegenerative diseases in older adults, thereby contributing to improved health outcomes in aging populations. As a second, synthesis-driven aim, we sought to translate the analytical domains and structural challenges recurring across the reviewed evidence into a 4-layer reference pipeline architecture and an accompanying layered validation pathway, intended to guide the design and empirical evaluation of clinically deployable AI-circadian biomarker systems. This second aim is interpretive: the architecture and validation pathway were derived by the authors from the synthesized evidence and are not findings of the systematic review itself.</p>
      </sec>
      <sec>
        <title>Data Sources and Search Strategy</title>
        <sec>
          <title>Original Search</title>
          <p>A primary literature search was conducted in June 2025 across PubMed, Scopus, and JMIR journals, targeting studies published between January 2015 and June 2025. Search terms were organized around three conceptual blocks: (1) wearable and objective sensing technologies, (2) circadian parameters and 24-hour behavioral constructs, and (3) aging or neurocognitive populations. This search identified 8441 records and, after duplicate removal and screening, yielded 12 included studies.</p>
        </sec>
        <sec>
          <title>Complementary Search</title>
          <p>Following peer-review recommendations, a complementary systematic search was conducted on March 24, 2026, across 3 additional bibliographic databases: PubMed (to capture publications beyond the original cutoff), IEEE Xplore, and the ACM Digital Library. These databases were selected to capture engineering, biomedical informatics, and human-computer interaction literature not adequately represented in the original search. Two parallel queries were run per database: a main search and a neurocognitive sensitivity search. Full database-specific Boolean search strings, including all field tags, filters (peer-reviewed journals only, 2015 onward), and date limits, are reported in <xref ref-type="supplementary-material" rid="app2">Multimedia Appendix 2</xref> in accordance with PRISMA-S (Preferred Reporting Items for Systematic Reviews and Meta-Analyses Literature Search Extension) guidelines [<xref ref-type="bibr" rid="ref10">10</xref>]. This search identified 162 records; 28 duplicates were removed, yielding 134 records for screening. Full-text assessment was conducted on 27 records. The publication window (2015-2025) was the same as in the original search; the later search date extended database coverage, not the period of eligible publications.</p>
        </sec>
        <sec>
          <title>Eligibility Criteria</title>
          <p>Studies were included if they met all the following criteria:</p>
          <list list-type="bullet">
            <list-item>
              <p>Original peer-reviewed journal article.</p>
            </list-item>
            <list-item>
              <p>Study population comprising adults aged ≥45 years, or individuals of any age with a documented neurodegenerative condition relevant to this review (mild cognitive impairment [MCI], AD, PD, Lewy body dementia, or related dementia syndromes).</p>
            </list-item>
            <list-item>
              <p>Use of at least 1 wearable or objective sensing device for physiological or behavioral data collection.</p>
            </list-item>
            <list-item>
              <p>Application of at least 1 AI or ML method in data analysis, pattern detection, feature extraction, or predictive modeling.</p>
            </list-item>
            <list-item>
              <p>Report of at least 1 explicit circadian parameter (IS, IV, RA, MESOR, acrophase, cosinor-derived parameter, circadian stimulus, or circadian rhythm strength) or a 24-hour behavioral construct interpreted within a circadian or chronobiological framework.</p>
            </list-item>
          </list>
          <p>Studies were excluded if they were reviews, meta-analyses, editorials, letters, conference abstracts, or protocols; used exclusively nonhuman participants; lacked full-text access; or did not incorporate an AI or ML component in the analysis.</p>
          <p>Sensitivity references were defined as records excluded at full-text assessment that failed exactly 1 eligibility criterion, met all the others, and addressed a methodological element directly relevant to the review question: validation of a wearable sleep-wake algorithm against a reference standard, AI-based prediction from continuous sensor data, or extraction of circadian parameters in an institutional population. Sensitivity references are not counted as eligible studies. They are identified by a footnote in all tables, reported separately in the risk-of-bias assessment, and used to delimit the methodological boundary of the corpus; no conclusion of this review rests on them alone.</p>
        </sec>
        <sec>
          <title>Study Selection Process</title>
          <p>Records were imported into Zotero for deduplication and screening. Title and abstract screening were conducted using Rayyan Qatar Computing Research Institute (original search) and a standardized manual protocol (complementary search). Full-text eligibility assessment was performed by 2 independent reviewers (JMH-Z and JM-R for the original search; JMH-Z and PCS-M for the complementary search). Disagreements were resolved through consensus or by a third reviewer (JG-M).</p>
          <p>The 12 studies from the original search were rescreened against the updated eligibility criteria during revision. In total, 5 were excluded: 3 because no AI or ML method was used in the analysis [<xref ref-type="bibr" rid="ref11">11</xref>-<xref ref-type="bibr" rid="ref13">13</xref>], and 2 because the enrolled population did not meet the population criterion or no explicit circadian parameter was analyzed [<xref ref-type="bibr" rid="ref14">14</xref>,<xref ref-type="bibr" rid="ref15">15</xref>]. A total of 7 studies were retained from the original corpus.</p>
        </sec>
        <sec>
          <title>Data Extraction and Synthesis</title>
          <p>Data were extracted from each included study using a standardized form covering authors, year and country, study design, population characteristics (sample size, age, condition), sensing device and modality, circadian construct and parameter names, AI/ML method and family, analytic approach, and main findings. Two reviewers (JMH-Z and JM-R) extracted data independently; discrepancies were resolved by discussion. For each study, we also extracted how computational methods were reported: validation scheme, unit of data partitioning, hyperparameter reporting, availability of code and data, explainability or variable-importance reporting, and adherence to AI-specific reporting guidelines (eg, TRIPOD+AI [Transparent Reporting of a Multivariable Prediction Model for Individual Prognosis or Diagnosis+Artificial Intelligence] [<xref ref-type="bibr" rid="ref16">16</xref>], CLAIM [Checklist for Artificial Intelligence in Medical Imaging] [<xref ref-type="bibr" rid="ref17">17</xref>], DECIDE-AI [Developmental and Exploratory Clinical Investigations of Decision Support Systems Driven by Artificial Intelligence] [<xref ref-type="bibr" rid="ref18">18</xref>], and CONSORT-AI [Consolidated Standards of Reporting Trials-Artificial Intelligence] [<xref ref-type="bibr" rid="ref19">19</xref>]). A codebook with controlled vocabularies separated absent reporting (not reported) from inapplicability (not applicable and when no predictive model was fitted). A large language model–based agentic workflow (powered by Anthropic’s Claude) produced a first-pass extraction from the full texts, recording a verbatim evidence locator for every value other than not reported; 2 reviewers (JMH-Z and PCS-M) then independently verified each value against the full text and supplementary materials of each source article, resolving disagreements by discussion or by a third reviewer (JG-M). Because AI assistance in scholarly evaluation can speed up the work but raises concerns about the depth and authenticity of the resulting judgments [<xref ref-type="bibr" rid="ref20">20</xref>], the assistant’s output was treated as a draft only, and no extracted value was retained without human verification. Consistent with the eligibility criterion, which counts feature extraction as a qualifying use of AI or ML, studies applying deterministic algorithms to sensor data were eligible even when they fitted no predictive model; this distinction is reported for every study in the Results section.</p>
          <p>Narrative synthesis was organized around four prespecified analytical domains: (1) device class and sensing modality, (2) circadian construct and parametric specificity, (3) AI/ML method family, and (4) population and neurocognitive context. Studies were first cataloged descriptively within each domain, then analyzed comparatively to identify convergent evidence, methodological gaps, and key contributions. Three sensitivity-reference studies [<xref ref-type="bibr" rid="ref21">21</xref>-<xref ref-type="bibr" rid="ref23">23</xref>] were used to interpret the methodological boundary of the corpus and are indicated by a footnote in all tables.</p>
        </sec>
        <sec>
          <title>Risk of Bias Assessment</title>
          <p>Risk of bias was assessed using the Mixed Methods Appraisal Tool (MMAT) [<xref ref-type="bibr" rid="ref24">24</xref>], which provides design-specific quality criteria for quantitative descriptive (type 4), quantitative nonrandomized (type 3), and quantitative randomized controlled trial (RCT; type 2) designs. Two reviewers (JMH-Z and JM-R) independently assessed each included study, applying the MMAT screening questions (S1 and S2) and 5 design-specific quality criteria (Q1-Q5). Disagreements were resolved by consensus. Domain-level judgments (Yes/No/Cannot tell) and overall appraisal (low, moderate, or high risk) are reported in the next section.</p>
        </sec>
      </sec>
    </sec>
    <sec sec-type="results">
      <title>Results</title>
      <sec>
        <title>Study Selection</title>
        <p>The 2-stage search process yielded a final corpus of 13 records: 10 eligible studies and 3 sensitivity references (<xref rid="figure1" ref-type="fig">Figure 1</xref>). From the original search corpus, 7 of 12 previously included studies were retained after rescreening; 5 were excluded. From the complementary search, 162 records were identified across 3 databases (PubMed: n=69; IEEE Xplore: n=31; ACM Digital Library: n=62). After removing 28 duplicates, 134 records were screened at the title and abstract; 107 were excluded. In total, 27 records were assessed at full text; 24 were excluded (wrong age group: n=4; no explicit circadian parameter measured: n=5; no AI/ML component: n=1; no circadian or 24-hour framework: n=14). A total of 3 records met all eligibility criteria and were included; 3 of the 24 excluded records met the definition of sensitivity references and were retained as such. <xref rid="figure1" ref-type="fig">Figure 1</xref> presents the PRISMA flow diagram with the full search strategy.</p>
        <fig id="figure1" position="float">
          <label>Figure 1</label>
          <caption>
            <p>PRISMA flowchart for study selection (publications 2015-2025; searches run in June 2025 and March 2026). Of the 8603 records identified, 13 were included in the final corpus: 10 met all eligibility criteria, and 3 were retained as sensitivity references. ML: machine learning; PRISMA: Preferred Reporting Items for Systematic Reviews and Meta-Analyses.</p>
          </caption>
          <graphic xlink:href="aging_v9i1e83031_fig1.png" alt-version="no" mimetype="image" position="float" xlink:type="simple"/>
        </fig>
      </sec>
      <sec>
        <title>Study Characteristics</title>
        <p>The 13 records (10 eligible studies and 3 sensitivity references) were published between 2015 and 2025. They were conducted in the United States (n=7), the United Kingdom (n=2), Taiwan (n=2), the Netherlands (n=1), and Finland (n=1). Sample sizes ranged from 16 to 7297 participants. In total, 6 studies enrolled populations with documented or high-risk neurodegenerative conditions; 4 enrolled community-dwelling or population-based samples without documented cognitive impairment; and 3 used middle-aged or mixed-age samples as methodological anchors. Full characteristics are presented in <xref ref-type="table" rid="table1">Table 1</xref>.</p>
        <table-wrap position="float" id="table1">
          <label>Table 1</label>
          <caption>
            <p>Characteristics of included studies on AI-enhanced wearable devices for circadian monitoring in aging populations.</p>
          </caption>
          <table width="1000" cellpadding="5" cellspacing="0" border="1" rules="groups" frame="hsides">
            <col width="90"/>
            <col width="120"/>
            <col width="290"/>
            <col width="150"/>
            <col width="180"/>
            <col width="170"/>
            <thead>
              <tr valign="top">
                <td>Author (year)</td>
                <td>Population and setting</td>
                <td>Device</td>
                <td>Circadian parameters</td>
                <td>AI/ML<sup>a</sup> method</td>
                <td>Key findings</td>
              </tr>
            </thead>
            <tbody>
              <tr valign="top">
                <td>Chuang et al (2024) [<xref ref-type="bibr" rid="ref25">25</xref>]</td>
                <td>Taiwan; n=78 adults (mean 41.5 years; 59% women; 29 obese, 16 overweight, 33 controls)</td>
                <td>Wrist actigraphy+Rhythm smartphone app</td>
                <td>IS<sup>b</sup>, social jet lag, sleep indicators</td>
                <td>Actigraphy-based algorithm; multivariate regression</td>
                <td>IS strongly correlated with adiposity. App-based measurement provided stronger associations than traditional actigraphy; IS from app correlated with obesity index, unlike social jet lag.</td>
              </tr>
              <tr valign="top">
                <td>Razjouyan et al (2017) [<xref ref-type="bibr" rid="ref26">26</xref>]</td>
                <td>United States; n=21 adults (mean 50.8, SD 12.8 years; 47.8% women; poor sleep)</td>
                <td>Chest sensor (BioPatch), wrist Actiwatch, PSG<sup>c</sup> (reference)</td>
                <td>Sleep-wake epoch; implicit rest-activity rhythm</td>
                <td>Posture+acceleration algorithm; logistic regression</td>
                <td>Chest sensor achieved 85.8% accuracy versus 79.8% for Actiwatch. Combined posture and acceleration improved sleep/wake epoch detection.</td>
              </tr>
              <tr valign="top">
                <td>Gabb et al (2025) [<xref ref-type="bibr" rid="ref27">27</xref>]</td>
                <td>United Kingdom; n=41 older adults (mean 70.9 years); 10 AD<sup>d</sup>, 11 LBD<sup>e</sup>, 20 controls</td>
                <td>Axivity AX3, Dreem 2 EEG<sup>f</sup> headband, MyDignio app</td>
                <td>Sleep continuity, circadian rhythm,<break/>cognitive task performance</td>
                <td>GGIR<sup>g</sup> automated pipeline; Dreem EEG algorithm</td>
                <td>Remote longitudinal monitoring feasible; 98% retention, 79%-97% data completeness. Multimodal approach viable in populations with MCI<sup>h</sup> and dementia.</td>
              </tr>
              <tr valign="top">
                <td>Cavaillès et al (2025) [<xref ref-type="bibr" rid="ref28">28</xref>]</td>
                <td>United States (MrOS<sup>i</sup> cohort); n=2667 older men</td>
                <td>SleepWatch-O wrist actigraph</td>
                <td>24-hour activity profiles; sleep/circadian phenotypes</td>
                <td>Unsupervised ML clustering (MCGHD<sup>j</sup>+PCA<sup>k</sup>)</td>
                <td>Three phenotypes: active healthy (64%), fragmented poor (14.1%), and long nappers (21.9%). Fragmented poor sleepers showed higher dementia and CVD<sup>l</sup> event risk.</td>
              </tr>
              <tr valign="top">
                <td>Wang et al (2024) [<xref ref-type="bibr" rid="ref29">29</xref>]</td>
                <td>Taiwan; n=191 community adults ≥50 years</td>
                <td>Actiwatch (3-axis accelerometer+light+temperature sensors), sleep diaries</td>
                <td>IS, IV<sup>m</sup>, RA<sup>n</sup>, chronotype</td>
                <td>Kuo actigraphy algorithm; ANOVA+regression</td>
                <td>Older adults had lower activity, worse sleep efficiency, earlier chronotype. Women showed greater IS and RA. Age-gender interactions for life satisfaction and RA.</td>
              </tr>
              <tr valign="top">
                <td>Shim et al (2023) [<xref ref-type="bibr" rid="ref30">30</xref>]</td>
                <td>United States (NHANES<sup>o</sup>); n=7297 adults</td>
                <td>ActiGraph GT3X+wrist accelerometer</td>
                <td>24-hour activity profiles as digital biomarkers (inflammation, biological age, mortality)</td>
                <td>Hierarchical clustering (Ward method); recursive feature elimination</td>
                <td>Five distinct 24-hour activity profiles identified. Circadian misalignment linked to systemic inflammation, accelerated biological age, and increased mortality risk.</td>
              </tr>
              <tr valign="top">
                <td>Jones et al (2025) [<xref ref-type="bibr" rid="ref31">31</xref>]</td>
                <td>United States; n=40 older adults (mean 67.2 years; 72.5% women); 19 MCI, 21 controls</td>
                <td>Apple Watch+myRhythmWatch platform</td>
                <td>IS, IV, RA, cosinor (MESOR<sup>p</sup>, acrophase, rhythm strength)</td>
                <td>myRW<sup>q</sup> algorithm; cosinor analysis; linear regression</td>
                <td>Later activity onset associated with worse memory. Fragmented rhythms linked to slower processing speed. Platform feasible for circadian assessment in MCI.</td>
              </tr>
              <tr valign="top">
                <td>Beiglary et al (2025) [<xref ref-type="bibr" rid="ref32">32</xref>]</td>
                <td>United States; n=29 older adults with dementia; 2 assisted-living facilities</td>
                <td>Wearable light sensor calibrated versus spectrophotometer</td>
                <td>Circadian stimulus, personal circadian lighting exposure</td>
                <td>Random forest (ML); laboratory calibration+field deployment</td>
                <td>Random forest accurately predicted circadian stimulus. Substantial interindividual variability in circadian light exposure. In total, 40% device loss; validated against spectrophotometric ground truth.</td>
              </tr>
              <tr valign="top">
                <td>Merilahti et al (2016) [<xref ref-type="bibr" rid="ref23">23</xref>]<sup>r</sup></td>
                <td>Finland; n=16 nursing home residents (mean 90.7 years; 7 with dementia)</td>
                <td>Wrist-worn device with integrated activity sensor</td>
                <td>MESOR, amplitude, acrophase, IS, IV, RA, CRS<sup>s</sup></td>
                <td>Single cosinor function (24 hours); nonparametric RAS<sup>t</sup> analysis</td>
                <td>Physical functioning associated with rhythm strength and stability. Functional decline tracked with reduced activity, lower MESOR/amplitude, and increased IV.</td>
              </tr>
              <tr valign="top">
                <td>Froggatt et al (2020) [<xref ref-type="bibr" rid="ref33">33</xref>]</td>
                <td>United Kingdom; n=32 nursing home residents with advanced dementia (mean 81.5 years)</td>
                <td>ActiGraph wrist actigraphy</td>
                <td>IS, IV (circadian fragmentation and synchronization)</td>
                <td>Nonparametric actigraphy analysis (IS/IV computation)</td>
                <td>Feasibility cluster RCT<sup>u</sup>. Highly inactive population with fragmented rhythms (IS: mean 0.08; IV: mean 0.7). Actigraphy feasible in advanced dementia.</td>
              </tr>
              <tr valign="top">
                <td>Sprint et al (2022) [<xref ref-type="bibr" rid="ref34">34</xref>]</td>
                <td>United States; n=14 amnestic MCI (mean 74.4 years) in smart-home environment</td>
                <td>Smart-home sensors+EMMA tablet app+text markers</td>
                <td>Circadian rhythm strength, 24-hour circadian rhythm, regularity indices</td>
                <td>Multimodal fusion; joint prediction (regression/ML)</td>
                <td>Moderate-to-large prediction correlations (<italic>r</italic>=0.601-0.871) for 10 clinical assessments. Joint prediction outperformed independent modality models.</td>
              </tr>
              <tr valign="top">
                <td>de Zambotti et al (2015) [<xref ref-type="bibr" rid="ref21">21</xref>]<sup>r</sup></td>
                <td>United States; n=28 midlife women (mean 50.1 years); sleep laboratory</td>
                <td>Jawbone UP consumer tracker vs PSG (reference)</td>
                <td>Sleep-wake rhythm (nightly sleep metrics)</td>
                <td>Proprietary consumer algorithm; PSG comparison</td>
                <td>High sleep sensitivity, poor wake specificity; overestimated TST<sup>v</sup>, underestimated WASO<sup>w</sup>. Potentially useful for daily sleep-wake rhythm tracking over multiple days.</td>
              </tr>
              <tr valign="top">
                <td>Robben et al (2017) [<xref ref-type="bibr" rid="ref22">22</xref>]<sup>r</sup></td>
                <td>Netherlands; n=13 older adults (mean 80, SD 7 years); independent living</td>
                <td>Ambient smart-home sensors (PIR<sup>x</sup>, pressure, switches); no on-body device</td>
                <td>24-hour behavioral structure as temporal proxy</td>
                <td>Delta-feature temporal modeling; random forest prediction</td>
                <td>Delta features predicted functional health change, outperforming static baselines. Temporal behavioral regularity exploited as implicit circadian proxy.</td>
              </tr>
            </tbody>
          </table>
          <table-wrap-foot>
            <fn id="table1fn1">
              <p><sup>a</sup>ML: machine learning.</p>
