Search Results (1 to 10 of 1932 Results)
Download search results: CSV END BibTex RIS
Skip search results from other journals and go to results- 736 Journal of Medical Internet Research
- 239 JMIR Public Health and Surveillance
- 206 JMIR mHealth and uHealth
- 176 JMIR Medical Informatics
- 131 JMIR Formative Research
- 121 JMIR Research Protocols
- 54 JMIR Serious Games
- 34 Online Journal of Public Health Informatics
- 31 JMIR Aging
- 26 JMIR Mental Health
- 25 Interactive Journal of Medical Research
- 23 JMIR Medical Education
- 20 JMIR Cancer
- 19 JMIR Human Factors
- 17 JMIR Infodemiology
- 13 JMIR AI
- 11 JMIR Diabetes
- 10 JMIR Pediatrics and Parenting
- 7 Iproceedings
- 4 JMIR Bioinformatics and Biotechnology
- 4 JMIR Dermatology
- 4 JMIR Perioperative Medicine
- 4 JMIR Rehabilitation and Assistive Technologies
- 4 JMIRx Med
- 4 Journal of Participatory Medicine
- 3 Asian/Pacific Island Nursing Journal
- 2 JMIR Cardio
- 2 JMIR Nursing
- 1 JMIR Biomedical Engineering
- 1 JMIR XR and Spatial Computing (JMXR)
- 0 Medicine 2.0
- 0 iProceedings
- 0 JMIR Preprints
- 0 JMIR Challenges
- 0 JMIR Data
- 0 JMIRx Bio
- 0 Transfer Hub (manuscript eXchange)
- 0 JMIR Neurotechnology

In contrast, other studies, such as Li et al [19] and Wang et al [20], observed considerably lower performance, with AUC values of 0.72 and 0.73, respectively. These discrepancies can be attributed to factors such as data quality, sample size, and model architecture. Low-quality datasets, such as retrospective studies or single-center studies, may introduce selection bias and limit the generalizability of models, thereby affecting the reliability of radiomics approaches in clinical practice [21].
J Med Internet Res 2025;27:e71091
Download Citation: END BibTex RIS
Go back to the top of the page Skip and go to footer section
Go back to the top of the page Skip and go to footer section
Go back to the top of the page Skip and go to footer section

In the 2 studies that analyzed data at the nonpatient level, Li et al [12] in 2021 integrated 14 radiomic features and compared the RF model with a conventional LR model, achieving a sensitivity of 89.02%, a specificity of 64.91%, and an AUROC of 0.82 at the plaque level in the training set, with no statistically significant difference observed between the training set and the validation set (P=.58).
J Med Internet Res 2025;27:e68872
Download Citation: END BibTex RIS
Go back to the top of the page Skip and go to footer section
Go back to the top of the page Skip and go to footer section

Li et al [75] conducted electronic database searches for studies published until April 14, 2021, and included 28 studies (all were RCTs). Wu et al [67] conducted electronic database searches for studies published from 2013 to 2023 and included 20 studies (RCTs=4; quasi-experiments=3; prospective observational cohort studies=4; cross-sectional design=1; qualitative design=3; and mixed methods design=5).
JMIR Mhealth Uhealth 2025;13:e71196
Download Citation: END BibTex RIS
Go back to the top of the page Skip and go to footer section
Go back to the top of the page Skip and go to footer section