e.g. mhealth
Search Results (1 to 10 of 284 Results)
Download search results: CSV END BibTex RIS
Skip search results from other journals and go to results- 96 Journal of Medical Internet Research
- 94 JMIR Medical Education
- 26 JMIR Formative Research
- 16 JMIR Medical Informatics
- 13 JMIR Dermatology
- 10 JMIR Mental Health
- 7 JMIR AI
- 4 JMIR Human Factors
- 3 JMIR Nursing
- 3 JMIR mHealth and uHealth
- 2 Asian/Pacific Island Nursing Journal
- 2 Interactive Journal of Medical Research
- 2 JMIR Bioinformatics and Biotechnology
- 2 JMIR Research Protocols
- 1 JMIR Aging
- 1 JMIR Cancer
- 1 JMIR Cardio
- 1 JMIR Infodemiology
- 0 Medicine 2.0
- 0 iProceedings
- 0 JMIR Public Health and Surveillance
- 0 JMIR Serious Games
- 0 JMIR Rehabilitation and Assistive Technologies
- 0 JMIR Preprints
- 0 JMIR Challenges
- 0 JMIR Diabetes
- 0 JMIR Biomedical Engineering
- 0 JMIR Data
- 0 Journal of Participatory Medicine
- 0 JMIR Pediatrics and Parenting
- 0 JMIR Perioperative Medicine
- 0 JMIRx Med
- 0 JMIRx Bio
- 0 Transfer Hub (manuscript eXchange)
- 0 JMIR Neurotechnology
- 0 Online Journal of Public Health Informatics
- 0 JMIR XR and Spatial Computing (JMXR)
Authors’ Reply: Citation Accuracy Challenges Posed by Large Language Models
We appreciate the thoughtful critique of our manuscript “Perceptions and earliest experiences of medical students and faculty with Chat GPT in medical education: qualitative study” [1] by Zhao and Zhang [2]. Concerns over the generation of hallucinated citations by large language models (LLMs), such as Open AI’s Chat GPT, Google’s Gemini, and Hangzhou’s Deep Seek, warrant exploring advanced and novel methodologies to ensure citation accuracy and overall output integrity [3].
JMIR Med Educ 2025;11:e73698
Download Citation: END BibTex RIS
Citation Accuracy Challenges Posed by Large Language Models
Large language models (LLMs) such as Deep Seek, Chat GPT, and Chat GLM have significant limitations in generating citations, raising concerns about the quality and reliability of academic research. These models tend to produce citations that are correctly formatted but fictional in content, misleading users and undermining academic rigor.
JMIR Med Educ 2025;11:e72998
Download Citation: END BibTex RIS
Juels Parker commented on our study comparing the sufficiency of Chat GPT, Google Bard, and Bing artificial intelligence (AI) in generating patient-facing responses to questions about five dermatological diagnoses [1,2]. He highlights an important need to compare AI to existing patient education tools, such as handouts, peer-reviewed articles, and patient-centered websites.
We agree that AI is not a benign entity, and many resources exist for patients to learn about their conditions, aside from AI [3,4].
JMIR Dermatol 2025;8:e72540
Download Citation: END BibTex RIS
Reference 1: The comparative sufficiency of ChatGPT, Google Bard, and Bing AI in answering diagnosis,chatgptTheme Issue (2023): AI and ChatGPT in Dermatology
JMIR Dermatol 2025;8:e71768
Download Citation: END BibTex RIS
The doctors’ responses and Chat GPT responses for the same question were anonymized and randomly labeled as Response 1 and Response 2.
J Med Internet Res 2025;27:e62857
Download Citation: END BibTex RIS
There has a been rapid popularization of the LLM, Chat GPT for home-based medical inquiries [3]. Minimal research exists on Chat GPT’s accuracy in detecting melanoma. Given that patients are increasingly presenting internet-derived diagnostics during cancer consultations, it is imperative to understand the capabilities of commonly used AI engines, such as Chat GPT [4].
JMIR Dermatol 2025;8:e67551
Download Citation: END BibTex RIS
To determine the extent of exposure to Chat GPT, for each participant in MIBot (version 5.2 A), we included an additional short survey in the 1-week-later survey referred to as the Chat GPT survey. It contained 8 new questions designed to evaluate the participant’s knowledge and use of Chat GPT prior to engaging in MIBot (version 5.2 A). The full Chat GPT survey can be found in Multimedia Appendix 1.
JMIR Form Res 2025;9:e56973
Download Citation: END BibTex RIS
We assessed the ability of Chat GPT to distinguish images of SCC and BCC from other lesions.
Open AI’s application programming interface was used to query Chat GPT-4 Omni (Chat GPT-4 O) for assessing the performance in classifying 200 dermatoscopic images each of SCC, BCC, BK, melanocytic nevi, and 150 images of AK from the HAM10 K database [7]. Images were verified using histopathology (>50%), follow-up examination, expert consensus, or in-vivo confocal microscopy.
JMIR Dermatol 2025;8:e67299
Download Citation: END BibTex RIS
While it is not the user (eg, researchers or clinicians), but the manufacturer (eg, Open AI for the Chat GPT models) who assigns an intended medical use—which itself comes with further regulatory requirements—the clinical use of the currently available and mostly all-purpose LLMs remains challenging.
JMIR Med Educ 2025;11:e58375
Download Citation: END BibTex RIS
For example, Chat GPT 3.5 performed poorly in diagnosing an infectious disease known to be widely underdiagnosed [26]. Furthermore, Chat GPT 3.5 made different treatment recommendations based on insurance status, which might introduce health disparities [27]. When generating clinical cases, Chat GPT-4 failed to create cases that depicted demographic diversity and relied on stereotypes when choosing gender or ethnicity [28].
JMIR Ment Health 2025;12:e57986
Download Citation: END BibTex RIS