            </fn>
            <fn id="table1fn2">
              <p><sup>b</sup>IS: interdaily stability.</p>
            </fn>
            <fn id="table1fn3">
              <p><sup>c</sup>PSG: polysomnography.</p>
            </fn>
            <fn id="table1fn4">
              <p><sup>d</sup>AD: Alzheimer disease.</p>
            </fn>
            <fn id="table1fn5">
              <p><sup>e</sup>LBD: Lewy body dementia.</p>
            </fn>
            <fn id="table1fn6">
              <p><sup>f</sup>EEG: electroencephalography.</p>
            </fn>
            <fn id="table1fn7">
              <p><sup>g</sup>GGIR: R-package for accelerometry.</p>
            </fn>
            <fn id="table1fn8">
              <p><sup>h</sup>MCI: mild cognitive impairment.</p>
            </fn>
            <fn id="table1fn9">
              <p><sup>i</sup>MrOS: Osteoporotic Fractures in Men Study.</p>
            </fn>
            <fn id="table1fn10">
              <p><sup>j</sup>MCGHD: multiple coalesced generalized hyperbolic distribution.</p>
            </fn>
            <fn id="table1fn11">
              <p><sup>k</sup>PCA: principal component analysis</p>
            </fn>
            <fn id="table1fn12">
              <p><sup>l</sup>CVD: cardiovascular disease.</p>
            </fn>
            <fn id="table1fn13">
              <p><sup>m</sup>IV: intradaily variability.</p>
            </fn>
            <fn id="table1fn14">
              <p><sup>n</sup>RA: relative amplitude.</p>
            </fn>
            <fn id="table1fn15">
              <p><sup>o</sup>NHANES: National Health and Nutrition Examination Survey.</p>
            </fn>
            <fn id="table1fn16">
              <p><sup>p</sup>MESOR: midline estimating statistic of rhythm.</p>
            </fn>
            <fn id="table1fn17">
              <p><sup>q</sup>myRW: myRhythmWatch.</p>
            </fn>
            <fn id="table1fn18">
              <p><sup>r</sup>Sensitivity-reference study.</p>
            </fn>
            <fn id="table1fn19">
              <p><sup>s</sup>CRS: circadian rhythm strength.</p>
            </fn>
            <fn id="table1fn20">
              <p><sup>t</sup>RAS: rest-activity rhythm analysis.</p>
            </fn>
            <fn id="table1fn21">
              <p><sup>u</sup>RCT: randomized controlled trial.</p>
            </fn>
            <fn id="table1fn22">
              <p><sup>v</sup>TST: total sleep time.</p>
            </fn>
            <fn id="table1fn23">
              <p><sup>w</sup>WASO: wake after sleep onset.</p>
            </fn>
            <fn id="table1fn24">
              <p><sup>x</sup>PIR: passive infrared.</p>
            </fn>
          </table-wrap-foot>
        </table-wrap>
      </sec>
      <sec>
        <title>Risk of Bias and Quality Assessment</title>
        <p>Risk of bias was assessed with the MMAT (version 2018) [<xref ref-type="bibr" rid="ref24">24</xref>], as described in the Methods section. The complete dataset to enhance transparency is available as <xref ref-type="supplementary-material" rid="app3">Multimedia Appendix 3</xref>.</p>
        <p>MMAT results are presented in <xref ref-type="table" rid="table2">Table 2</xref>. In total, 8 studies were classified as quantitative descriptive (type 4), 4 as quantitative nonrandomized (type 3), and 1 as a quantitative RCT (type 2). A total of 6 studies received an overall low-risk appraisal. Wang et al [<xref ref-type="bibr" rid="ref29">29</xref>] was rated moderate risk due to a nonrepresentative sample (Q2: No). Beiglary et al [<xref ref-type="bibr" rid="ref32">32</xref>] was rated moderate risk because sample representativeness could not be determined (Q2: Cannot tell), and Merilahti et al [<xref ref-type="bibr" rid="ref23">23</xref>] because low reporting bias could not be confirmed (Q4: Cannot tell). Sprint et al [<xref ref-type="bibr" rid="ref34">34</xref>] was rated as moderate-to-high risk, given a small sample size (n=14), no control of confounders, and uncertain completeness of outcome data. Froggatt et al [<xref ref-type="bibr" rid="ref33">33</xref>] received a high-risk appraisal, as expected for a complex-care feasibility RCT with nonblinded outcomes; this does not diminish the descriptive validity of the actigraphy findings. All 3 sensitivity-reference studies were rated moderate risk.</p>
        <table-wrap position="float" id="table2">
          <label>Table 2</label>
          <caption>
            <p>Mixed Methods Appraisal Tool (MMAT) results for included studies<sup>a</sup>.</p>
          </caption>
          <table width="1000" cellpadding="5" cellspacing="0" border="1" rules="groups" frame="hsides">
            <col width="310"/>
            <col width="160"/>
            <col width="50"/>
            <col width="50"/>
            <col width="50"/>
            <col width="50"/>
            <col width="50"/>
            <col width="50"/>
            <col width="50"/>
            <col width="180"/>
            <thead>
              <tr valign="top">
                <td>Author (year)</td>
                <td>MMAT type</td>
                <td>S1</td>
                <td>S2</td>
                <td>Q1</td>
                <td>Q2</td>
                <td>Q3</td>
                <td>Q4</td>
                <td>Q5</td>
                <td>Appraisal</td>
              </tr>
            </thead>
            <tbody>
              <tr valign="top">
                <td>Chuang et al (2024) [<xref ref-type="bibr" rid="ref25">25</xref>]</td>
                <td>Type 4</td>
                <td>Y<sup>b</sup></td>
                <td>Y</td>
                <td>Y</td>
                <td>Y</td>
                <td>Y</td>
                <td>Y</td>
                <td>Y</td>
                <td>Low risk</td>
              </tr>
              <tr valign="top">
                <td>Razjouyan et al (2017) [<xref ref-type="bibr" rid="ref26">26</xref>]</td>
                <td>Type 4</td>
                <td>Y</td>
                <td>Y</td>
                <td>Y</td>
                <td>Y</td>
                <td>Y</td>
                <td>Y</td>
                <td>Y</td>
                <td>Low risk</td>
              </tr>
              <tr valign="top">
                <td>Gabb et al (2025) [<xref ref-type="bibr" rid="ref27">27</xref>]</td>
                <td>Type 4</td>
                <td>Y</td>
                <td>Y</td>
                <td>Y</td>
                <td>Y</td>
                <td>Y</td>
                <td>Y</td>
                <td>Y</td>
                <td>Low risk</td>
              </tr>
              <tr valign="top">
                <td>Cavaillès et al (2025) [<xref ref-type="bibr" rid="ref28">28</xref>]</td>
                <td>Type 3</td>
                <td>Y</td>
                <td>Y</td>
                <td>Y</td>
                <td>Y</td>
                <td>Y</td>
                <td>Y</td>
                <td>Y</td>
                <td>Low risk</td>
              </tr>
              <tr valign="top">
                <td>Wang et al (2024) [<xref ref-type="bibr" rid="ref29">29</xref>]</td>
                <td>Type 4</td>
                <td>Y</td>
                <td>Y</td>
                <td>Y</td>
                <td>N<sup>c</sup></td>
                <td>Y</td>
                <td>Y</td>
                <td>Y</td>
                <td>Moderate</td>
              </tr>
              <tr valign="top">
                <td>Shim et al (2023) [<xref ref-type="bibr" rid="ref30">30</xref>]</td>
                <td>Type 3</td>
                <td>Y</td>
                <td>Y</td>
                <td>Y</td>
                <td>Y</td>
                <td>Y</td>
                <td>Y</td>
                <td>Y</td>
                <td>Low risk</td>
              </tr>
              <tr valign="top">
                <td>Jones et al (2025) [<xref ref-type="bibr" rid="ref31">31</xref>]</td>
                <td>Type 4</td>
                <td>Y</td>
                <td>Y</td>
                <td>Y</td>
                <td>Y</td>
                <td>Y</td>
                <td>Y</td>
                <td>Y</td>
                <td>Low risk</td>
              </tr>
              <tr valign="top">
                <td>Beiglary et al (2025) [<xref ref-type="bibr" rid="ref32">32</xref>]</td>
                <td>Type 4</td>
                <td>Y</td>
                <td>Y</td>
                <td>Y</td>
                <td>C<sup>d</sup></td>
                <td>Y</td>
                <td>Y</td>
                <td>Y</td>
                <td>Moderate</td>
              </tr>
              <tr valign="top">
                <td>Merilahti et al (2016) [<xref ref-type="bibr" rid="ref23">23</xref>]<sup>e</sup></td>
                <td>Type 4</td>
                <td>Y</td>
                <td>Y</td>
                <td>Y</td>
                <td>Y</td>
                <td>Y</td>
                <td>C</td>
                <td>Y</td>
                <td>Moderate</td>
              </tr>
              <tr valign="top">
                <td>Froggatt et al (2020) [<xref ref-type="bibr" rid="ref33">33</xref>]</td>
                <td>Type 2</td>
                <td>Y</td>
                <td>Y</td>
                <td>Y</td>
                <td>N</td>
                <td>C</td>
                <td>N</td>
                <td>C</td>
                <td>High risk<sup>f</sup></td>
              </tr>
              <tr valign="top">
                <td>Sprint et al (2022) [<xref ref-type="bibr" rid="ref34">34</xref>]</td>
                <td>Type 3</td>
                <td>Y</td>
                <td>Y</td>
                <td>C</td>
                <td>Y</td>
                <td>C</td>
                <td>N</td>
                <td>Y</td>
                <td>Moderate-high</td>
              </tr>
              <tr valign="top">
                <td>de Zambotti et al (2015) [<xref ref-type="bibr" rid="ref21">21</xref>]<sup>e</sup></td>
                <td>Type 4</td>
                <td>Y</td>
                <td>Y</td>
                <td>Y</td>
                <td>C</td>
                <td>Y</td>
                <td>Y</td>
                <td>Y</td>
                <td>Moderate</td>
              </tr>
              <tr valign="top">
                <td>Robben et al (2017) [<xref ref-type="bibr" rid="ref22">22</xref>]<sup>e</sup></td>
                <td>Type 3</td>
                <td>Y</td>
                <td>Y</td>
                <td>Y</td>
                <td>Y</td>
                <td>Y</td>
                <td>C</td>
                <td>Y</td>
                <td>Moderate</td>
              </tr>
            </tbody>
          </table>
          <table-wrap-foot>
            <fn id="table2fn1">
              <p><sup>a</sup>Type 2 (randomized controlled trial): Q1=randomization appropriately performed; Q2=groups comparable at baseline; Q3=complete outcome data; Q4=outcome assessors blinded; Q5=participants adhered to the assigned intervention. Type 3 (nonrandomized): Q1=participants representative of the target population; Q2=appropriate measurement of outcome and exposure; Q3=complete outcome data; Q4=confounders accounted for in design and analysis; Q5=exposure occurred as intended. Type 4 (descriptive): Q1=sampling strategy relevant to the research question; Q2=sample representative of the target population; Q3=appropriate measurements; Q4=low risk of nonresponse bias; Q5=appropriate statistical analysis.</p>
            </fn>
            <fn id="table2fn2">
              <p><sup>b</sup>Y: yes.</p>
            </fn>
            <fn id="table2fn3">
              <p><sup>c</sup>N: no.</p>
            </fn>
            <fn id="table2fn4">
              <p><sup>d</sup>C: cannot tell.</p>
            </fn>
            <fn id="table2fn5">
              <p><sup>e</sup>Sensitivity-reference study.</p>
            </fn>
            <fn id="table2fn6">
              <p><sup>f</sup>High risk is expected and acceptable for a feasibility cluster randomized controlled trial.</p>
            </fn>
          </table-wrap-foot>
        </table-wrap>
      </sec>
      <sec>
        <title>Synthesis by Domain</title>
        <sec>
          <title>Device Class and Sensing Modality</title>
          <p>Wrist actigraphy was the most used modality, enabling derivation of nonparametric circadian metrics and cosinor parameters from continuous rest-activity data recorded over days to months. Devices deployed included ActiGraph GT3X+ [<xref ref-type="bibr" rid="ref30">30</xref>], SleepWatch-O [<xref ref-type="bibr" rid="ref28">28</xref>], Axivity AX3 [<xref ref-type="bibr" rid="ref27">27</xref>], Actiwatch [<xref ref-type="bibr" rid="ref29">29</xref>], Apple Watch with the myRhythmWatch platform [<xref ref-type="bibr" rid="ref31">31</xref>], an unspecified integrated activity-sensing wrist device [<xref ref-type="bibr" rid="ref23">23</xref>], and a spectrophotometer-calibrated wearable light sensor [<xref ref-type="bibr" rid="ref32">32</xref>]. Device validation varied widely: Beiglary et al [<xref ref-type="bibr" rid="ref32">32</xref>] performed explicit laboratory calibration against a spectrophotometric reference before field deployment; de Zambotti et al [<xref ref-type="bibr" rid="ref21">21</xref>] compared consumer tracker output with polysomnography; and Razjouyan et al [<xref ref-type="bibr" rid="ref26">26</xref>] validated a novel algorithm against polysomnography in a sleep laboratory setting. Most other studies used devices without concurrent within-study validation. Multisensor platforms that combine actigraphy with EEG headbands, apps, and pulse oximetry were used by Gabb et al [<xref ref-type="bibr" rid="ref27">27</xref>]. Smart-home ambient sensor arrays without on-body devices were used in Sprint et al [<xref ref-type="bibr" rid="ref34">34</xref>] and Robben et al [<xref ref-type="bibr" rid="ref22">22</xref>]. A smartphone-actigraphy combination formed the sensing platform in Chuang et al [<xref ref-type="bibr" rid="ref25">25</xref>].</p>
        </sec>
        <sec>
          <title>Circadian Construct and Parametric Specificity</title>
          <p>Studies differed substantially in how explicitly they operationalized circadian biology. Six studies derived or analyzed named chronobiological metrics as primary or clearly defined outcomes: IS, IV, RA, MESOR, acrophase, and circadian stimulus [<xref ref-type="bibr" rid="ref23">23</xref>,<xref ref-type="bibr" rid="ref25">25</xref>,<xref ref-type="bibr" rid="ref29">29</xref>,<xref ref-type="bibr" rid="ref31">31</xref>-<xref ref-type="bibr" rid="ref33">33</xref>]. Four studies embedded temporal or 24-hour rhythm features within predictive or clustering frameworks without foregrounding them as primary end points [<xref ref-type="bibr" rid="ref27">27</xref>,<xref ref-type="bibr" rid="ref28">28</xref>,<xref ref-type="bibr" rid="ref30">30</xref>,<xref ref-type="bibr" rid="ref34">34</xref>]. Two studies used continuous sleep-wake rhythm monitoring as a functional proxy for circadian structure without computing formal circadian parameters [<xref ref-type="bibr" rid="ref21">21</xref>,<xref ref-type="bibr" rid="ref26">26</xref>]. One study treated ambient behavioral regularity as an implicit temporal proxy [<xref ref-type="bibr" rid="ref22">22</xref>].</p>
          <p>This gradient from explicit to implicit circadian operationalization has important implications for interpreting studies as evidence for circadian biomarkers. Only studies reporting formal chronobiological metrics provide direct evidence for the circadian construct. Studies that use sleep-wake rhythm or behavioral regularity as proxies provide contextual support but cannot serve as substitutes for direct circadian measurement.</p>
        </sec>
        <sec>
          <title>AI/ML Method Family</title>
          <p>Four AI/ML families were identified (<xref ref-type="table" rid="table3">Table 3</xref>). Supervised predictive modeling was applied in 3 studies: a random forest predicting circadian light exposure from a calibrated wearable light sensor [<xref ref-type="bibr" rid="ref32">32</xref>]; decision tree, random forest, and gradient-boosted regressors predicting clinical assessment scores from multimodal behavioral markers in MCI [<xref ref-type="bibr" rid="ref34">34</xref>]; and regression forests predicting change in functional health from ambient sensor data [<xref ref-type="bibr" rid="ref22">22</xref>]. Unsupervised phenotyping identified distinct 24-hour activity profiles in 2 large-sample studies, using hierarchical clustering after recursive feature elimination [<xref ref-type="bibr" rid="ref30">30</xref>] and generalized hyperbolic mixture modeling on principal components of actigraphy variables [<xref ref-type="bibr" rid="ref28">28</xref>]. Algorithm development or validation was the primary contribution of 2 studies: a chest-worn sleep-wake classifier developed against polysomnography [<xref ref-type="bibr" rid="ref26">26</xref>] and the polysomnographic validation of a consumer tracker’s proprietary scoring algorithm [<xref ref-type="bibr" rid="ref21">21</xref>]. The remaining 6 studies applied deterministic algorithms (nonparametric rest-activity indices, cosinor fitting, and automated sleep-wake scoring) followed by conventional statistical inference to relate circadian metrics to functional, clinical, or feasibility outcomes [<xref ref-type="bibr" rid="ref23">23</xref>,<xref ref-type="bibr" rid="ref25">25</xref>,<xref ref-type="bibr" rid="ref27">27</xref>,<xref ref-type="bibr" rid="ref29">29</xref>,<xref ref-type="bibr" rid="ref31">31</xref>,<xref ref-type="bibr" rid="ref33">33</xref>].</p>
          <p>Reporting of these methods was uneven (<xref ref-type="table" rid="table3">Table 3</xref>). No model was validated on an independent external cohort, and no study reported following an AI-specific reporting guideline; 3 cited general reporting guidelines (STROBE [Strengthening the Reporting of Observational Studies in Epidemiology] [<xref ref-type="bibr" rid="ref27">27</xref>,<xref ref-type="bibr" rid="ref28">28</xref>], TIDieR [Template for Intervention Description and Replication], and CONSORT [Consolidated Standards of Reporting Trials] [<xref ref-type="bibr" rid="ref33">33</xref>]). Seven studies fitted no predictive model, so validation, partitioning, and hyperparameter criteria did not apply to them [<xref ref-type="bibr" rid="ref21">21</xref>,<xref ref-type="bibr" rid="ref23">23</xref>,<xref ref-type="bibr" rid="ref25">25</xref>,<xref ref-type="bibr" rid="ref27">27</xref>,<xref ref-type="bibr" rid="ref29">29</xref>,<xref ref-type="bibr" rid="ref31">31</xref>,<xref ref-type="bibr" rid="ref33">33</xref>]. Of the 4 studies with a training-testing distinction, 2 stated that data were partitioned at the participant level [<xref ref-type="bibr" rid="ref22">22</xref>,<xref ref-type="bibr" rid="ref26">26</xref>]; 1 reported leave-one-out cross-validation over a participant-level feature matrix without stating the partition unit [<xref ref-type="bibr" rid="ref34">34</xref>]; and 1 described a train-test split and 5-fold cross-validation for its linear models without stating the unit or proportion of the split, while its deposited random forest code fitted the complete training set without resampling [<xref ref-type="bibr" rid="ref32">32</xref>]. The 2 unsupervised studies performed no held-out evaluation: one assessed cluster stability by repeated subsampling [<xref ref-type="bibr" rid="ref28">28</xref>], and the other reported no validation of the clustering solution [<xref ref-type="bibr" rid="ref30">30</xref>]. In total, 5 studies reported hyperparameter values partially [<xref ref-type="bibr" rid="ref22">22</xref>,<xref ref-type="bibr" rid="ref26">26</xref>,<xref ref-type="bibr" rid="ref28">28</xref>,<xref ref-type="bibr" rid="ref30">30</xref>,<xref ref-type="bibr" rid="ref34">34</xref>], and 1 relied on package defaults that are visible only in its deposited code [<xref ref-type="bibr" rid="ref32">32</xref>]; none reported every setting needed to reproduce its model. A total of 3 studies deposited analysis code, in each case covering only part of the analysis [<xref ref-type="bibr" rid="ref28">28</xref>,<xref ref-type="bibr" rid="ref32">32</xref>,<xref ref-type="bibr" rid="ref34">34</xref>], and 3 made their data openly available [<xref ref-type="bibr" rid="ref28">28</xref>,<xref ref-type="bibr" rid="ref30">30</xref>,<xref ref-type="bibr" rid="ref32">32</xref>].</p>
          <p>Explainability was addressed only through model-level variable importance. One study applied a post-hoc attribution method, permutation importance for its random forest [<xref ref-type="bibr" rid="ref32">32</xref>], and 3 reported model-internal importance measures: Gini importance [<xref ref-type="bibr" rid="ref22">22</xref>], variables ranked by their contribution to cluster separation [<xref ref-type="bibr" rid="ref28">28</xref>], and random forest importance within the feature-selection step that preceded clustering, rather than for the clustering itself [<xref ref-type="bibr" rid="ref30">30</xref>]. No study used local attribution methods such as Shapley Additive Explanations (SHAP) or Local Interpretable Model-Agnostic Explanations (LIME), and none explained individual predictions.</p>
          <table-wrap position="float" id="table3">
            <label>Table 3</label>
            <caption>
              <p>Reporting of computational methods in the included studies<sup>a</sup>.</p>
            </caption>
            <table width="1000" cellpadding="5" cellspacing="0" border="1" rules="groups" frame="hsides">
              <col width="170"/>
              <col width="150"/>
              <col width="160"/>
              <col width="110"/>
              <col width="140"/>
              <col width="70"/>
              <col width="90"/>
              <col width="110"/>
              <thead>
                <tr valign="top">
                  <td>Author (year)</td>
                  <td>AI/ML<sup>b</sup> family</td>
                  <td>Validation</td>
                  <td>Partition unit</td>
                  <td>Hyperparameters</td>
                  <td>Code</td>
                  <td>Data</td>
                  <td>XAI<sup>c</sup></td>
                </tr>
              </thead>
              <tbody>
                <tr valign="top">
                  <td>Chuang et al (2024) [<xref ref-type="bibr" rid="ref25">25</xref>]</td>
