e.g. mhealth
Search Results (1 to 10 of 155 Results)
Download search results: CSV END BibTex RIS
CSV download: Download all 155 search results
(up to 5,000 articles maximum)
- 48 Journal of Medical Internet Research
- 19 JMIR Formative Research
- 14 JMIR Medical Informatics
- 14 JMIR mHealth and uHealth
- 8 JMIR Research Protocols
- 7 JMIR Public Health and Surveillance
- 6 JMIR Cardio
- 6 JMIR Medical Education
- 5 JMIR AI
- 4 Interactive Journal of Medical Research
- 4 JMIR Mental Health
- 3 JMIR Biomedical Engineering
- 3 JMIR Human Factors
- 3 JMIR Infodemiology
- 3 JMIR Pediatrics and Parenting
- 2 JMIR Cancer
- 2 JMIR Perioperative Medicine
- 1 Iproceedings
- 1 JMIR Aging
- 1 JMIR Dermatology
- 1 JMIR Rehabilitation and Assistive Technologies
- 0 Medicine 2.0
- 0 iProceedings
- 0 JMIR Serious Games
- 0 JMIR Preprints
- 0 JMIR Bioinformatics and Biotechnology
- 0 JMIR Challenges
- 0 JMIR Diabetes
- 0 JMIR Data
- 0 Journal of Participatory Medicine
- 0 JMIR Nursing
- 0 JMIRx Med
- 0 JMIRx Bio
- 0 Transfer Hub (manuscript eXchange)
- 0 JMIR Neurotechnology
- 0 Asian/Pacific Island Nursing Journal
- 0 Online Journal of Public Health Informatics
- 0 JMIR XR and Spatial Computing (JMXR)
The questionnaire, adapted from previous studies, assessed ease-of-use, convenience, reporting burden, perceived accuracy, likelihood of future use, and overall experience [36,37]. It included a question about parental assistance in home-cooked meals and app use, with responses recorded on a 5-point Likert scale from 1 (strongly disagree) to 5 (strongly agree).
JMIR Res Protoc 2025;14:e70194
Download Citation: END BibTex RIS
The results are organized by the primary performance dimensions assessed: accuracy across different question types and exam sections, repeatability between prompting strategies, confidence calibration, and robustness under adversarial conditions.
The accuracy rates of the 4 LLMs were evaluated through 2 independent attempts, with the results presented in Table 1. Overall, Gemini 2.0 Pro and Deep Seek V3 demonstrated significant advantages, with overall accuracy rates exceeding 83% in both attempts.
JMIR Med Inform 2025;13:e78279
Download Citation: END BibTex RIS
True positive rate, true negative rate, as well as accuracy rates above and below optimal discrimination threshold were estimated with 95% CIs and compared using Mc Nemar tests.
J Med Internet Res 2025;27:e64348
Download Citation: END BibTex RIS
Diagnostic accuracy was also evaluated by comparing the ground truth with diagnostic impressions provided by the model. Top-1 and top-2 diagnostic accuracy were calculated for both the Korean and English versions of the psychiatric notes.
To further examine whether translation errors affected diagnostic performance, we divided the 200 translated notes into two groups based on translation quality.
J Med Internet Res 2025;27:e69857
Download Citation: END BibTex RIS
Automatic Image Recognition Meal Reporting Among Young Adults: Randomized Controlled Trial
While the existing design was shown to be positive in terms of accuracy and was generally well received by users, there were concerns regarding the accuracy and time-consuming nature of completing meal reporting for an entire meal. Furthermore, in authentic dietary intake scenarios, voice reporting during meal consumption was not always convenient. Consequently, we developed the latest version to enhance the existing design.
JMIR Mhealth Uhealth 2025;13:e60070
Download Citation: END BibTex RIS
Deep Learning Multi-Modal Melanoma Detection: Algorithm Development and Validation
The best-performing model was a combination of Res Net-50 and Inception V3, with an accuracy of 80%. Most of these approaches aim to optimize models through transfer learning and various preprocessing techniques in an attempt to increase accuracy.
JMIR AI 2025;4:e66561
Download Citation: END BibTex RIS
Go back to the top of the page Skip and go to footer section
XGB achieved the highest overall performance, with an accuracy of 84.7% and an AUC-ROC score of 84.6%. Its F1-score of 84.0% and precision of 83.9% demonstrate its ability to consistently deliver high-accuracy predictions while minimizing false positives.
The SVM achieved an accuracy of 73.0%, comparable to that of LR, but it demonstrated an improvement in the AUC-ROC score of 65.7%. Its F1-score of 67.1% reflects a slight enhancement in predictive balance.
JMIR Med Inform 2025;13:e60204
Download Citation: END BibTex RIS
Artificial intelligence (AI) presents a solution by automating and streamlining these processes, potentially augmenting both efficiency and accuracy. However, the adoption of AI in breast cancer screening is not without challenges. Although there are over 20 Food and Drug Administration (FDA)–approved AI applications for breast imaging, their adoption and utilization in clinical settings remain highly variable and generally low [6].
J Med Internet Res 2025;27:e62941
Download Citation: END BibTex RIS
The average accuracy for v3.5 was 68.4%, while v4.0 achieved an outstanding 87.2% accuracy rate. This indicates an absolute difference of 18.8% between the two versions and a relative improvement of 27.4% in accuracy in the latest platform version.
JMIR AI 2025;4:e66552
Download Citation: END BibTex RIS