Search Articles

View query in Help articles search

Search Results (1 to 10 of 254 Results)

Download search results: CSV END BibTex RIS


Deep Learning and Image Generator Health Tabular Data (IGHT) for Predicting Overall Survival in Patients With Colorectal Cancer: Retrospective Study

Deep Learning and Image Generator Health Tabular Data (IGHT) for Predicting Overall Survival in Patients With Colorectal Cancer: Retrospective Study

Health care providers use AI for patient disease prediction and data anonymization, addressing the growing health care costs associated with increasing chronic diseases and life expectancy [2]. Recent studies have increasingly applied advanced machine learning and explainable artificial intelligence (XAI) methods to improve disease survival prediction.

Seo Hyun Oh, Youngho Lee, Jeong-Heum Baek, Woongsang Sunwoo

JMIR Med Inform 2025;13:e75022

AI and Machine Learning Terminology in Medicine, Psychology, and Social Sciences: Tutorial and Practical Recommendations

AI and Machine Learning Terminology in Medicine, Psychology, and Social Sciences: Tutorial and Practical Recommendations

In summary, most “prediction” studies predominantly explain the relationships between variables and outcomes within existing data, rather than establishing generalizable prediction through rigorous validation with new data. Even among studies using appropriate ML prediction models with both training and validation components, most implement only basic predictive methodologies, focusing on prediction with retrospective modeling and lacking external validation.

Bo Cao, Russell Greiner, Andrew Greenshaw, Jie Sui

J Med Internet Res 2025;27:e66100

Optimizing Feature Selection and Machine Learning Algorithms for Early Detection of Prediabetes Risk: Comparative Study

Optimizing Feature Selection and Machine Learning Algorithms for Early Detection of Prediabetes Risk: Comparative Study

Additionally, it reduces model complexity and boosts prediction accuracy by eliminating irrelevant or unnecessary data in ML. Models based on the most impactful clinical features, such as BMI, age, low-density lipoprotein cholesterol (LDL-C), and high-density lipoprotein cholesterol (HDL-C), can capture underlying patterns linked with prediabetes [6].

Mahmoud B Almadhoun, MA Burhanuddin

JMIR Bioinform Biotech 2025;6:e70621

A Responsible Framework for Assessing, Selecting, and Explaining Machine Learning Models in Cardiovascular Disease Outcomes Among People With Type 2 Diabetes: Methodology and Validation Study

A Responsible Framework for Assessing, Selecting, and Explaining Machine Learning Models in Cardiovascular Disease Outcomes Among People With Type 2 Diabetes: Methodology and Validation Study

However, efforts to quantitatively evaluate fairness in prediction models for clinical practice are still scarce [15]. A model with high predictive accuracy does not guarantee the best clinical usage, as it might display unfavorable biases [16]. As a result, it is important to understand and quantify the trade-offs between accuracy and fairness in model selection.

Yang Yang, Che-Yi Liao, Esmaeil Keyvanshokooh, Hui Shao, Mary Beth Weber, Francisco J Pasquel, Gian-Gabriel P Garcia

JMIR Med Inform 2025;13:e66200

Predicting Early-Onset Colorectal Cancer in Individuals Below Screening Age Using Machine Learning and Real-World Data: Case Control Study

Predicting Early-Onset Colorectal Cancer in Individuals Below Screening Age Using Machine Learning and Real-World Data: Case Control Study

We then incorporated a range of different observation periods and prediction windows (Figure 2) to test our prediction algorithms, considering the different use cases. We considered 4 different prediction windows: 0 years, 1 year, 3 years, and 5 years before CRC diagnosis. Visualization of the observation and prediction windows. For the prediction task. The index date for CRC cases is the date of diagnosis.

Chengkun Sun, Erin Mobley, Michael Quillen, Max Parker, Meghan Daly, Rui Wang, Isabela Visintin, Ziad Awad, Jennifer Fishe, Alexander Parker, Thomas George, Jiang Bian, Jie Xu

JMIR Cancer 2025;11:e64506

Peer Relationships Are a Direct Cause of the Adolescent Mental Health Crisis: Interpretable Machine Learning Analysis of 2 Large Cohort Studies

Peer Relationships Are a Direct Cause of the Adolescent Mental Health Crisis: Interpretable Machine Learning Analysis of 2 Large Cohort Studies

The reporting of the model development and validation followed the Transparent Reporting of a multivariable prediction model for Individual Prognosis Or Diagnosis (TRIPOD [37]) guidelines (Table S4 in Multimedia Appendix 1). The strongest statistically significant unadjusted linear relationships between baseline predictors and mental health at follow-up are shown in Figure 4 and coefficients of all predictors can be found in Tables S1 and S2 in Multimedia Appendix 1.

Heiner Stuke, Robert Schlack, Michael Erhart, Anne Kaman, Ulrike Ravens-Sieberer, Christopher Irrgang

JMIR Public Health Surveill 2025;11:e60125

Prediction of Reactivation After Antivascular Endothelial Growth Factor Monotherapy for Retinopathy of Prematurity: Multimodal Machine Learning Model Study

Prediction of Reactivation After Antivascular Endothelial Growth Factor Monotherapy for Retinopathy of Prematurity: Multimodal Machine Learning Model Study

Prediction models that can identify infants with a high risk of ROP reactivation are needed in clinical practice. Artificial intelligence has recently optimized medical practice [15-17]. Artificial intelligence has been mainly applied to ROP diagnosis and prediction based on imaging [17-19]. To our knowledge, studies on ROP reactivation after treatment are very limited, and there is no successful prediction model for clinical application.

Rong Wu, Yu Zhang, Peijie Huang, Yiying Xie, Jianxun Wang, Shuangyong Wang, Qiuxia Lin, Yichen Bai, Songfu Feng, Nian Cai, Xiaohe Lu

J Med Internet Res 2025;27:e60367

Effect of Uncertainty-Aware AI Models on Pharmacists’ Reaction Time and Decision-Making in a Web-Based Mock Medication Verification Task: Randomized Controlled Trial

Effect of Uncertainty-Aware AI Models on Pharmacists’ Reaction Time and Decision-Making in a Web-Based Mock Medication Verification Task: Randomized Controlled Trial

Uncertainty-aware AI models present the model’s uncertainty, or confidence in its decision, alongside its prediction [11], thus providing a metric for the user to assess the AI’s reliability [12]. CDSS reliability is an essential component of human evaluation of AI’s trustworthiness which determines the user’s acceptability of a technology [7].

Corey Lester, Brigid Rowell, Yifan Zheng, Zoe Co, Vincent Marshall, Jin Yong Kim, Qiyuan Chen, Raed Kontar, X Jessie Yang

JMIR Med Inform 2025;13:e64902