                  <td>Pattern association</td>
                  <td>Reference standard<sup>d</sup></td>
                  <td>N/A<sup>e</sup></td>
                  <td>N/A</td>
                  <td>NR<sup>f</sup></td>
                  <td>Request</td>
                  <td>N/A</td>
                </tr>
                <tr valign="top">
                  <td>Razjouyan et al (2017) [<xref ref-type="bibr" rid="ref26">26</xref>]</td>
                  <td>Algorithm validation</td>
                  <td>CV</td>
                  <td>Participant</td>
                  <td>Partial</td>
                  <td>NR</td>
                  <td>NR</td>
                  <td>NR</td>
                </tr>
                <tr valign="top">
                  <td>Gabb et al (2025) [<xref ref-type="bibr" rid="ref27">27</xref>]</td>
                  <td>Pattern association</td>
                  <td>N/A</td>
                  <td>N/A</td>
                  <td>N/A</td>
                  <td>NR</td>
                  <td>Request</td>
                  <td>N/A</td>
                </tr>
                <tr valign="top">
                  <td>Cavaillès et al (2025) [<xref ref-type="bibr" rid="ref28">28</xref>]</td>
                  <td>Unsupervised</td>
                  <td>Resubstitution<sup>g</sup></td>
                  <td>N/A</td>
                  <td>Partial</td>
                  <td>Open<sup>h</sup></td>
                  <td>Open</td>
                  <td>Importance</td>
                </tr>
                <tr valign="top">
                  <td>Wang et al (2024) [<xref ref-type="bibr" rid="ref29">29</xref>]</td>
                  <td>Pattern association</td>
                  <td>N/A</td>
                  <td>N/A</td>
                  <td>N/A</td>
                  <td>NR</td>
                  <td>Request</td>
                  <td>N/A</td>
                </tr>
                <tr valign="top">
                  <td>Shim et al (2023) [<xref ref-type="bibr" rid="ref30">30</xref>]</td>
                  <td>Unsupervised</td>
                  <td>NR</td>
                  <td>N/A</td>
                  <td>Partial</td>
                  <td>NR</td>
                  <td>Open</td>
                  <td>Importance<sup>i</sup></td>
                </tr>
                <tr valign="top">
                  <td>Jones et al (2025) [<xref ref-type="bibr" rid="ref31">31</xref>]</td>
                  <td>Pattern association</td>
                  <td>N/A</td>
                  <td>N/A</td>
                  <td>N/A</td>
                  <td>NR</td>
                  <td>NR</td>
                  <td>N/A</td>
                </tr>
                <tr valign="top">
                  <td>Beiglary et al (2025) [<xref ref-type="bibr" rid="ref32">32</xref>]</td>
                  <td>Supervised</td>
                  <td>Unclear<sup>j</sup></td>
                  <td>NR</td>
                  <td>Defaults<sup>k</sup></td>
                  <td>Open<sup>h</sup></td>
                  <td>Open</td>
                  <td>Post hoc</td>
                </tr>
                <tr valign="top">
                  <td>Merilahti et al (2016) [<xref ref-type="bibr" rid="ref23">23</xref>]<sup>l</sup></td>
                  <td>Pattern association</td>
                  <td>N/A</td>
                  <td>N/A</td>
                  <td>N/A</td>
                  <td>NR<sup>m</sup></td>
                  <td>NR</td>
                  <td>N/A</td>
                </tr>
                <tr valign="top">
                  <td>Froggatt et al (2020) [<xref ref-type="bibr" rid="ref33">33</xref>]</td>
                  <td>Pattern association</td>
                  <td>N/A</td>
                  <td>N/A</td>
                  <td>N/A</td>
                  <td>NR</td>
                  <td>Request</td>
                  <td>N/A</td>
                </tr>
                <tr valign="top">
                  <td>Sprint et al (2022) [<xref ref-type="bibr" rid="ref34">34</xref>]</td>
                  <td>Supervised</td>
                  <td>CV</td>
                  <td>Unclear<sup>n</sup></td>
                  <td>Partial</td>
                  <td>Open<sup>h</sup></td>
                  <td>NR</td>
                  <td>NR</td>
                </tr>
                <tr valign="top">
                  <td>de Zambotti et al (2015) [<xref ref-type="bibr" rid="ref21">21</xref>]<sup>l</sup></td>
                  <td>Algorithm validation<sup>o</sup></td>
                  <td>Reference standard</td>
                  <td>N/A</td>
                  <td>N/A</td>
                  <td>NR<sup>m</sup></td>
                  <td>NR</td>
                  <td>N/A</td>
                </tr>
                <tr valign="top">
                  <td>Robben et al (2017) [<xref ref-type="bibr" rid="ref22">22</xref>]<sup>l</sup></td>
                  <td>Supervised</td>
                  <td>CV</td>
                  <td>Participant</td>
                  <td>Partial</td>
                  <td>NR</td>
                  <td>NR</td>
                  <td>Importance<sup>p</sup></td>
                </tr>
              </tbody>
            </table>
            <table-wrap-foot>
              <fn id="table3fn1">
                <p><sup>a</sup>AI/ML family: supervised=supervised predictive modeling; unsupervised=unsupervised phenotyping or clustering; algorithm validation=algorithm development or validation; pattern association=pattern-association modeling with conventional statistical inference. Validation: CV=cross-validation; reference standard=comparison against a concurrent reference standard; resubstitution=performance reported on the fitting data. Partition unit: participant=data partitioned at the participant level. Hyperparameters: partial=some values reported; defaults=package defaults used. Code and data: open=publicly available; request=available on request. XAI: post hoc=post hoc attribution method; importance=variable importance without a named attribution framework; unclear=reporting insufficient to determine. No study reported external validation on an independent cohort or adherence to an AI-specific reporting guideline.</p>
              </fn>
              <fn id="table3fn2">
                <p><sup>b</sup>ML: machine learning.</p>
              </fn>
              <fn id="table3fn3">
                <p><sup>c</sup>XAI: explainable AI.</p>
              </fn>
              <fn id="table3fn4">
                <p><sup>d</sup>Reference standard was research-grade actigraphy, not polysomnography.</p>
              </fn>
              <fn id="table3fn5">
                <p><sup>e</sup>N/A: not applicable (no fitted predictive model).</p>
              </fn>
              <fn id="table3fn6">
                <p><sup>f</sup>NR: not reported.</p>
              </fn>
              <fn id="table3fn7">
                <p><sup>g</sup>Internal stability only (100 subsamples excluding 5% of participants; adjusted Rand index reported); no held-out performance evaluation.</p>
              </fn>
              <fn id="table3fn8">
                <p><sup>h</sup>Deposited code covers only part of the analysis: cluster identification [<xref ref-type="bibr" rid="ref28">28</xref>], the circadian light model [<xref ref-type="bibr" rid="ref32">32</xref>], and digital marker extraction [<xref ref-type="bibr" rid="ref34">34</xref>].</p>
              </fn>
              <fn id="table3fn9">
                <p><sup>i</sup>Variable importance reported for the random forest used in recursive feature elimination, not for the clustering model.</p>
              </fn>
              <fn id="table3fn10">
                <p><sup>j</sup>Five-fold cross-validation and a train-test split are described for the linear regression models; the deposited random forest code fits the full training set without resampling.</p>
              </fn>
              <fn id="table3fn11">
                <p><sup>k</sup>Package defaults evident in the deposited code; not stated in the text; no random seed.</p>
              </fn>
              <fn id="table3fn12">
                <p><sup>l</sup>Sensitivity-reference study.</p>
              </fn>
              <fn id="table3fn13">
                <p><sup>m</sup>The device algorithm is stated to be proprietary; no statement on the authors’ analysis code.</p>
              </fn>
              <fn id="table3fn14">
                <p><sup>n</sup>Leave-one-out cross-validation over a participant-level feature matrix; partition unit not stated explicitly.</p>
              </fn>
              <fn id="table3fn15">
                <p><sup>o</sup>Validation of a commercial algorithm against polysomnography; no algorithm was developed by the authors.</p>
              </fn>
              <fn id="table3fn16">
                <p><sup>p</sup>The authors also state that the learned decision rules are interpretable.</p>
              </fn>
            </table-wrap-foot>
          </table-wrap>
        </sec>
        <sec>
          <title>Population and Neurocognitive Context</title>
          <p>Six studies enrolled populations with documented or high-risk neurodegenerative conditions. Three targeted residents of nursing homes or assisted-living facilities with advanced dementia [<xref ref-type="bibr" rid="ref23">23</xref>,<xref ref-type="bibr" rid="ref32">32</xref>,<xref ref-type="bibr" rid="ref33">33</xref>], where circadian fragmentation was most severe (IS mean ~0.08, IV mean ~0.7 in Froggatt et al [<xref ref-type="bibr" rid="ref33">33</xref>]), and actigraphy feasibility was explicitly evaluated. Two enrolled participants with amnestic MCI [<xref ref-type="bibr" rid="ref31">31</xref>,<xref ref-type="bibr" rid="ref34">34</xref>], with circadian metrics linked to cognitive test scores and clinical assessment predictions. One enrolled a mixed MCI/dementia cohort alongside a cognitively normal control [<xref ref-type="bibr" rid="ref27">27</xref>]. Four studies used community-dwelling or population-based samples without documented cognitive impairment [<xref ref-type="bibr" rid="ref22">22</xref>,<xref ref-type="bibr" rid="ref28">28</xref>-<xref ref-type="bibr" rid="ref30">30</xref>], examining 24-hour activity profiles in relation to dementia risk, biological age, or functional health as outcome proxies. Three studies provided methodological contributions using middle-aged or mixed-age samples [<xref ref-type="bibr" rid="ref21">21</xref>,<xref ref-type="bibr" rid="ref25">25</xref>,<xref ref-type="bibr" rid="ref26">26</xref>].</p>
          <p>Sex composition was notably imbalanced: Cavaillès et al [<xref ref-type="bibr" rid="ref28">28</xref>] enrolled exclusively older men (Osteoporotic Fractures in Men Study [MrOS] cohort), and de Zambotti et al [<xref ref-type="bibr" rid="ref21">21</xref>] enrolled exclusively women. Racial and ethnic diversity was rarely reported. No study was conducted in low- or middle-income country settings, despite the global burden of neurodegeneration in those populations.</p>
          <p>The 4 analytical domains examined across this synthesis do not operate independently. Their recurrence across the included studies reveals a set of structural dependencies: sensing modality choices constrain the circadian constructs that can be extracted; construct operationalization specificity determines the inferential ceiling available to AI/ML methods; and the characteristics of the target population bound the generalizability of any biomarker claim. Taken together, these dependencies constitute the functional logic of a pipeline, not merely a descriptive taxonomy.</p>
          <p>Across domains, the evidence shows a consistent pattern: continuous circadian-related data capture is technically feasible, but biomarker interpretation remains constrained by construct heterogeneity, limited external validation, and population-level generalizability gaps. The most robust evidence comes from actigraphy-derived rest-activity fragmentation in institutional or cognitively vulnerable populations, whereas AI-based prediction in community and preclinical settings remains exploratory. The Evidence-Derived Proposal: Toward Clinically Deployable AI-Circadian Biomarker Systems: Structural Challenges, a Reference Pipeline Architecture, and Validation Priorities section translates that logic into a proposed reference architecture and layered validation pathway grounded in the evidence synthesized earlier.</p>
        </sec>
      </sec>
      <sec>
        <title>Evidence-Derived Proposal: Toward Clinically Deployable AI-Circadian Biomarker Systems: Structural Challenges, a Reference Pipeline Architecture, and Validation Priorities</title>
        <p>The structural challenges, reference architecture, and validation pathway that follow were derived by the authors from the synthesized evidence; they are not outcomes of the screening and appraisal process described in the Methods section.</p>
        <sec>
          <title>Structural Challenges at the Sensing-to-Biomarker Interface</title>
          <p>The 13-study corpus reveals a field in which technical capacity has outpaced analytical standardization. Six structural challenges constrain the transition from ecologically valid continuous monitoring to validated predictive biomarkers for neurodegenerative disease, and each aligns with a specific stage of the pipeline that connects sensing hardware to clinical interpretation.</p>
          <p>At the sensing layer, device heterogeneity is compounded by inconsistent within-study validation. Only 3 studies [<xref ref-type="bibr" rid="ref21">21</xref>,<xref ref-type="bibr" rid="ref26">26</xref>,<xref ref-type="bibr" rid="ref32">32</xref>] performed explicit concurrent validation against a reference standard (polysomnography or laboratory spectrophotometry), while the remaining studies assumed device validity in populations for which independent performance evidence is often lacking. This limitation intensifies in advanced dementia, where motor disorganization, behavioral symptoms, and adherence constraints may invalidate calibration obtained from younger or healthier samples, as illustrated by the 40% device loss reported by Beiglary et al [<xref ref-type="bibr" rid="ref32">32</xref>].</p>
          <p>At the parameter extraction layer, circadian end-point heterogeneity prevents cross-study integration. Studies using nonparametric metrics (IS, IV, and RA), cosinor-derived parameters (MESOR, acrophase, and rhythm strength), circadian stimulus, and unsupervised activity phenotypes differ not only in end-point choice but also in epoch length, analysis software, and time window, precluding quantitative synthesis despite consistent directional findings. A parallel concern is the gradient from explicit chronobiological operationalization to implicit proxies based on sleep-wake rhythm or behavioral regularity; the distinction matters because only the former supports direct circadian biomarker inference; yet, it is frequently blurred in cross-study interpretation.</p>
          <p>At the AI/ML inference layer, 2 interlocking challenges stand out. The first is the absence of external validation, compounded by incomplete reporting of how models were validated. Every model in the corpus was developed within a single dataset, and none was evaluated on an independent cohort; in 2 of the 4 studies with a training-testing distinction, the unit of data partitioning could not be determined from the report; no study reported every hyperparameter needed for reproduction; and none followed an AI-specific reporting guideline such as TRIPOD+AI [<xref ref-type="bibr" rid="ref16">16</xref>] (<xref ref-type="table" rid="table3">Table 3</xref>). Even the large-sample clustering studies [<xref ref-type="bibr" rid="ref28">28</xref>,<xref ref-type="bibr" rid="ref30">30</xref>] relied on retrospective datasets without prospective biomarker validation designs. The same pattern has been documented for AI-based risk prediction in sports medicine, where only 5 of 39 injury prediction studies reported external validation [<xref ref-type="bibr" rid="ref35">35</xref>], which points to a methodological gap shared across physiological monitoring domains rather than one specific to circadian research. The second challenge is the shallowness of explainability. Four studies reported variable importance [<xref ref-type="bibr" rid="ref22">22</xref>,<xref ref-type="bibr" rid="ref28">28</xref>,<xref ref-type="bibr" rid="ref30">30</xref>,<xref ref-type="bibr" rid="ref32">32</xref>], but only at the level of the whole model: no study used local attribution methods such as SHAP [<xref ref-type="bibr" rid="ref36">36</xref>] or LIME [<xref ref-type="bibr" rid="ref37">37</xref>], and none explained individual predictions, which is the level at which a clinician would act on a biomarker result. Converging regulatory expectations require transparency and human oversight before AI-based biomarkers are deployed clinically; for circadian models intended to inform neurodegeneration risk, this gap is a direct barrier to regulatory clearance and clinician trust.</p>
          <p>At the population and deployment layer, 2 further challenges emerge. All included studies were conducted in high-income countries, with limited reporting of racial and ethnic diversity and pronounced sex imbalances, with exclusively male [<xref ref-type="bibr" rid="ref28">28</xref>] or exclusively female [<xref ref-type="bibr" rid="ref21">21</xref>] samples. Given that the global burden of neurodegeneration falls disproportionately on populations underrepresented in this corpus, biomarkers validated in this sample frame cannot be assumed to transfer to the populations for whom early detection is most urgent. Compounding this, the descriptive-to-predictive gap limits clinical actionability: 7 of the 13 studies fitted no predictive model, applying deterministic algorithms for feature extraction followed by conventional statistical inference, and the 6 that did were small, single-site, or retrospective (<xref ref-type="table" rid="table3">Table 3</xref>).</p>
          <p>These 6 challenges are not independent; they constrain one another. Device validation failures propagate into parameter extraction errors, which in turn exacerbate AI model overfitting and confound the association with clinical outcomes. Addressing them individually is insufficient; they must be addressed in structural relation to the pipeline that generates a circadian biomarker (<xref ref-type="table" rid="table4">Table 4</xref>).</p>
          <table-wrap position="float" id="table4">
            <label>Table 4</label>
            <caption>
              <p>Structural challenges at the sensing-to-biomarker interface and their architectural responses.</p>
            </caption>
            <table width="1000" cellpadding="5" cellspacing="0" border="1" rules="groups" frame="hsides">
              <col width="240"/>
              <col width="180"/>
              <col width="290"/>
              <col width="290"/>
              <thead>
                <tr valign="top">
                  <td>Identified challenge/gap</td>
                  <td>Why it matters</td>
                  <td>Architectural/design response</td>
                  <td>Future validation implication</td>
                </tr>
              </thead>
              <tbody>
                <tr valign="top">
                  <td>Device validation heterogeneity across populations</td>
                  <td>Calibration from younger or healthier samples may not hold in advanced dementia, where device loss and motion artifact are substantial</td>
                  <td>Layer 1: concurrent validation against reference standards in the target population as a precondition, not an add-on</td>
                  <td>Population-specific device validation required before biomarker claims</td>
                </tr>
                <tr valign="top">
                  <td>Circadian end-point heterogeneity (metrics, epoch lengths, and pipelines)</td>
                  <td>Prevents quantitative synthesis and direct effect-size comparison across studies</td>
                  <td>Layer 2: harmonized parameter set, standardized analysis pipelines, explicit distinction between direct circadian measurement and behavioral proxies</td>
                  <td>Minimum standard reporting of end points, epochs, and pipelines in biomarker studies</td>
                </tr>
                <tr valign="top">
                  <td>Absence of external validation and incomplete AI reporting</td>
                  <td>All included AI models are developed within a single dataset; validation procedures and hyperparameters are incompletely reported, and no study follows an AI-specific reporting guideline</td>
                  <td>Layer 3: external replication on independent cohorts as a design requirement</td>
                  <td>Cross-cohort generalization evidence and guideline-concordant reporting (eg, TRIPOD+AI<sup>a</sup>) before clinical translation claims</td>
                </tr>
                <tr valign="top">
                  <td>Explainable AI (XAI) gap</td>
                  <td>Regulatory deployment and clinical trust require transparency; importance reporting in the corpus is model-level only, with no explanation of individual predictions</td>
                  <td>Layer 3: explainability mechanisms integrated at design time, with model outputs mapped to chronobiologically interpretable parameters</td>
                  <td>Prespecified XAI reporting, including explanations of individual predictions (eg, SHAP<sup>b</sup> and LIME<sup>c</sup>), as a standard element of biomarker studies</td>
                </tr>
                <tr valign="top">
                  <td>Population diversity deficit</td>
                  <td>High-income, demographically narrow samples do not represent the global burden of neurodegeneration</td>
                  <td>Layer 4: demographic representativeness as an architectural concern, not a discussion caveat</td>
                  <td>Quantified transferability across sex, race, ethnicity, and country-income setting as a validation criterion</td>
                </tr>
                <tr valign="top">
                  <td>Descriptive-to-predictive gap</td>
                  <td>In total, 7 of 13 studies fit no predictive model; the 6 that do are small, single-site, or retrospective</td>
                  <td>Layer 3: principled distinction between algorithmic processing and learned representation, with implications for evidence strength</td>
                  <td>Prospective biomarker validation designs rather than retrospective cross-sectional association</td>
                </tr>
              </tbody>
            </table>
            <table-wrap-foot>
              <fn id="table4fn1">
                <p><sup>a</sup>TRIPOD+AI: Transparent Reporting of a Multivariable Prediction Model for Individual Prognosis or Diagnosis + Artificial Intelligence</p>
              </fn>
              <fn id="table4fn2">
                <p><sup>b</sup>SHAP: Shapley Additive Explanations.</p>
              </fn>
              <fn id="table4fn3">
                <p><sup>c</sup>LIME: Local Interpretable Model-Agnostic Explanations.</p>
              </fn>
            </table-wrap-foot>
          </table-wrap>
        </sec>
        <sec>
          <title>A Reference Pipeline Architecture for AI-Enhanced Circadian Monitoring</title>
          <sec>
            <title>Overview</title>
            <p>The recurrence of 4 analytical domains across the reviewed corpus supports a 4-layer reference architecture for AI-enhanced circadian biomarker pipelines (as <xref rid="figure2" ref-type="fig">Figure 2</xref> shows). The architecture is conceptual: it does not prescribe a specific implementation but articulates the functional dependencies that any credible biomarker system in this field must respect.</p>
            <fig id="figure2" position="float">
              <label>Figure 2</label>
              <caption>
                <p>Reference pipeline architecture for AI-enhanced circadian monitoring. EEG: electroencephalography; GGIR: R package for processing raw accelerometer data; IS: interdaily stability; IV: intradaily variability; LIME: Local Interpretable Model-Agnostic Explanations; MESOR: midline estimating statistic of rhythm; ML: machine learning; RA: relative amplitude; SHAP: Shapley Additive Explanations; STROBE: Strengthening the Reporting of Observational Studies in Epidemiology; TRIPOD+AI: Transparent Reporting of a Multivariable Prediction Model for Individual Prognosis or Diagnosis+Artificial Intelligence.</p>
              </caption>
              <graphic xlink:href="aging_v9i1e83031_fig2.png" alt-version="no" mimetype="image" position="float" xlink:type="simple"/>
            </fig>
          </sec>
          <sec>
            <title>Sensing and Acquisition Layer</title>
            <p>This layer encompasses the modalities observed in the corpus: wrist actigraphy as the dominant channel, supplemented by EEG headbands, multisensor wearables, calibrated wearable light sensors, ambient smart-home sensor arrays, and smartphone-based platforms. Its function is to produce continuous, ecologically valid physiological and behavioral data streams with explicit, population-specific validation against reference measurements. The cross-cutting concerns at this layer are device calibration, data completeness under real-world conditions, and feasibility in the populations for whom the biomarker is intended, particularly institutionalized individuals with advanced dementia, where adherence, motion artifact, and device loss are nontrivial [<xref ref-type="bibr" rid="ref32">32</xref>,<xref ref-type="bibr" rid="ref33">33</xref>]. Calls to standardize how sensor technologies and the algorithms built on them are assessed have also been made in applied fields outside medicine [<xref ref-type="bibr" rid="ref38">38</xref>]; in circadian biomarker pipelines, this standardization matters most at layers 1 and 2, because device validation and parameter extraction set the ceiling for every inference downstream.</p>
          </sec>
          <sec>
            <title>Circadian Parameter Extraction Layer</title>
            <p>This layer converts raw sensor streams into chronobiologically interpretable metrics. The corpus supports a core parameter set comprising nonparametric metrics (IS, IV, and RA), cosinor-derived parameters (MESOR, acrophase, and rhythm strength), and specialized metrics such as circadian stimulus for light exposure studies. Its function is to produce reproducible, comparable circadian end points. The cross-cutting concerns are harmonized epoch length, standardized analysis pipelines (such as GGIR or equivalently documented alternatives), explicit operationalization of whether a study measures circadian biology directly or uses behavioral proxies, and transparent reporting that enables downstream meta-analytic integration.</p>
          </sec>
          <sec>
            <title>AI/ML Inference Layer</title>
            <p>This layer applies computational reasoning to extracted parameters or, in more advanced designs, directly to raw sensor streams. The corpus demonstrates 4 viable method families: supervised predictive modeling, unsupervised phenotyping, algorithm development and validation, and pattern-association modeling combined with conventional statistical inference. Its function is to extract clinically meaningful representations beyond the reach of single-metric description. The cross-cutting concerns are external validation across independent datasets, explainability mechanisms compatible with regulatory expectations, overfitting control under small-sample conditions, and the principled distinction between genuine learned representations and algorithmic feature engineering, a distinction that directly affects how results should be interpreted as biomarker evidence.</p>
          </sec>
          <sec>
            <title>Clinical and Validation Interface Layer</title>
            <p>This layer situates pipeline outputs within clinical, epidemiological, and regulatory contexts. Its function is to connect circadian biomarker outputs to actionable constructs: functional decline tracking, dementia risk stratification, cognitive assessment prediction, and preclinical pathology screening. The cross-cutting concerns are population representativeness across sex, race, ethnicity, socioeconomic status, and geography; regulatory alignment (including XAI requirements and medical device pathways); governance of longitudinal biometric data; and the evidence certainty calibration that distinguishes moderate-certainty institutional findings from lower-certainty community preclinical claims.</p>
            <p>The layers are not merely sequential; they are mutually constraining. Layer 1 decisions determine the circadian constructs that layer 2 can produce. Layer 2 decisions determine the inferential ceiling available to layer 3. Layer 3 decisions determine the evidence strength with which layer 4 claims can be made. This dependency structure is the architectural translation of the cross-layer failure propagation described in the Structural Challenges at the Sensing-to-Biomarker Interface subsection.</p>
          </sec>
        </sec>
        <sec>
          <title>Conceptual Validation Logic and Implementation Propositions</title>
          <p>The proposed architecture suggests a layered validation logic that may guide subsequent empirical work. Rather than evaluating AI-circadian biomarker systems solely by end-point predictive accuracy, each layer should be assessed against criteria appropriate to its function, with the overall pipeline’s evidence certainty bounded by its weakest layer.</p>
          <p>At layer 1, validation should establish device performance in the specific population of interest, with concurrent comparison against an accepted reference (polysomnography, laboratory spectrophotometry, or supervised clinical observation, depending on modality) and transparent reporting of data completeness, device loss, and calibration drift over the monitoring period. At layer 2, validation should demonstrate that extracted circadian parameters are reproducible under standardized pipelines and interpretable within chronobiological theory, with an explicit statement of whether the study measures circadian biology directly or uses a behavioral proxy. At layer 3, validation should require external replication on independent cohorts, prespecified explainability reporting, and honest characterization of overfitting risk, particularly when sample sizes approach those observed in most included clinical studies. At layer 4, validation should evidence diagnostic or prognostic utility in prospectively followed, demographically representative cohorts, with certainty grading that respects the gradient already visible in the present corpus: moderate certainty for actigraphy-derived fragmentation-function associations in institutional settings and low certainty for AI-based predictive claims in community or preclinical populations.</p>
          <p>From this logic, 4 implementation propositions follow directly from the evidence base. First, prospective longitudinal multisite cohort studies using harmonized circadian end points (nonparametric metrics and cosinor parameters computed with consistent epoch lengths and transparently reported pipelines) are required across cognitively normal aging, MCI, and dementia populations to move the field beyond retrospective and cross-sectional evidence. Second, AI models intended for clinical deployment should incorporate explainability mechanisms at design time rather than retrofit them, with explicit mapping between model outputs and chronobiologically interpretable parameters. Third, device and algorithm validation should be treated as population-specific rather than generic: performance in younger or healthier samples does not warrant extrapolation to institutionalized older adults or individuals with advanced dementia, where the biomarker is most clinically consequential. Fourth, demographic diversity should be operationalized as a validation criterion (not an ethical aspiration) with biomarker transferability quantified across sex, race, ethnicity, and country-income setting before generalizability is claimed.</p>
          <p>Taken together, the challenge synthesis, reference architecture, and layered validation logic articulate what a clinically deployable AI–circadian biomarker system would require, and where the existing evidence in this review stops short of those requirements. Circadian disruption remains a biologically plausible, technically measurable, and increasingly computable candidate biomarker for neurodegeneration. Its maturation into a validated clinical tool depends less on additional sensing innovation than on structural coherence across the pipeline from device to clinical interface, on external and population-specific validation of AI components, and on explainability and governance frameworks commensurate with the clinical stakes.</p>
        </sec>
      </sec>
    </sec>
    <sec sec-type="discussion">
      <title>Discussion</title>
      <sec>
        <title>Principal Findings</title>
        <p>Building on the cross-domain analysis, 3 convergent findings emerge from the 13 studies examining AI-enhanced wearable and objective sensing systems for circadian monitoring in older or cognitively vulnerable populations. First, circadian fragmentation (measurable as elevated IV, reduced IS, and attenuated RA from wrist actigraphy) is consistently documented in populations with advanced dementia and is associated with concurrent functional decline [<xref ref-type="bibr" rid="ref23">23</xref>,<xref ref-type="bibr" rid="ref33">33</xref>]. Second, AI and ML contribute meaningful value beyond descriptive actigraphy by enabling automated phenotyping of 24-hour activity profiles at a population scale [<xref ref-type="bibr" rid="ref28">28</xref>,<xref ref-type="bibr" rid="ref30">30</xref>], calibrated prediction of circadian light exposure [<xref ref-type="bibr" rid="ref32">32</xref>], and multimodal fusion of behavioral markers into clinically interpretable predictions [<xref ref-type="bibr" rid="ref34">34</xref>]. Third, the field’s technical capacity to collect ecologically valid continuous data has advanced faster than its ability to standardize circadian end points, harmonize device-agnostic reporting, and independently validate AI models across diverse populations.</p>
      </sec>
      <sec>
        <title>Contextual Interpretation and Recent Evidence</title>
        <p>The association between actigraphy-derived circadian fragmentation and dementia risk is supported by prospective cohort evidence extending well beyond this review’s corpus. Lim et al [<xref ref-type="bibr" rid="ref39">39</xref>] and Musiek et al [<xref ref-type="bibr" rid="ref40">40</xref>] demonstrated that rest-activity fragmentation predicts incident dementia and tracks preclinical Alzheimer pathology, supporting the biological plausibility of IS and IV as early biomarkers. The present review advances this framing by showing that AI approaches (particularly unsupervised clustering [<xref ref-type="bibr" rid="ref28">28</xref>,<xref ref-type="bibr" rid="ref30">30</xref>] and supervised learning [<xref ref-type="bibr" rid="ref32">32</xref>]) can extract richer, phenotypically meaningful representations from these data streams that go beyond single-metric descriptors.</p>
        <p>The multimodal sensing approach in Gabb et al [<xref ref-type="bibr" rid="ref27">27</xref>] and Sprint et al [<xref ref-type="bibr" rid="ref34">34</xref>] aligns with emerging frameworks for remote monitoring of cognitive decline that combine multiple behavioral data streams to improve predictive validity. However, neither study achieved external validation, and both were small. In Sprint et al [<xref ref-type="bibr" rid="ref34">34</xref>], leave-one-out cross-validation on 14 participants leaves substantial overfitting risk; Gabb et al [<xref ref-type="bibr" rid="ref27">27</xref>] fitted no predictive model, so its contribution is feasibility evidence rather than model performance.</p>
        <p>Explainability was addressed only partially. One study applied permutation importance to a random forest [<xref ref-type="bibr" rid="ref32">32</xref>], and 3 reported model-internal variable importance [<xref ref-type="bibr" rid="ref22">22</xref>,<xref ref-type="bibr" rid="ref28">28</xref>,<xref ref-type="bibr" rid="ref30">30</xref>]; no study used local attribution methods such as SHAP [<xref ref-type="bibr" rid="ref36">36</xref>] or LIME [<xref ref-type="bibr" rid="ref37">37</xref>], and none explained individual predictions. Regulatory frameworks, including the US Food and Drug Administration’s action plan for AI/ML-based software as a medical device [<xref ref-type="bibr" rid="ref41">41</xref>], emphasize that clinical deployment of AI-based biomarkers requires transparency, interpretability, and human oversight, expectations that already apply to a growing number of approved AI/ML-enabled devices [<xref ref-type="bibr" rid="ref42">42</xref>]. Knowing which hours of the day or which light-sensor channels drive a model helps during development, but a clinician deciding whether one patient’s rest-activity rhythm signals elevated risk needs to know why the model flagged that patient. Circadian biomarker models that cannot give that account are unlikely to earn clinical trust or regulatory clearance in neurodegenerative contexts.</p>
        <p>Population diversity remains a critical shortcoming. The Lancet Commission on Dementia has underscored that the global burden of neurodegenerative disease falls disproportionately on populations underrepresented in biomarker research [<xref ref-type="bibr" rid="ref43">43</xref>]. All included studies were conducted in high-income countries, with limited reporting of racial and ethnic diversity. Wearable-based circadian biomarkers validated exclusively in these populations cannot be assumed to generalize to the populations most affected by neurodegeneration worldwide. In ML terms, this is a problem of dataset shift [<xref ref-type="bibr" rid="ref44">44</xref>]: a model learns the relationship between sensor signals and outcomes in the population it was trained on, and that relationship may not hold when age, sex, ethnicity, comorbidity, or device-wearing behavior are distributed differently. Performance in the development sample is therefore not evidence of performance in underrepresented groups.</p>
      </sec>
      <sec>
        <title>Limitations of the Evidence and of This Review</title>
        <p>Sample sizes were predominantly small (n=16 to ~50 in clinical studies), with adequate statistical power only in the 2 large-scale epidemiological studies [<xref ref-type="bibr" rid="ref28">28</xref>,<xref ref-type="bibr" rid="ref30">30</xref>] that used retrospective data without prospective biomarker validation designs. Sex imbalances (male-only in MrOS [<xref ref-type="bibr" rid="ref28">28</xref>]; female-only in de Zambotti et al [<xref ref-type="bibr" rid="ref21">21</xref>]) limit generalizability. In total, 7 of the 13 studies fitted no predictive model, so conclusions about AI model performance rest on 6 studies [<xref ref-type="bibr" rid="ref22">22</xref>,<xref ref-type="bibr" rid="ref26">26</xref>,<xref ref-type="bibr" rid="ref28">28</xref>,<xref ref-type="bibr" rid="ref30">30</xref>,<xref ref-type="bibr" rid="ref32">32</xref>,<xref ref-type="bibr" rid="ref34">34</xref>], each developed within a single dataset and none externally validated. The assessment of methods reporting relied on the published articles and their supplementary materials; details available only in protocols or repositories not referenced by the articles were not captured. Circadian end-point heterogeneity across studies (different parametric conventions, epoch lengths, device positions, analysis software, and time windows) prevented quantitative synthesis and precluded direct cross-study comparison of effect estimates. Certainty in the evidence is moderate for actigraphy-derived fragmentation-function associations in institutional settings and low to very low for AI-based circadian biomarker claims in community or preclinical populations.</p>
        <p>Regarding the review itself, the original search did not cover IEEE Xplore or the ACM Digital Library, which were added in the complementary search following peer-review recommendations. The complementary search is a documented amendment to the deposited protocol, as described in the Methods. The reference architecture and validation pathway are an interpretive synthesis by the authors; they were not produced by the systematic review process and have not been empirically validated.</p>
      </sec>
      <sec>
        <title>Implications and Future Directions</title>
        <p>Four research priorities are indicated for the field to progress toward clinically deployable AI-based circadian biomarkers. First, prospective longitudinal cohort studies with standardized circadian metric reporting (including IS, IV, RA, and cosinor parameters calculated with harmonized epoch lengths and analysis pipelines) are needed across cognitively normal aging, MCI, and dementia populations. Second, AI models should incorporate explainability mechanisms to facilitate clinical interpretation and regulatory compliance. Third, device validation must be population-specific, accounting for the performance differences expected in very old, institutionalized, or demented populations. Fourth, population diversity in sex, race, ethnicity, and socioeconomic status must be actively prioritized to support generalizable biomarker development.</p>
        <p>The structural interdependence of these 4 priorities (across sensing, parameter extraction, AI/ML inference, and clinical validation) motivates the reference architecture and layered validation pathway developed in the previous section, which maps each challenge to its corresponding pipeline layer and proposes implementation-oriented responses grounded in the reviewed evidence.</p>
      </sec>
      <sec>
        <title>Conclusions</title>
        <p>The 13-study corpus reviewed here documents a field in productive transition. AI-enhanced wearable and objective sensing devices can measure circadian fragmentation, derive 24-hour activity phenotypes, estimate circadian light exposure, and predict clinical assessments in older and cognitively vulnerable populations. Current evidence is strongest in institutional settings using actigraphy-derived nonparametric circadian metrics, where fragmentation is most pronounced and most consistently associated with functional outcomes. Circadian disruption, quantified through objective wearable sensing and AI-based analysis, constitutes a biologically plausible and technically measurable candidate biomarker for neurodegeneration. Transitioning this candidate into a validated predictive tool requires prospective longitudinal validation, cross-device harmonization, and transparent XAI approaches that the existing literature does not yet provide. The structural challenges, reference architecture, and layered validation pathway developed from this synthesis provide a conceptual map for researchers and system designers working toward that transition.</p>
      </sec>
    </sec>
  </body>
  <back>
    <app-group>
      <supplementary-material id="app1">
        <label>Multimedia Appendix 1</label>
        <p>PRISMA checklist.</p>
        <media xlink:href="aging_v9i1e83031_app1.pdf" xlink:title="PDF File  (Adobe PDF File), 254 KB"/>
      </supplementary-material>
      <supplementary-material id="app2">
        <label>Multimedia Appendix 2</label>
        <p>Search strategy (PRISMA-S).</p>
        <media xlink:href="aging_v9i1e83031_app2.pdf" xlink:title="PDF File  (Adobe PDF File), 112 KB"/>
      </supplementary-material>
      <supplementary-material id="app3">
        <label>Multimedia Appendix 3</label>
        <p>Mixed Methods Appraisal Tool.</p>
        <media xlink:href="aging_v9i1e83031_app3.xlsx" xlink:title="XLSX File  (Microsoft Excel File), 15 KB"/>
      </supplementary-material>
    </app-group>
    <glossary>
      <title>Abbreviations</title>
      <def-list>
        <def-item>
          <term id="abb1">AD</term>
          <def>
            <p>Alzheimer disease</p>
          </def>
        </def-item>
        <def-item>
          <term id="abb2">CLAIM</term>
          <def>
            <p>Checklist for Artificial Intelligence in Medical Imaging</p>
          </def>
        </def-item>
        <def-item>
          <term id="abb3">CONSORT</term>
          <def>
            <p>Consolidated Standards of Reporting Trials</p>
          </def>
        </def-item>
        <def-item>
          <term id="abb4">CONSORT-AI</term>
          <def>
            <p>Consolidated Standards of Reporting Trials-Artificial Intelligence</p>
          </def>
        </def-item>
        <def-item>
          <term id="abb5">DECIDE-AI</term>
          <def>
            <p>Developmental and Exploratory Clinical Investigations of Decision Support Systems Driven by Artificial Intelligence</p>
          </def>
        </def-item>
        <def-item>
          <term id="abb6">EEG</term>
          <def>
            <p>electroencephalography</p>
          </def>
        </def-item>
        <def-item>
          <term id="abb7">IS</term>
          <def>
            <p>interdaily stability</p>
          </def>
        </def-item>
        <def-item>
          <term id="abb8">IV</term>
          <def>
            <p>intradaily variability</p>
          </def>
        </def-item>
        <def-item>
          <term id="abb9">LIME</term>
          <def>
            <p>Local Interpretable Model-Agnostic Explanations</p>
          </def>
        </def-item>
        <def-item>
          <term id="abb10">MESOR</term>
          <def>
            <p>midline estimating statistic of rhythm</p>
          </def>
        </def-item>
        <def-item>
          <term id="abb11">ML</term>
          <def>
            <p>machine learning</p>
          </def>
        </def-item>
        <def-item>
          <term id="abb12">MMAT</term>
          <def>
            <p>Mixed Methods Appraisal Tool</p>
          </def>
        </def-item>
        <def-item>
          <term id="abb13">MrOS</term>
          <def>
            <p>Osteoporotic Fractures in Men Study</p>
          </def>
        </def-item>
        <def-item>
          <term id="abb14">PD</term>
          <def>
            <p>Parkinson disease</p>
          </def>
        </def-item>
        <def-item>
          <term id="abb15">PRISMA</term>
          <def>
            <p>Preferred Reporting Items for Systematic Reviews and Meta-Analyses</p>
          </def>
        </def-item>
        <def-item>
          <term id="abb16">PRISMA-S</term>
          <def>
            <p>Preferred Reporting Items for Systematic Reviews and Meta-Analyses Literature Search Extension</p>
          </def>
        </def-item>
        <def-item>
          <term id="abb17">RA</term>
          <def>
            <p>relative amplitude</p>
          </def>
        </def-item>
        <def-item>
          <term id="abb18">RCT</term>
          <def>
            <p>randomized controlled trial</p>
          </def>
        </def-item>
        <def-item>
          <term id="abb19">SHAP</term>
          <def>
            <p>Shapley Additive Explanations</p>
          </def>
        </def-item>
        <def-item>
          <term id="abb20">STROBE</term>
          <def>
            <p>Strengthening the Reporting of Observational Studies in Epidemiology</p>
          </def>
        </def-item>
        <def-item>
          <term id="abb21">TIDieR</term>
          <def>
            <p>Template for Intervention Description and Replication</p>
          </def>
        </def-item>
        <def-item>
          <term id="abb22">TRIPOD+AI</term>
          <def>
            <p>Transparent Reporting of a Multivariable Prediction Model for Individual Prognosis or Diagnosis+Artificial Intelligence</p>
          </def>
        </def-item>
        <def-item>
          <term id="abb23">XAI</term>
          <def>
            <p>explainable AI</p>
          </def>
        </def-item>
      </def-list>
    </glossary>
    <ack>
      <p>During the preparation of this manuscript, the authors used Claude (Anthropic) and Grammarly to refine language and improve clarity. An agentic workflow using Claude was also used to produce a first-pass draft of the extraction of computational-method reporting summarized in <xref ref-type="table" rid="table3">Table 3</xref>, as described in the Methods section; every extracted value was verified independently by 2 authors (JMH-Z and PCS-M) against the source articles and their supplementary materials. The authors reviewed and edited all AI-generated content and take full responsibility for the final version of the manuscript.</p>
    </ack>
    <notes>
      <sec>
        <title>Funding</title>
        <p>The authors declared no financial support was received for this work.</p>
      </sec>
    </notes>
    <notes>
      <sec>
        <title>Data Availability</title>
        <p>All data generated or analyzed during this study are included in this published article and its supplementary information files (<xref ref-type="supplementary-material" rid="app1">Multimedia Appendices 1</xref>-<xref ref-type="supplementary-material" rid="app3">3</xref>). The title-abstract and full-text screening records are available from the corresponding author on reasonable request.</p>
      </sec>
    </notes>
    <fn-group>
      <fn fn-type="con">
        <p>All authors contributed equally to the conception and design of this review. The literature search and analysis were performed collaboratively. The manuscript was written by JMH-Z, JM-R, JG-M, and PCS-M, and critically reviewed and edited by all authors. All authors read and approved the final version of the manuscript.</p>
      </fn>
      <fn fn-type="conflict">
        <p>PCS-M is an associate editor of <italic>JMIR Human Factors</italic> at the time of this publication. All other authors declare no conflicts of interest.</p>
      </fn>
    </fn-group>
    <ref-list>
      <ref id="ref1">
        <label>1</label>
        <nlm-citation citation-type="journal">
          <person-group person-group-type="author">
            <collab>GBD 2019 Dementia Forecasting Collaborators</collab>
          </person-group>
          <article-title>Estimation of the global prevalence of dementia in 2019 and forecasted prevalence in 2050: an analysis for the Global Burden of Disease Study 2019</article-title>
          <source>Lancet Public Health</source>
          <year>2022</year>
          <volume>7</volume>
          <issue>2</issue>
          <fpage>e105</fpage>
          <lpage>e125</lpage>
          <comment>
            <ext-link ext-link-type="uri" xlink:type="simple" xlink:href="https://linkinghub.elsevier.com/retrieve/pii/S2468-2667(21)00249-8"/>
          </comment>
          <pub-id pub-id-type="doi">10.1016/S2468-2667(21)00249-8</pub-id>
          <pub-id pub-id-type="medline">34998485</pub-id>
          <pub-id pub-id-type="pii">S2468-2667(21)00249-8</pub-id>
          <pub-id pub-id-type="pmcid">PMC8810394</pub-id>
        </nlm-citation>
      </ref>
      <ref id="ref2">
        <label>2</label>
        <nlm-citation citation-type="journal">
          <person-group person-group-type="author">
            <name name-style="western">
              <surname>Hood</surname>
              <given-names>S</given-names>
            </name>
            <name name-style="western">
              <surname>Amir</surname>
              <given-names>S</given-names>
            </name>
          </person-group>
          <article-title>Neurodegeneration and the circadian clock</article-title>
          <source>Front Aging Neurosci</source>
          <year>2017</year>
          <volume>9</volume>
          <fpage>170</fpage>
          <comment>
            <ext-link ext-link-type="uri" xlink:type="simple" xlink:href="https://europepmc.org/abstract/MED/28611660"/>
          </comment>
          <pub-id pub-id-type="doi">10.3389/fnagi.2017.00170</pub-id>
          <pub-id pub-id-type="medline">28611660</pub-id>
          <pub-id pub-id-type="pmcid">PMC5447688</pub-id>
        </nlm-citation>
      </ref>
      <ref id="ref3">
        <label>3</label>
        <nlm-citation citation-type="journal">
          <person-group person-group-type="author">
            <name name-style="western">
              <surname>Aubourg</surname>
              <given-names>T</given-names>
            </name>
            <name name-style="western">
              <surname>Demongeot</surname>
              <given-names>J</given-names>
            </name>
            <name name-style="western">
              <surname>Provost</surname>
              <given-names>H</given-names>
            </name>
            <name name-style="western">
              <surname>Vuillerme</surname>
              <given-names>N</given-names>
            </name>
          </person-group>
          <article-title>Exploitation of outgoing and incoming telephone calls in the context of circadian rhythms of social activity among elderly people: observational descriptive study</article-title>
          <source>JMIR Mhealth Uhealth</source>
          <year>2020</year>
          <volume>8</volume>
          <issue>11</issue>
          <fpage>e13535</fpage>
          <comment>
            <ext-link ext-link-type="uri" xlink:type="simple" xlink:href="https://mhealth.jmir.org/2020/11/e13535/"/>
          </comment>
          <pub-id pub-id-type="doi">10.2196/13535</pub-id>
          <pub-id pub-id-type="medline">33242018</pub-id>
          <pub-id pub-id-type="pii">v8i11e13535</pub-id>
          <pub-id pub-id-type="pmcid">PMC7728541</pub-id>
        </nlm-citation>
      </ref>
      <ref id="ref4">
        <label>4</label>
        <nlm-citation citation-type="journal">
          <person-group person-group-type="author">
            <name name-style="western">
              <surname>Dose</surname>
              <given-names>B</given-names>
            </name>
            <name name-style="western">
              <surname>Yalçin</surname>
              <given-names>M</given-names>
            </name>
            <name name-style="western">
              <surname>Dries</surname>
              <given-names>SPM</given-names>
            </name>
            <name name-style="western">
              <surname>Relógio</surname>
              <given-names>A</given-names>
            </name>
          </person-group>
          <article-title>TimeTeller for timing health: the potential of circadian medicine to improve performance, prevent disease and optimize treatment</article-title>
          <source>Front Digit Health</source>
          <year>2023</year>
          <volume>5</volume>
          <fpage>1157654</fpage>
          <comment>
            <ext-link ext-link-type="uri" xlink:type="simple" xlink:href="https://europepmc.org/abstract/MED/37153516"/>
          </comment>
          <pub-id pub-id-type="doi">10.3389/fdgth.2023.1157654</pub-id>
          <pub-id pub-id-type="medline">37153516</pub-id>
          <pub-id pub-id-type="pmcid">PMC10155816</pub-id>
        </nlm-citation>
      </ref>
      <ref id="ref5">
        <label>5</label>
        <nlm-citation citation-type="journal">
          <person-group person-group-type="author">
            <name name-style="western">
              <surname>Fifel</surname>
              <given-names>K</given-names>
            </name>
            <name name-style="western">
              <surname>Videnovic</surname>
              <given-names>A</given-names>
            </name>
          </person-group>
          <article-title>Circadian and sleep dysfunctions in neurodegenerative disorders—an update</article-title>
          <source>Front Neurosci</source>
          <year>2020</year>
          <volume>14</volume>
          <fpage>627330</fpage>
          <comment>
            <ext-link ext-link-type="uri" xlink:type="simple" xlink:href="https://europepmc.org/abstract/MED/33536872"/>
          </comment>
          <pub-id pub-id-type="doi">10.3389/fnins.2020.627330</pub-id>
          <pub-id pub-id-type="medline">33536872</pub-id>
          <pub-id pub-id-type="pmcid">PMC7848154</pub-id>
        </nlm-citation>
      </ref>
      <ref id="ref6">
        <label>6</label>
        <nlm-citation citation-type="journal">
          <person-group person-group-type="author">
            <name name-style="western">
              <surname>Dhahbi</surname>
              <given-names>W</given-names>
            </name>
            <name name-style="western">
              <surname>Chamari</surname>
              <given-names>K</given-names>
            </name>
            <name name-style="western">
              <surname>Dergaa</surname>
              <given-names>I</given-names>
            </name>
            <name name-style="western">
              <surname>Trabelsi</surname>
              <given-names>K</given-names>
            </name>
            <name name-style="western">
              <surname>BaHammam</surname>
              <given-names>AS</given-names>
            </name>
            <name name-style="western">
              <surname>Ben Saad</surname>
              <given-names>H</given-names>
            </name>
          </person-group>
          <article-title>Sleep under fire: towards a tripartite circadian-neuroendocrine framework for conflict-zone athletes</article-title>
          <source>Br J Sports Med</source>
          <year>2026</year>
          <volume>60</volume>
          <issue>19</issue>
          <fpage>1453</fpage>
          <lpage>1456</lpage>
          <pub-id pub-id-type="doi">10.1136/bjsports-2026-111904</pub-id>
          <pub-id pub-id-type="medline">42242882</pub-id>
          <pub-id pub-id-type="pii">bjsports-2026-111904</pub-id>
        </nlm-citation>
      </ref>
      <ref id="ref7">
        <label>7</label>
        <nlm-citation citation-type="journal">
          <person-group person-group-type="author">
            <name name-style="western">
              <surname>Reddy</surname>
              <given-names>A</given-names>
            </name>
            <name name-style="western">
              <surname>Reddy</surname>
              <given-names>RP</given-names>
            </name>
            <name name-style="western">
              <surname>Roghani</surname>
              <given-names>AK</given-names>
            </name>
            <name name-style="western">
              <surname>Garcia</surname>
              <given-names>RI</given-names>
            </name>
            <name name-style="western">
              <surname>Khemka</surname>
              <given-names>S</given-names>
            </name>
            <name name-style="western">
              <surname>Pattoor</surname>
              <given-names>V</given-names>
            </name>
            <name name-style="western">
              <surname>Jacob</surname>
              <given-names>M</given-names>
            </name>
            <name name-style="western">
              <surname>Reddy</surname>
              <given-names>PH</given-names>
            </name>
            <name name-style="western">
              <surname>Sehar</surname>
              <given-names>U</given-names>
            </name>
          </person-group>
          <article-title>Artificial intelligence in parkinson's disease: early detection and diagnostic advancements</article-title>
          <source>Ageing Res Rev</source>
          <year>2024</year>
          <volume>99</volume>
          <fpage>102410</fpage>
          <pub-id pub-id-type="doi">10.1016/j.arr.2024.102410</pub-id>
          <pub-id pub-id-type="medline">38972602</pub-id>
          <pub-id pub-id-type="pii">S1568-1637(24)00228-9</pub-id>
        </nlm-citation>
      </ref>
      <ref id="ref8">
        <label>8</label>
        <nlm-citation citation-type="journal">
          <person-group person-group-type="author">
            <name name-style="western">
              <surname>Page</surname>
              <given-names>MJ</given-names>
            </name>
            <name name-style="western">
              <surname>McKenzie</surname>
              <given-names>JE</given-names>
            </name>
            <name name-style="western">
              <surname>Bossuyt</surname>
              <given-names>PM</given-names>
            </name>
            <name name-style="western">
              <surname>Boutron</surname>
              <given-names>I</given-names>
            </name>
            <name name-style="western">
              <surname>Hoffmann</surname>
              <given-names>TC</given-names>
            </name>
            <name name-style="western">
              <surname>Mulrow</surname>
              <given-names>CD</given-names>
            </name>
            <name name-style="western">
              <surname>Shamseer</surname>
              <given-names>L</given-names>
            </name>
            <name name-style="western">
              <surname>Tetzlaff</surname>
              <given-names>JM</given-names>
            </name>
            <name name-style="western">
              <surname>Akl</surname>
              <given-names>EA</given-names>
            </name>
            <name name-style="western">
              <surname>Brennan</surname>
              <given-names>SE</given-names>
            </name>
            <name name-style="western">
              <surname>Chou</surname>
              <given-names>R</given-names>
            </name>
            <name name-style="western">
              <surname>Glanville</surname>
              <given-names>J</given-names>
            </name>
            <name name-style="western">
              <surname>Grimshaw</surname>
              <given-names>JM</given-names>
            </name>
            <name name-style="western">
              <surname>Hróbjartsson</surname>
              <given-names>A</given-names>
            </name>
            <name name-style="western">
              <surname>Lalu</surname>
              <given-names>MM</given-names>
            </name>
            <name name-style="western">
              <surname>Li</surname>
              <given-names>T</given-names>
            </name>
            <name name-style="western">
              <surname>Loder</surname>
              <given-names>EW</given-names>
            </name>
            <name name-style="western">
              <surname>Mayo-Wilson</surname>
              <given-names>E</given-names>
            </name>
            <name name-style="western">
              <surname>McDonald</surname>
              <given-names>S</given-names>
            </name>
            <name name-style="western">
              <surname>McGuinness</surname>
              <given-names>LA</given-names>
            </name>
            <name name-style="western">
              <surname>Stewart</surname>
              <given-names>LA</given-names>
            </name>
            <name name-style="western">
              <surname>Thomas</surname>
              <given-names>J</given-names>
            </name>
            <name name-style="western">
              <surname>Tricco</surname>
              <given-names>AC</given-names>
            </name>
            <name name-style="western">
              <surname>Welch</surname>
              <given-names>VA</given-names>
            </name>
            <name name-style="western">
              <surname>Whiting</surname>
              <given-names>P</given-names>
            </name>
            <name name-style="western">
              <surname>Moher</surname>
              <given-names>D</given-names>
            </name>
          </person-group>
          <article-title>The PRISMA 2020 statement: an updated guideline for reporting systematic reviews</article-title>
          <source>BMJ</source>
          <year>2021</year>
          <volume>372</volume>
          <fpage>n71</fpage>
          <comment>
            <ext-link ext-link-type="uri" xlink:type="simple" xlink:href="https://www.bmj.com/lookup/pmidlookup?view=long&amp;pmid=33782057"/>
          </comment>
          <pub-id pub-id-type="doi">10.1136/bmj.n71</pub-id>
          <pub-id pub-id-type="medline">33782057</pub-id>
          <pub-id pub-id-type="pmcid">PMC8005924</pub-id>
        </nlm-citation>
      </ref>
      <ref id="ref9">
        <label>9</label>
        <nlm-citation citation-type="web">
          <person-group person-group-type="author">
            <name name-style="western">
              <surname>Herrera</surname>
              <given-names>JM</given-names>
            </name>
            <name name-style="western">
              <surname>Monroy-Rodriguez</surname>
              <given-names>J</given-names>
            </name>
            <name name-style="western">
              <surname>Santana-Mancilla</surname>
              <given-names>PC</given-names>
            </name>
          </person-group>
          <article-title>Artificial intelligence-based monitoring devices for the identification of circadian parameters as digital predictive biomarkers of neurodegenerative diseases in older adults</article-title>
          <source>Zenodo</source>
          <year>2025</year>
          <access-date>2025-11-10</access-date>
          <comment>
            <ext-link ext-link-type="uri" xlink:type="simple" xlink:href="https://zenodo.org/records/16883416">https://zenodo.org/records/16883416</ext-link>
          </comment>
        </nlm-citation>
      </ref>
      <ref id="ref10">
        <label>10</label>
        <nlm-citation citation-type="journal">
          <person-group person-group-type="author">
            <name name-style="western">
              <surname>Rethlefsen</surname>
              <given-names>ML</given-names>
            </name>
            <name name-style="western">
              <surname>Kirtley</surname>
              <given-names>S</given-names>
            </name>
            <name name-style="western">
              <surname>Waffenschmidt</surname>
              <given-names>S</given-names>
            </name>
            <name name-style="western">
              <surname>Ayala</surname>
              <given-names>AP</given-names>
            </name>
            <name name-style="western">
              <surname>Moher</surname>
              <given-names>D</given-names>
            </name>
            <name name-style="western">
              <surname>Page</surname>
              <given-names>MJ</given-names>
            </name>
            <name name-style="western">
              <surname>Koffel</surname>
              <given-names>JB</given-names>
            </name>
            <collab>PRISMA-S Group</collab>
          </person-group>
          <article-title>PRISMA-S: an extension to the PRISMA Statement for Reporting Literature Searches in Systematic Reviews</article-title>
          <source>Syst Rev</source>
          <year>2021</year>
          <volume>10</volume>
          <issue>1</issue>
          <fpage>39</fpage>
          <comment>
            <ext-link ext-link-type="uri" xlink:type="simple" xlink:href="https://systematicreviewsjournal.biomedcentral.com/articles/10.1186/s13643-020-01542-z"/>
          </comment>
          <pub-id pub-id-type="doi">10.1186/s13643-020-01542-z</pub-id>
          <pub-id pub-id-type="medline">33499930</pub-id>
          <pub-id pub-id-type="pii">10.1186/s13643-020-01542-z</pub-id>
          <pub-id pub-id-type="pmcid">PMC7839230</pub-id>
        </nlm-citation>
      </ref>
      <ref id="ref11">
        <label>11</label>
        <nlm-citation citation-type="journal">
          <person-group person-group-type="author">
            <name name-style="western">
              <surname>Wei</surname>
              <given-names>J</given-names>
            </name>
            <name name-style="western">
              <surname>Boger</surname>
              <given-names>J</given-names>
            </name>
          </person-group>
          <article-title>Sleep detection for younger adults, healthy older adults, and older adults living with dementia using wrist temperature and actigraphy: prototype testing and case study analysis</article-title>
          <source>JMIR Mhealth Uhealth</source>
          <year>2021</year>
          <volume>9</volume>
          <issue>6</issue>
          <fpage>e26462</fpage>
          <comment>
            <ext-link ext-link-type="uri" xlink:type="simple" xlink:href="https://mhealth.jmir.org/2021/6/e26462/"/>
          </comment>
          <pub-id pub-id-type="doi">10.2196/26462</pub-id>
          <pub-id pub-id-type="medline">34061038</pub-id>
          <pub-id pub-id-type="pii">v9i6e26462</pub-id>
          <pub-id pub-id-type="pmcid">PMC8207254</pub-id>
        </nlm-citation>
      </ref>
      <ref id="ref12">
        <label>12</label>
        <nlm-citation citation-type="journal">
          <person-group person-group-type="author">
            <name name-style="western">
              <surname>Kuula</surname>
              <given-names>L</given-names>
            </name>
            <name name-style="western">
              <surname>Gradisar</surname>
              <given-names>M</given-names>
            </name>
            <name name-style="western">
              <surname>Martinmäki</surname>
              <given-names>K</given-names>
            </name>
            <name name-style="western">
              <surname>Richardson</surname>
              <given-names>C</given-names>
            </name>
            <name name-style="western">
              <surname>Bonnar</surname>
              <given-names>D</given-names>
            </name>
            <name name-style="western">
              <surname>Bartel</surname>
              <given-names>K</given-names>
            </name>
            <name name-style="western">
              <surname>Lang</surname>
              <given-names>C</given-names>
            </name>
            <name name-style="western">
              <surname>Leinonen</surname>
              <given-names>L</given-names>
            </name>
            <name name-style="western">
              <surname>Pesonen</surname>
              <given-names>A</given-names>
            </name>
          </person-group>
          <article-title>Using big data to explore worldwide trends in objective sleep in the transition to adulthood</article-title>
          <source>Sleep Med</source>
          <year>2019</year>
          <volume>62</volume>
          <fpage>69</fpage>
          <lpage>76</lpage>
          <pub-id pub-id-type="doi">10.1016/j.sleep.2019.07.024</pub-id>
          <pub-id pub-id-type="medline">31563008</pub-id>
          <pub-id pub-id-type="pii">S1389-9457(19)30262-X</pub-id>
        </nlm-citation>
      </ref>
      <ref id="ref13">
        <label>13</label>
        <nlm-citation citation-type="journal">
          <person-group person-group-type="author">
            <name name-style="western">
              <surname>Cai</surname>
              <given-names>R</given-names>
            </name>
            <name name-style="western">
              <surname>Chao</surname>
              <given-names>J</given-names>
            </name>
            <name name-style="western">
              <surname>Gao</surname>
              <given-names>C</given-names>
            </name>
            <name name-style="western">
              <surname>Gao</surname>
              <given-names>L</given-names>
            </name>
            <name name-style="western">
              <surname>Hu</surname>
              <given-names>K</given-names>
            </name>
            <name name-style="western">
              <surname>Li</surname>
              <given-names>P</given-names>
            </name>
          </person-group>
          <article-title>Association between sleep duration and cognitive frailty in older Chinese adults: prospective cohort study</article-title>
          <source>JMIR Aging</source>
          <year>2025</year>
          <volume>8</volume>
          <fpage>e65183</fpage>
          <comment>
            <ext-link ext-link-type="uri" xlink:type="simple" xlink:href="https://aging.jmir.org/2025//e65183/"/>
          </comment>
          <pub-id pub-id-type="doi">10.2196/65183</pub-id>
          <pub-id pub-id-type="medline">40267503</pub-id>
          <pub-id pub-id-type="pii">v8i1e65183</pub-id>
          <pub-id pub-id-type="pmcid">PMC12043274</pub-id>
        </nlm-citation>
      </ref>
      <ref id="ref14">
        <label>14</label>
        <nlm-citation citation-type="journal">
          <person-group person-group-type="author">
            <name name-style="western">
              <surname>Tsanas</surname>
              <given-names>A</given-names>
            </name>
            <name name-style="western">
              <surname>Woodward</surname>
              <given-names>E</given-names>
            </name>
            <name name-style="western">
              <surname>Ehlers</surname>
              <given-names>A</given-names>
            </name>
          </person-group>
          <article-title>Objective characterization of activity, sleep, and circadian rhythm patterns using a wrist-worn actigraphy sensor: insights into posttraumatic stress disorder</article-title>
          <source>JMIR Mhealth Uhealth</source>
          <year>2020</year>
          <volume>8</volume>
          <issue>4</issue>
          <fpage>e14306</fpage>
          <comment>
            <ext-link ext-link-type="uri" xlink:type="simple" xlink:href="https://mhealth.jmir.org/2020/4/e14306/"/>
          </comment>
          <pub-id pub-id-type="doi">10.2196/14306</pub-id>
          <pub-id pub-id-type="medline">32310142</pub-id>
          <pub-id pub-id-type="pii">v8i4e14306</pub-id>
          <pub-id pub-id-type="pmcid">PMC7199134</pub-id>
        </nlm-citation>
      </ref>
      <ref id="ref15">
        <label>15</label>
        <nlm-citation citation-type="journal">
          <person-group person-group-type="author">
            <name name-style="western">
              <surname>Hu</surname>
              <given-names>B</given-names>
            </name>
            <name name-style="western">
              <surname>Dixon</surname>
              <given-names>PC</given-names>
            </name>
            <name name-style="western">
              <surname>Jacobs</surname>
              <given-names>JV</given-names>
            </name>
            <name name-style="western">
              <surname>Dennerlein</surname>
              <given-names>JT</given-names>
            </name>
            <name name-style="western">
              <surname>Schiffman</surname>
              <given-names>JM</given-names>
            </name>
          </person-group>
          <article-title>Machine learning algorithms based on signals from a single wearable inertial sensor can detect surface- and age-related differences in walking</article-title>
          <source>J Biomech</source>
          <year>2018</year>
          <volume>71</volume>
          <fpage>37</fpage>
          <lpage>42</lpage>
          <pub-id pub-id-type="doi">10.1016/j.jbiomech.2018.01.005</pub-id>
          <pub-id pub-id-type="medline">29452755</pub-id>
          <pub-id pub-id-type="pii">S0021-9290(18)30019-8</pub-id>
        </nlm-citation>
      </ref>
      <ref id="ref16">
        <label>16</label>
        <nlm-citation citation-type="journal">
          <person-group person-group-type="author">
            <name name-style="western">
              <surname>Collins</surname>
              <given-names>GS</given-names>
            </name>
            <name name-style="western">
              <surname>Moons</surname>
              <given-names>KGM</given-names>
            </name>
            <name name-style="western">
              <surname>Dhiman</surname>
              <given-names>P</given-names>
            </name>
            <name name-style="western">
              <surname>Riley</surname>
              <given-names>RD</given-names>
            </name>
            <name name-style="western">
              <surname>Beam</surname>
              <given-names>AL</given-names>
            </name>
            <name name-style="western">
              <surname>Van Calster</surname>
              <given-names>B</given-names>
            </name>
            <name name-style="western">
              <surname>Ghassemi</surname>
              <given-names>M</given-names>
            </name>
            <name name-style="western">
              <surname>Liu</surname>
              <given-names>X</given-names>
            </name>
            <name name-style="western">
              <surname>Reitsma</surname>
              <given-names>B</given-names>
            </name>
            <name name-style="western">
              <surname>van Smeden</surname>
              <given-names>M</given-names>
            </name>
            <name name-style="western">
              <surname>Boulesteix</surname>
              <given-names>A-L</given-names>
            </name>
            <name name-style="western">
              <surname>Camaradou</surname>
              <given-names>JC</given-names>
            </name>
            <name name-style="western">
              <surname>Celi</surname>
              <given-names>LA</given-names>
            </name>
            <name name-style="western">
              <surname>Denaxas</surname>
              <given-names>S</given-names>
            </name>
            <name name-style="western">
              <surname>Denniston</surname>
              <given-names>AK</given-names>
            </name>
            <name name-style="western">
              <surname>Glocker</surname>
              <given-names>B</given-names>
            </name>
            <name name-style="western">
              <surname>Golub</surname>
              <given-names>RM</given-names>
            </name>
            <name name-style="western">
              <surname>Harvey</surname>
              <given-names>H</given-names>
            </name>
            <name name-style="western">
              <surname>Heinze</surname>
              <given-names>G</given-names>
            </name>
            <name name-style="western">
              <surname>Hoffman</surname>
              <given-names>MM</given-names>
            </name>
            <name name-style="western">
              <surname>Kengne</surname>
              <given-names>AP</given-names>
            </name>
            <name name-style="western">
              <surname>Lam</surname>
              <given-names>E</given-names>
            </name>
            <name name-style="western">
              <surname>Lee</surname>
              <given-names>N</given-names>
            </name>
            <name name-style="western">
              <surname>Loder</surname>
              <given-names>EW</given-names>
            </name>
            <name name-style="western">
              <surname>Maier-Hein</surname>
              <given-names>L</given-names>
            </name>
            <name name-style="western">
              <surname>Mateen</surname>
              <given-names>BA</given-names>
            </name>
            <name name-style="western">
              <surname>McCradden</surname>
              <given-names>MD</given-names>
            </name>
            <name name-style="western">
              <surname>Oakden-Rayner</surname>
              <given-names>L</given-names>
            </name>
            <name name-style="western">
              <surname>Ordish</surname>
              <given-names>J</given-names>
            </name>
            <name name-style="western">
              <surname>Parnell</surname>
              <given-names>R</given-names>
            </name>
            <name name-style="western">
              <surname>Rose</surname>
              <given-names>S</given-names>
            </name>
            <name name-style="western">
              <surname>Singh</surname>
              <given-names>K</given-names>
            </name>
            <name name-style="western">
              <surname>Wynants</surname>
              <given-names>L</given-names>
            </name>
            <name name-style="western">
              <surname>Logullo</surname>
              <given-names>P</given-names>
            </name>
          </person-group>
          <article-title>TRIPOD+AI statement: updated guidance for reporting clinical prediction models that use regression or machine learning methods</article-title>
          <source>BMJ</source>
          <year>2024</year>
          <volume>385</volume>
          <fpage>e078378</fpage>
          <comment>
            <ext-link ext-link-type="uri" xlink:type="simple" xlink:href="https://www.bmj.com/lookup/pmidlookup?view=long&amp;pmid=38626948"/>
          </comment>
          <pub-id pub-id-type="doi">10.1136/bmj-2023-078378</pub-id>
          <pub-id pub-id-type="medline">38626948</pub-id>
          <pub-id pub-id-type="pmcid">PMC11019967</pub-id>
        </nlm-citation>
      </ref>
      <ref id="ref17">
        <label>17</label>
        <nlm-citation citation-type="journal">
          <person-group person-group-type="author">
            <name name-style="western">
              <surname>Tejani</surname>
              <given-names>AS</given-names>
            </name>
            <name name-style="western">
              <surname>Klontzas</surname>
              <given-names>ME</given-names>
            </name>
            <name name-style="western">
              <surname>Gatti</surname>
              <given-names>AA</given-names>
            </name>
            <name name-style="western">
              <surname>Mongan</surname>
              <given-names>JT</given-names>
            </name>
            <name name-style="western">
              <surname>Moy</surname>
              <given-names>L</given-names>
            </name>
            <name name-style="western">
              <surname>Park</surname>
              <given-names>SH</given-names>
            </name>
            <name name-style="western">
              <surname>Kahn</surname>
              <given-names>CE</given-names>
            </name>
            <name name-style="western">
              <surname>Abbara</surname>
              <given-names>S</given-names>
            </name>
            <name name-style="western">
              <surname>Afat</surname>
              <given-names>S</given-names>
            </name>
          </person-group>
          <article-title>Checklist for Artificial Intelligence in Medical Imaging (CLAIM): 2024 update</article-title>
          <source>Radiol Artif Intell</source>
          <year>2024</year>
          <volume>6</volume>
          <issue>4</issue>
          <fpage>e240300</fpage>
          <comment>
            <ext-link ext-link-type="uri" xlink:type="simple" xlink:href="https://escholarship.org/uc/item/qt9zf6q2rb"/>
          </comment>
          <pub-id pub-id-type="doi">10.1148/ryai.240300</pub-id>
          <pub-id pub-id-type="medline">38809149</pub-id>
          <pub-id pub-id-type="pmcid">PMC11304031</pub-id>
        </nlm-citation>
      </ref>
      <ref id="ref18">
        <label>18</label>
        <nlm-citation citation-type="journal">
          <person-group person-group-type="author">
            <name name-style="western">
              <surname>Vasey</surname>
              <given-names>B</given-names>
            </name>
            <name name-style="western">
              <surname>Nagendran</surname>
              <given-names>M</given-names>
            </name>
            <name name-style="western">
              <surname>Campbell</surname>
              <given-names>B</given-names>
            </name>
            <name name-style="western">
              <surname>Clifton</surname>
              <given-names>DA</given-names>
            </name>
            <name name-style="western">
              <surname>Collins</surname>
              <given-names>GS</given-names>
            </name>
            <name name-style="western">
              <surname>Denaxas</surname>
              <given-names>S</given-names>
            </name>
            <name name-style="western">
              <surname>Denniston</surname>
              <given-names>AK</given-names>
            </name>
            <name name-style="western">
              <surname>Faes</surname>
              <given-names>L</given-names>
            </name>
            <name name-style="western">
              <surname>Geerts</surname>
              <given-names>B</given-names>
            </name>
            <name name-style="western">
              <surname>Ibrahim</surname>
              <given-names>M</given-names>
            </name>
            <name name-style="western">
              <surname>Liu</surname>
              <given-names>X</given-names>
            </name>
            <name name-style="western">
              <surname>Mateen</surname>
              <given-names>BA</given-names>
            </name>
            <name name-style="western">
              <surname>Mathur</surname>
              <given-names>P</given-names>
            </name>
            <name name-style="western">
              <surname>McCradden</surname>
              <given-names>MD</given-names>
            </name>
            <name name-style="western">
              <surname>Morgan</surname>
              <given-names>L</given-names>
            </name>
            <name name-style="western">
              <surname>Ordish</surname>
              <given-names>J</given-names>
            </name>
            <name name-style="western">
              <surname>Rogers</surname>
              <given-names>C</given-names>
            </name>
            <name name-style="western">
              <surname>Saria</surname>
              <given-names>S</given-names>
            </name>
            <name name-style="western">
              <surname>Ting</surname>
              <given-names>DSW</given-names>
            </name>
            <name name-style="western">
              <surname>Watkinson</surname>
              <given-names>P</given-names>
            </name>
            <name name-style="western">
              <surname>Weber</surname>
              <given-names>W</given-names>
            </name>
            <name name-style="western">
              <surname>Wheatstone</surname>
              <given-names>P</given-names>
            </name>
            <name name-style="western">
              <surname>McCulloch</surname>
              <given-names>P</given-names>
            </name>
          </person-group>
          <article-title>Reporting guideline for the early-stage clinical evaluation of decision support systems driven by artificial intelligence: DECIDE-AI</article-title>
          <source>Nat Med</source>
          <year>2022</year>
          <volume>28</volume>
          <issue>5</issue>
          <fpage>924</fpage>
          <lpage>933</lpage>
          <pub-id pub-id-type="doi">10.1038/s41591-022-01772-9</pub-id>
          <pub-id pub-id-type="medline">35585198</pub-id>
          <pub-id pub-id-type="pii">10.1038/s41591-022-01772-9</pub-id>
        </nlm-citation>
      </ref>
      <ref id="ref19">
        <label>19</label>
        <nlm-citation citation-type="journal">
          <person-group person-group-type="author">
            <name name-style="western">
              <surname>Liu</surname>
              <given-names>X</given-names>
            </name>
            <name name-style="western">
              <surname>Cruz Rivera</surname>
              <given-names>S</given-names>
            </name>
            <name name-style="western">
              <surname>Moher</surname>
              <given-names>D</given-names>
            </name>
            <name name-style="western">
              <surname>Calvert</surname>
              <given-names>MJ</given-names>
            </name>
            <name name-style="western">
              <surname>Denniston</surname>
              <given-names>AK</given-names>
            </name>
            <collab>SPIRIT-AICONSORT-AI Working Group</collab>
          </person-group>
          <article-title>Reporting guidelines for clinical trial reports for interventions involving artificial intelligence: the CONSORT-AI extension</article-title>
          <source>Nat Med</source>
          <year>2020</year>
          <volume>26</volume>
          <issue>9</issue>
          <fpage>1364</fpage>
          <lpage>1374</lpage>
          <comment>
            <ext-link ext-link-type="uri" xlink:type="simple" xlink:href="https://europepmc.org/abstract/MED/32908283"/>
          </comment>
          <pub-id pub-id-type="doi">10.1038/s41591-020-1034-x</pub-id>
          <pub-id pub-id-type="medline">32908283</pub-id>
          <pub-id pub-id-type="pii">10.1038/s41591-020-1034-x</pub-id>
          <pub-id pub-id-type="pmcid">PMC7598943</pub-id>
        </nlm-citation>
      </ref>
      <ref id="ref20">
        <label>20</label>
        <nlm-citation citation-type="journal">
          <person-group person-group-type="author">
            <name name-style="western">
              <surname>Ben Saad</surname>
              <given-names>H</given-names>
            </name>
            <name name-style="western">
              <surname>Dergaa</surname>
              <given-names>I</given-names>
            </name>
            <name name-style="western">
              <surname>Ghouili</surname>
              <given-names>H</given-names>
            </name>
            <name name-style="western">
              <surname>Ceylan</surname>
              <given-names>H?</given-names>
            </name>
            <name name-style="western">
              <surname>Chamari</surname>
              <given-names>K</given-names>
            </name>
            <name name-style="western">
              <surname>Dhahbi</surname>
              <given-names>W</given-names>
            </name>
          </person-group>
          <article-title>The assisted technology dilemma: a reflection on AI chatbots use and risks while reshaping the peer review process in scientific research</article-title>
          <source>AI Soc</source>
          <year>2025</year>
          <volume>40</volume>
          <issue>7</issue>
          <fpage>5649</fpage>
          <lpage>5656</lpage>
          <pub-id pub-id-type="doi">10.1007/s00146-025-02299-6</pub-id>
        </nlm-citation>
      </ref>
      <ref id="ref21">
        <label>21</label>
        <nlm-citation citation-type="journal">
          <person-group person-group-type="author">
            <name name-style="western">
              <surname>de Zambotti</surname>
              <given-names>M</given-names>
            </name>
            <name name-style="western">
              <surname>Claudatos</surname>
              <given-names>S</given-names>
            </name>
            <name name-style="western">
              <surname>Inkelis</surname>
              <given-names>S</given-names>
            </name>
            <name name-style="western">
              <surname>Colrain</surname>
              <given-names>IM</given-names>
            </name>
            <name name-style="western">
              <surname>Baker</surname>
              <given-names>FC</given-names>
            </name>
          </person-group>
          <article-title>Evaluation of a consumer fitness-tracking device to assess sleep in adults</article-title>
          <source>Chronobiol Int</source>
          <year>2015</year>
          <volume>32</volume>
          <issue>7</issue>
          <fpage>1024</fpage>
          <lpage>1028</lpage>
          <comment>
            <ext-link ext-link-type="uri" xlink:type="simple" xlink:href="https://europepmc.org/abstract/MED/26158542"/>
          </comment>
          <pub-id pub-id-type="doi">10.3109/07420528.2015.1054395</pub-id>
          <pub-id pub-id-type="medline">26158542</pub-id>
          <pub-id pub-id-type="pmcid">PMC4780439</pub-id>
        </nlm-citation>
      </ref>
      <ref id="ref22">
        <label>22</label>
        <nlm-citation citation-type="journal">
          <person-group person-group-type="author">
            <name name-style="western">
              <surname>Robben</surname>
              <given-names>S</given-names>
            </name>
            <name name-style="western">
              <surname>Englebienne</surname>
              <given-names>G</given-names>
            </name>
            <name name-style="western">
              <surname>Krose</surname>
              <given-names>B</given-names>
            </name>
          </person-group>
          <article-title>Delta features from ambient sensor data are good predictors of change in functional health</article-title>
          <source>IEEE J Biomed Health Inform</source>
          <year>2017</year>
          <volume>21</volume>
          <issue>4</issue>
          <fpage>986</fpage>
          <lpage>993</lpage>
          <pub-id pub-id-type="doi">10.1109/JBHI.2016.2593980</pub-id>
          <pub-id pub-id-type="medline">27455530</pub-id>
        </nlm-citation>
      </ref>
      <ref id="ref23">
        <label>23</label>
        <nlm-citation citation-type="journal">
          <person-group person-group-type="author">
            <name name-style="western">
              <surname>Merilahti</surname>
              <given-names>J</given-names>
            </name>
            <name name-style="western">
              <surname>Viramo</surname>
              <given-names>P</given-names>
            </name>
            <name name-style="western">
              <surname>Korhonen</surname>
              <given-names>I</given-names>
            </name>
          </person-group>
          <article-title>Wearable monitoring of physical functioning and disability changes, circadian rhythms and sleep patterns in nursing home residents</article-title>
          <source>IEEE J Biomed Health Inform</source>
          <year>2016</year>
          <volume>20</volume>
          <issue>3</issue>
          <fpage>856</fpage>
          <lpage>864</lpage>
          <pub-id pub-id-type="doi">10.1109/JBHI.2015.2420680</pub-id>
          <pub-id pub-id-type="medline">25861091</pub-id>
        </nlm-citation>
      </ref>
      <ref id="ref24">
        <label>24</label>
        <nlm-citation citation-type="journal">
          <person-group person-group-type="author">
            <name name-style="western">
              <surname>Hong</surname>
              <given-names>QN</given-names>
            </name>
            <name name-style="western">
              <surname>Fàbregues</surname>
              <given-names>S</given-names>
            </name>
            <name name-style="western">
              <surname>Bartlett</surname>
              <given-names>G</given-names>
            </name>
            <name name-style="western">
              <surname>Boardman</surname>
              <given-names>F</given-names>
            </name>
            <name name-style="western">
              <surname>Cargo</surname>
              <given-names>M</given-names>
            </name>
            <name name-style="western">
              <surname>Dagenais</surname>
              <given-names>P</given-names>
            </name>
            <name name-style="western">
              <surname>Gagnon</surname>
              <given-names>M</given-names>
            </name>
            <name name-style="western">
              <surname>Griffiths</surname>
              <given-names>F</given-names>
            </name>
            <name name-style="western">
              <surname>Nicolau</surname>
              <given-names>B</given-names>
            </name>
            <name name-style="western">
              <surname>O’Cathain</surname>
              <given-names>A</given-names>
            </name>
            <name name-style="western">
              <surname>Rousseau</surname>
              <given-names>M</given-names>
            </name>
            <name name-style="western">
              <surname>Vedel</surname>
              <given-names>I</given-names>
            </name>
            <name name-style="western">
              <surname>Pluye</surname>
              <given-names>P</given-names>
            </name>
          </person-group>
          <article-title>The Mixed Methods Appraisal Tool (MMAT) version 2018 for information professionals and researchers</article-title>
          <source>EFI</source>
          <year>2018</year>
          <volume>34</volume>
          <issue>4</issue>
          <fpage>285</fpage>
          <lpage>291</lpage>
          <pub-id pub-id-type="doi">10.3233/EFI-180221</pub-id>
        </nlm-citation>
      </ref>
      <ref id="ref25">
        <label>25</label>
        <nlm-citation citation-type="journal">
          <person-group person-group-type="author">
            <name name-style="western">
              <surname>Chuang</surname>
              <given-names>H</given-names>
            </name>
            <name name-style="western">
              <surname>Lin</surname>
              <given-names>C</given-names>
            </name>
            <name name-style="western">
              <surname>Lee</surname>
              <given-names>L</given-names>
            </name>
            <name name-style="western">
              <surname>Chang</surname>
              <given-names>H</given-names>
            </name>
            <name name-style="western">
              <surname>She</surname>
              <given-names>G</given-names>
            </name>
            <name name-style="western">
              <surname>Lin</surname>
              <given-names>Y</given-names>
            </name>
          </person-group>
          <article-title>Comparing human-smartphone interactions and actigraphy measurements for circadian rhythm stability and adiposity: algorithm development and validation study</article-title>
          <source>J Med Internet Res</source>
          <year>2024</year>
          <volume>26</volume>
          <fpage>e50149</fpage>
          <comment>
            <ext-link ext-link-type="uri" xlink:type="simple" xlink:href="https://www.jmir.org/2024//e50149/"/>
          </comment>
          <pub-id pub-id-type="doi">10.2196/50149</pub-id>
          <pub-id pub-id-type="medline">38838328</pub-id>
          <pub-id pub-id-type="pii">v26i1e50149</pub-id>
          <pub-id pub-id-type="pmcid">PMC11187513</pub-id>
        </nlm-citation>
      </ref>
      <ref id="ref26">
        <label>26</label>
        <nlm-citation citation-type="journal">
          <person-group person-group-type="author">
            <name name-style="western">
              <surname>Razjouyan</surname>
              <given-names>J</given-names>
            </name>
            <name name-style="western">
              <surname>Lee</surname>
              <given-names>H</given-names>
            </name>
            <name name-style="western">
              <surname>Parthasarathy</surname>
              <given-names>S</given-names>
            </name>
            <name name-style="western">
              <surname>Mohler</surname>
              <given-names>J</given-names>
            </name>
            <name name-style="western">
              <surname>Sharafkhaneh</surname>
              <given-names>A</given-names>
            </name>
            <name name-style="western">
              <surname>Najafi</surname>
              <given-names>B</given-names>
            </name>
          </person-group>
          <article-title>Improving sleep quality assessment using wearable sensors by including information from postural/sleep position changes and body acceleration: a comparison of chest-worn sensors, wrist actigraphy, and polysomnography</article-title>
          <source>J Clin Sleep Med</source>
          <year>2017</year>
          <volume>13</volume>
          <issue>11</issue>
          <fpage>1301</fpage>
          <lpage>1310</lpage>
          <comment>
            <ext-link ext-link-type="uri" xlink:type="simple" xlink:href="https://europepmc.org/abstract/MED/28992827"/>
          </comment>
          <pub-id pub-id-type="doi">10.5664/jcsm.6802</pub-id>
          <pub-id pub-id-type="medline">28992827</pub-id>
          <pub-id pub-id-type="pii">jc-17-00122</pub-id>
          <pub-id pub-id-type="pmcid">PMC5656479</pub-id>
        </nlm-citation>
      </ref>
      <ref id="ref27">
        <label>27</label>
        <nlm-citation citation-type="journal">
          <person-group person-group-type="author">
            <name name-style="western">
              <surname>Gabb</surname>
              <given-names>VG</given-names>
            </name>
            <name name-style="western">
              <surname>Blackman</surname>
              <given-names>J</given-names>
            </name>
            <name name-style="western">
              <surname>Morrison</surname>
              <given-names>H</given-names>
            </name>
            <name name-style="western">
              <surname>Li</surname>
              <given-names>H</given-names>
            </name>
            <name name-style="western">
              <surname>Kendrick</surname>
              <given-names>A</given-names>
            </name>
            <name name-style="western">
              <surname>Turner</surname>
              <given-names>N</given-names>
            </name>
            <name name-style="western">
              <surname>Greenwood</surname>
              <given-names>R</given-names>
            </name>
            <name name-style="western">
              <surname>Biswas</surname>
              <given-names>B</given-names>
            </name>
            <name name-style="western">
              <surname>Heslegrave</surname>
              <given-names>A</given-names>
            </name>
            <name name-style="western">
              <surname>Coulthard</surname>
              <given-names>E</given-names>
            </name>
          </person-group>
          <article-title>Longitudinal remote sleep and cognitive research in older adults with mild cognitive impairment and dementia: prospective feasibility cohort study</article-title>
          <source>JMIR Aging</source>
          <year>2025</year>
          <volume>8</volume>
          <fpage>e72824</fpage>
          <comment>
            <ext-link ext-link-type="uri" xlink:type="simple" xlink:href="https://aging.jmir.org/2025//e72824/"/>
          </comment>
          <pub-id pub-id-type="doi">10.2196/72824</pub-id>
          <pub-id pub-id-type="medline">40435500</pub-id>
          <pub-id pub-id-type="pii">v8i1e72824</pub-id>
          <pub-id pub-id-type="pmcid">PMC12159556</pub-id>
        </nlm-citation>
      </ref>
      <ref id="ref28">
        <label>28</label>
        <nlm-citation citation-type="journal">
          <person-group person-group-type="author">
            <name name-style="western">
              <surname>Cavaillès</surname>
              <given-names>C</given-names>
            </name>
            <name name-style="western">
              <surname>Wallace</surname>
              <given-names>M</given-names>
            </name>
            <name name-style="western">
              <surname>Leng</surname>
              <given-names>Y</given-names>
            </name>
            <name name-style="western">
              <surname>Stone</surname>
              <given-names>KL</given-names>
            </name>
            <name name-style="western">
              <surname>Ancoli-Israel</surname>
              <given-names>S</given-names>
            </name>
            <name name-style="western">
              <surname>Yaffe</surname>
              <given-names>K</given-names>
            </name>
          </person-group>
          <article-title>Multidimensional sleep profiles via machine learning and risk of dementia and cardiovascular disease</article-title>
          <source>Commun Med (Lond)</source>
          <year>2025</year>
          <volume>5</volume>
          <issue>1</issue>
          <fpage>306</fpage>
          <comment>
            <ext-link ext-link-type="uri" xlink:type="simple" xlink:href="https://doi.org/10.1038/s43856-025-01019-x"/>
          </comment>
          <pub-id pub-id-type="doi">10.1038/s43856-025-01019-x</pub-id>
          <pub-id pub-id-type="medline">40696147</pub-id>
          <pub-id pub-id-type="pii">10.1038/s43856-025-01019-x</pub-id>
          <pub-id pub-id-type="pmcid">PMC12283935</pub-id>
        </nlm-citation>
      </ref>
      <ref id="ref29">
        <label>29</label>
        <nlm-citation citation-type="journal">
          <person-group person-group-type="author">
            <name name-style="western">
              <surname>Wang</surname>
              <given-names>C</given-names>
            </name>
            <name name-style="western">
              <surname>Li</surname>
              <given-names>C</given-names>
            </name>
            <name name-style="western">
              <surname>Liang</surname>
              <given-names>S</given-names>
            </name>
          </person-group>
          <article-title>The lifestyle of new middle-aged and older adults in Taiwan described by wearable device: age and gender differences</article-title>
          <source>Eur J Ageing</source>
          <year>2024</year>
          <volume>21</volume>
          <issue>1</issue>
          <fpage>27</fpage>
          <pub-id pub-id-type="doi">10.1007/s10433-024-00824-y</pub-id>
          <pub-id pub-id-type="medline">39340574</pub-id>
          <pub-id pub-id-type="pii">10.1007/s10433-024-00824-y</pub-id>
          <pub-id pub-id-type="pmcid">PMC11438744</pub-id>
        </nlm-citation>
      </ref>
      <ref id="ref30">
        <label>30</label>
        <nlm-citation citation-type="journal">
          <person-group person-group-type="author">
            <name name-style="western">
              <surname>Shim</surname>
              <given-names>J</given-names>
            </name>
            <name name-style="western">
              <surname>Fleisch</surname>
              <given-names>E</given-names>
            </name>
            <name name-style="western">
              <surname>Barata</surname>
              <given-names>F</given-names>
            </name>
          </person-group>
          <article-title>Wearable-based accelerometer activity profile as digital biomarker of inflammation, biological age, and mortality using hierarchical clustering analysis in NHANES 2011-2014</article-title>
          <source>Sci Rep</source>
          <year>2023</year>
          <volume>13</volume>
          <issue>1</issue>
          <fpage>9326</fpage>
          <comment>
            <ext-link ext-link-type="uri" xlink:type="simple" xlink:href="https://doi.org/10.1038/s41598-023-36062-y"/>
          </comment>
          <pub-id pub-id-type="doi">10.1038/s41598-023-36062-y</pub-id>
          <pub-id pub-id-type="medline">37291134</pub-id>
          <pub-id pub-id-type="pii">10.1038/s41598-023-36062-y</pub-id>
          <pub-id pub-id-type="pmcid">PMC10250365</pub-id>
        </nlm-citation>
      </ref>
      <ref id="ref31">
        <label>31</label>
        <nlm-citation citation-type="journal">
          <person-group person-group-type="author">
            <name name-style="western">
              <surname>Jones</surname>
              <given-names>CD</given-names>
            </name>
            <name name-style="western">
              <surname>Wasilko</surname>
              <given-names>R</given-names>
            </name>
            <name name-style="western">
              <surname>Zhang</surname>
              <given-names>G</given-names>
            </name>
            <name name-style="western">
              <surname>Stone</surname>
              <given-names>KL</given-names>
            </name>
            <name name-style="western">
              <surname>Gujral</surname>
              <given-names>S</given-names>
            </name>
            <name name-style="western">
              <surname>Rodakowski</surname>
              <given-names>J</given-names>
            </name>
            <name name-style="western">
              <surname>Smagula</surname>
              <given-names>SF</given-names>
            </name>
          </person-group>
          <article-title>Detecting sleep/wake rhythm disruption related to cognition in older adults with and without mild cognitive impairment using the myRhythmWatch platform: feasibility and correlation study</article-title>
          <source>JMIR Aging</source>
          <year>2025</year>
          <volume>8</volume>
          <fpage>e67294</fpage>
          <comment>
            <ext-link ext-link-type="uri" xlink:type="simple" xlink:href="https://aging.jmir.org/2025//e67294/"/>
          </comment>
          <pub-id pub-id-type="doi">10.2196/67294</pub-id>
          <pub-id pub-id-type="medline">40194310</pub-id>
          <pub-id pub-id-type="pii">v8i1e67294</pub-id>
          <pub-id pub-id-type="pmcid">PMC11996143</pub-id>
        </nlm-citation>
      </ref>
      <ref id="ref32">
        <label>32</label>
        <nlm-citation citation-type="journal">
          <person-group person-group-type="author">
            <name name-style="western">
              <surname>Beiglary</surname>
              <given-names>S</given-names>
            </name>
            <name name-style="western">
              <surname>Feng</surname>
              <given-names>Y</given-names>
            </name>
            <name name-style="western">
              <surname>Wang</surname>
              <given-names>N</given-names>
            </name>
            <name name-style="western">
              <surname>Ghaeili</surname>
              <given-names>N</given-names>
            </name>
            <name name-style="western">
              <surname>Jao</surname>
              <given-names>Y</given-names>
            </name>
            <name name-style="western">
              <surname>Liao</surname>
              <given-names>Y</given-names>
            </name>
            <name name-style="western">
              <surname>Li</surname>
              <given-names>Y</given-names>
            </name>
            <name name-style="western">
              <surname>Wang</surname>
              <given-names>J</given-names>
            </name>
          </person-group>
          <article-title>Using wearable sensors to measure and predict personal circadian lighting exposure in nursing home residents: model development and validation</article-title>
          <source>JMIR Aging</source>
          <year>2025</year>
          <volume>8</volume>
          <fpage>e72338</fpage>
          <comment>
            <ext-link ext-link-type="uri" xlink:type="simple" xlink:href="https://aging.jmir.org/2025//e72338/"/>
          </comment>
          <pub-id pub-id-type="doi">10.2196/72338</pub-id>
          <pub-id pub-id-type="medline">41055091</pub-id>
          <pub-id pub-id-type="pii">v8i1e72338</pub-id>
          <pub-id pub-id-type="pmcid">PMC12501904</pub-id>
        </nlm-citation>
      </ref>
      <ref id="ref33">
        <label>33</label>
        <nlm-citation citation-type="journal">
          <person-group person-group-type="author">
            <name name-style="western">
              <surname>Froggatt</surname>
              <given-names>K</given-names>
            </name>
            <name name-style="western">
              <surname>Best</surname>
              <given-names>A</given-names>
            </name>
            <name name-style="western">
              <surname>Bunn</surname>
              <given-names>F</given-names>
            </name>
            <name name-style="western">
              <surname>Burnside</surname>
              <given-names>G</given-names>
            </name>
            <name name-style="western">
              <surname>Coast</surname>
              <given-names>J</given-names>
            </name>
            <name name-style="western">
              <surname>Dunleavy</surname>
              <given-names>L</given-names>
            </name>
            <name name-style="western">
              <surname>Goodman</surname>
              <given-names>C</given-names>
            </name>
            <name name-style="western">
              <surname>Hardwick</surname>
              <given-names>B</given-names>
            </name>
            <name name-style="western">
              <surname>Jackson</surname>
              <given-names>C</given-names>
            </name>
            <name name-style="western">
              <surname>Kinley</surname>
              <given-names>J</given-names>
            </name>
            <name name-style="western">
              <surname>Davidson Lund</surname>
              <given-names>A</given-names>
            </name>
            <name name-style="western">
              <surname>Lynch</surname>
              <given-names>J</given-names>
            </name>
            <name name-style="western">
              <surname>Mitchell</surname>
              <given-names>P</given-names>
            </name>
            <name name-style="western">
              <surname>Myring</surname>
              <given-names>G</given-names>
            </name>
            <name name-style="western">
              <surname>Patel</surname>
              <given-names>S</given-names>
            </name>
            <name name-style="western">
              <surname>Algorta</surname>
              <given-names>GP</given-names>
            </name>
            <name name-style="western">
              <surname>Preston</surname>
              <given-names>N</given-names>
            </name>
            <name name-style="western">
              <surname>Scott</surname>
              <given-names>D</given-names>
            </name>
            <name name-style="western">
              <surname>Silvera</surname>
              <given-names>K</given-names>
            </name>
            <name name-style="western">
              <surname>Walshe</surname>
              <given-names>C</given-names>
            </name>
          </person-group>
          <article-title>A group intervention to improve quality of life for people with advanced dementia living in care homes: the namaste feasibility cluster RCT</article-title>
          <source>Health Technol Assess</source>
          <year>2020</year>
          <volume>24</volume>
          <issue>6</issue>
          <fpage>1</fpage>
          <lpage>140</lpage>
          <comment>
            <ext-link ext-link-type="uri" xlink:type="simple" xlink:href="https://doi.org/10.3310/hta24060"/>
          </comment>
          <pub-id pub-id-type="doi">10.3310/hta24060</pub-id>
          <pub-id pub-id-type="medline">31971506</pub-id>
          <pub-id pub-id-type="pmcid">PMC7008353</pub-id>
        </nlm-citation>
      </ref>
      <ref id="ref34">
        <label>34</label>
        <nlm-citation citation-type="journal">
          <person-group person-group-type="author">
            <name name-style="western">
              <surname>Sprint</surname>
              <given-names>G</given-names>
            </name>
            <name name-style="western">
              <surname>Cook</surname>
              <given-names>DJ</given-names>
            </name>
            <name name-style="western">
              <surname>Schmitter-Edgecombe</surname>
              <given-names>M</given-names>
            </name>
            <name name-style="western">
              <surname>Holder</surname>
              <given-names>LB</given-names>
            </name>
          </person-group>
          <article-title>Multimodal fusion of smart home and text-based behavior markers for clinical assessment prediction</article-title>
          <source>ACM Trans Comput Healthc</source>
          <year>2022</year>
          <volume>3</volume>
          <issue>4</issue>
          <fpage>41</fpage>
          <comment>
            <ext-link ext-link-type="uri" xlink:type="simple" xlink:href="https://europepmc.org/abstract/MED/36381500"/>
          </comment>
          <pub-id pub-id-type="doi">10.1145/3531231</pub-id>
          <pub-id pub-id-type="medline">36381500</pub-id>
          <pub-id pub-id-type="pii">41</pub-id>
          <pub-id pub-id-type="pmcid">PMC9645787</pub-id>
        </nlm-citation>
      </ref>
      <ref id="ref35">
        <label>35</label>
        <nlm-citation citation-type="journal">
          <person-group person-group-type="author">
            <name name-style="western">
              <surname>Dhahbi</surname>
              <given-names>W</given-names>
            </name>
            <name name-style="western">
              <surname>Jebabli</surname>
              <given-names>N</given-names>
            </name>
            <name name-style="western">
              <surname>Souaifi</surname>
              <given-names>M</given-names>
            </name>
            <name name-style="western">
              <surname>Ceylan</surname>
              <given-names>Hİ</given-names>
            </name>
            <name name-style="western">
              <surname>Ben Saad</surname>
              <given-names>H</given-names>
            </name>
            <name name-style="western">
              <surname>Chamari</surname>
              <given-names>K</given-names>
            </name>
            <name name-style="western">
              <surname>Pyne</surname>
              <given-names>DB</given-names>
            </name>
            <name name-style="western">
              <surname>Chaabene</surname>
              <given-names>H</given-names>
            </name>
          </person-group>
          <article-title>Application of artificial intelligence for predicting sports injuries and customizing personalized prevention strategies: a scoping review</article-title>
          <source>Bioengineering (Basel)</source>
          <year>2026</year>
          <volume>13</volume>
          <issue>6</issue>
          <fpage>692</fpage>
          <comment>
            <ext-link ext-link-type="uri" xlink:type="simple" xlink:href="https://www.mdpi.com/resolver?pii=bioengineering13060692"/>
          </comment>
          <pub-id pub-id-type="doi">10.3390/bioengineering13060692</pub-id>
          <pub-id pub-id-type="medline">42351936</pub-id>
          <pub-id pub-id-type="pii">bioengineering13060692</pub-id>
          <pub-id pub-id-type="pmcid">PMC13296066</pub-id>
        </nlm-citation>
      </ref>
      <ref id="ref36">
        <label>36</label>
        <nlm-citation citation-type="journal">
          <person-group person-group-type="author">
            <name name-style="western">
              <surname>Lundberg</surname>
              <given-names>SM</given-names>
            </name>
            <name name-style="western">
              <surname>Lee</surname>
              <given-names>SI</given-names>
            </name>
          </person-group>
          <article-title>Unified approach to interpreting model predictions</article-title>
          <source>ArXiv. Preprint posted online on November 25, 2017</source>
          <year>2017</year>
          <pub-id pub-id-type="doi">10.48550/arXiv.1705.07874</pub-id>
        </nlm-citation>
      </ref>
      <ref id="ref37">
        <label>37</label>
        <nlm-citation citation-type="confproc">
          <person-group person-group-type="author">
            <name name-style="western">
              <surname>Ribeiro</surname>
              <given-names>MT</given-names>
            </name>
            <name name-style="western">
              <surname>Singh</surname>
              <given-names>S</given-names>
            </name>
            <name name-style="western">
              <surname>Guestrin</surname>
              <given-names>C</given-names>
            </name>
          </person-group>
          <article-title>"Why Should I Trust You?": explaining the predictions of any classifier</article-title>
          <year>2016</year>
          <conf-name>KDD '16: Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining</conf-name>
          <conf-date>August 13-17, 2016</conf-date>
          <conf-loc>San Francisco, CA, United States</conf-loc>
          <fpage>1135</fpage>
          <lpage>1144</lpage>
          <pub-id pub-id-type="doi">10.1145/2939672.2939778</pub-id>
        </nlm-citation>
      </ref>
      <ref id="ref38">
        <label>38</label>
        <nlm-citation citation-type="journal">
          <person-group person-group-type="author">
            <name name-style="western">
              <surname>Dhahbi</surname>
              <given-names>W</given-names>
            </name>
            <name name-style="western">
              <surname>Chamari</surname>
              <given-names>K</given-names>
            </name>
          </person-group>
          <article-title>The algorithmic athlete: a call to standardize assessment of sensor technologies and artificial intelligence</article-title>
          <source>Int J Sports Physiol Perform</source>
          <year>2026</year>
          <volume>21</volume>
          <issue>4</issue>
          <fpage>505</fpage>
          <lpage>506</lpage>
          <comment>
            <ext-link ext-link-type="uri" xlink:type="simple" xlink:href="https://journals.humankinetics.com/doi/10.1123/ijspp.2025-0547"/>
          </comment>
          <pub-id pub-id-type="doi">10.1123/ijspp.2025-0547</pub-id>
          <pub-id pub-id-type="medline">41493778</pub-id>
        </nlm-citation>
      </ref>
      <ref id="ref39">
        <label>39</label>
        <nlm-citation citation-type="journal">
          <person-group person-group-type="author">
            <name name-style="western">
              <surname>Lim</surname>
              <given-names>ASP</given-names>
            </name>
            <name name-style="western">
              <surname>Yu</surname>
              <given-names>L</given-names>
            </name>
            <name name-style="western">
              <surname>Costa</surname>
              <given-names>MD</given-names>
            </name>
            <name name-style="western">
              <surname>Leurgans</surname>
              <given-names>SE</given-names>
            </name>
            <name name-style="western">
              <surname>Buchman</surname>
              <given-names>AS</given-names>
            </name>
            <name name-style="western">
              <surname>Bennett</surname>
              <given-names>DA</given-names>
            </name>
            <name name-style="western">
              <surname>Saper</surname>
              <given-names>CB</given-names>
            </name>
          </person-group>
          <article-title>Increased fragmentation of rest-activity patterns is associated with a characteristic pattern of cognitive impairment in older individuals</article-title>
          <source>Sleep</source>
          <year>2012</year>
          <volume>35</volume>
          <issue>5</issue>
          <fpage>633</fpage>
          <lpage>640B</lpage>
          <comment>
            <ext-link ext-link-type="uri" xlink:type="simple" xlink:href="https://europepmc.org/abstract/MED/22547889"/>
          </comment>
          <pub-id pub-id-type="doi">10.5665/sleep.1820</pub-id>
          <pub-id pub-id-type="medline">22547889</pub-id>
          <pub-id pub-id-type="pmcid">PMC3321422</pub-id>
        </nlm-citation>
      </ref>
      <ref id="ref40">
        <label>40</label>
        <nlm-citation citation-type="journal">
          <person-group person-group-type="author">
            <name name-style="western">
              <surname>Musiek</surname>
              <given-names>ES</given-names>
            </name>
            <name name-style="western">
              <surname>Bhimasani</surname>
              <given-names>M</given-names>
            </name>
            <name name-style="western">
              <surname>Zangrilli</surname>
              <given-names>MA</given-names>
            </name>
            <name name-style="western">
              <surname>Morris</surname>
              <given-names>JC</given-names>
            </name>
            <name name-style="western">
              <surname>Holtzman</surname>
              <given-names>DM</given-names>
            </name>
            <name name-style="western">
              <surname>Ju</surname>
              <given-names>YS</given-names>
            </name>
          </person-group>
          <article-title>Circadian rest-activity pattern changes in aging and preclinical alzheimer disease</article-title>
          <source>JAMA Neurol</source>
          <year>2018</year>
          <volume>75</volume>
          <issue>5</issue>
          <fpage>582</fpage>
          <lpage>590</lpage>
          <comment>
            <ext-link ext-link-type="uri" xlink:type="simple" xlink:href="https://europepmc.org/abstract/MED/29379963"/>
          </comment>
          <pub-id pub-id-type="doi">10.1001/jamaneurol.2017.4719</pub-id>
          <pub-id pub-id-type="medline">29379963</pub-id>
          <pub-id pub-id-type="pii">2670749</pub-id>
          <pub-id pub-id-type="pmcid">PMC5885197</pub-id>
        </nlm-citation>
      </ref>
      <ref id="ref41">
        <label>41</label>
        <nlm-citation citation-type="web">
          <article-title>Artificial Intelligence/Machine Learning (AI/ML)-Based Software as a Medical Device (SaMD) Action Plan</article-title>
          <source>Food &amp; Drug Administration</source>
          <year>2021</year>
          <access-date>2026-10-27</access-date>
          <comment>
            <ext-link ext-link-type="uri" xlink:type="simple" xlink:href="https://www.fda.gov/media/145022/download">https://www.fda.gov/media/145022/download</ext-link>
          </comment>
        </nlm-citation>
      </ref>
      <ref id="ref42">
        <label>42</label>
        <nlm-citation citation-type="journal">
          <person-group person-group-type="author">
            <name name-style="western">
              <surname>Joshi</surname>
              <given-names>G</given-names>
            </name>
            <name name-style="western">
              <surname>Jain</surname>
              <given-names>A</given-names>
            </name>
            <name name-style="western">
              <surname>Araveeti</surname>
              <given-names>SR</given-names>
            </name>
            <name name-style="western">
              <surname>Adhikari</surname>
              <given-names>S</given-names>
            </name>
            <name name-style="western">
              <surname>Garg</surname>
              <given-names>H</given-names>
            </name>
            <name name-style="western">
              <surname>Bhandari</surname>
              <given-names>M</given-names>
            </name>
          </person-group>
          <article-title>FDA-approved artificial intelligence and machine learning (AI/ML)-enabled medical devices: an updated landscape</article-title>
          <source>Electronics</source>
          <year>2024</year>
          <month>01</month>
          <day>24</day>
          <volume>13</volume>
          <issue>3</issue>
          <fpage>498</fpage>
          <pub-id pub-id-type="doi">10.3390/electronics13030498</pub-id>
        </nlm-citation>
      </ref>
      <ref id="ref43">
        <label>43</label>
        <nlm-citation citation-type="journal">
          <person-group person-group-type="author">
            <name name-style="western">
              <surname>Livingston</surname>
              <given-names>G</given-names>
            </name>
            <name name-style="western">
              <surname>Huntley</surname>
              <given-names>J</given-names>
            </name>
            <name name-style="western">
              <surname>Liu</surname>
              <given-names>KY</given-names>
            </name>
            <name name-style="western">
              <surname>Costafreda</surname>
              <given-names>SG</given-names>
            </name>
            <name name-style="western">
              <surname>Selbæk</surname>
              <given-names>G</given-names>
            </name>
            <name name-style="western">
              <surname>Alladi</surname>
              <given-names>S</given-names>
            </name>
            <name name-style="western">
              <surname>Ames</surname>
              <given-names>D</given-names>
            </name>
            <name name-style="western">
              <surname>Banerjee</surname>
              <given-names>S</given-names>
            </name>
            <name name-style="western">
              <surname>Burns</surname>
              <given-names>A</given-names>
            </name>
            <name name-style="western">
              <surname>Brayne</surname>
              <given-names>C</given-names>
            </name>
            <name name-style="western">
              <surname>Fox</surname>
              <given-names>NC</given-names>
            </name>
            <name name-style="western">
              <surname>Ferri</surname>
              <given-names>CP</given-names>
            </name>
            <name name-style="western">
              <surname>Gitlin</surname>
              <given-names>LN</given-names>
            </name>
            <name name-style="western">
              <surname>Howard</surname>
              <given-names>R</given-names>
            </name>
            <name name-style="western">
              <surname>Kales</surname>
              <given-names>HC</given-names>
            </name>
            <name name-style="western">
              <surname>Kivimäki</surname>
              <given-names>M</given-names>
            </name>
            <name name-style="western">
              <surname>Larson</surname>
              <given-names>EB</given-names>
            </name>
            <name name-style="western">
              <surname>Nakasujja</surname>
              <given-names>N</given-names>
            </name>
            <name name-style="western">
              <surname>Rockwood</surname>
              <given-names>K</given-names>
            </name>
            <name name-style="western">
              <surname>Samus</surname>
              <given-names>Q</given-names>
            </name>
            <name name-style="western">
              <surname>Shirai</surname>
              <given-names>K</given-names>
            </name>
            <name name-style="western">
              <surname>Singh-Manoux</surname>
              <given-names>A</given-names>
            </name>
            <name name-style="western">
              <surname>Schneider</surname>
              <given-names>LS</given-names>
            </name>
            <name name-style="western">
              <surname>Walsh</surname>
              <given-names>S</given-names>
            </name>
            <name name-style="western">
              <surname>Yao</surname>
              <given-names>Y</given-names>
            </name>
            <name name-style="western">
              <surname>Sommerlad</surname>
              <given-names>A</given-names>
            </name>
            <name name-style="western">
              <surname>Mukadam</surname>
              <given-names>N</given-names>
            </name>
          </person-group>
          <article-title>Dementia prevention, intervention, and care: 2024 report of the Lancet standing Commission</article-title>
          <source>Lancet</source>
          <year>2024</year>
          <month>08</month>
          <volume>404</volume>
          <issue>10452</issue>
          <fpage>572</fpage>
          <lpage>628</lpage>
          <pub-id pub-id-type="doi">10.1016/s0140-6736(24)01296-0</pub-id>
        </nlm-citation>
      </ref>
      <ref id="ref44">
        <label>44</label>
        <nlm-citation citation-type="journal">
          <person-group person-group-type="author">
            <name name-style="western">
              <surname>Finlayson</surname>
              <given-names>SG</given-names>
            </name>
            <name name-style="western">
              <surname>Subbaswamy</surname>
              <given-names>A</given-names>
            </name>
            <name name-style="western">
              <surname>Singh</surname>
              <given-names>K</given-names>
            </name>
            <name name-style="western">
              <surname>Bowers</surname>
              <given-names>J</given-names>
            </name>
            <name name-style="western">
              <surname>Kupke</surname>
              <given-names>A</given-names>
            </name>
            <name name-style="western">
              <surname>Zittrain</surname>
              <given-names>J</given-names>
            </name>
            <name name-style="western">
              <surname>Kohane</surname>
              <given-names>IS</given-names>
            </name>
            <name name-style="western">
              <surname>Saria</surname>
              <given-names>S</given-names>
            </name>
          </person-group>
          <article-title>The clinician and dataset shift in artificial intelligence</article-title>
          <source>N Engl J Med</source>
          <year>2021</year>
          <month>07</month>
          <day>15</day>
          <volume>385</volume>
          <issue>3</issue>
          <fpage>283</fpage>
          <lpage>286</lpage>
          <comment>
            <ext-link ext-link-type="uri" xlink:type="simple" xlink:href="https://europepmc.org/abstract/MED/34260843"/>
          </comment>
          <pub-id pub-id-type="doi">10.1056/NEJMc2104626</pub-id>
          <pub-id pub-id-type="medline">34260843</pub-id>
          <pub-id pub-id-type="pmcid">PMC8665481</pub-id>
        </nlm-citation>
      </ref>
    </ref-list>
  </back>
</article>
