Background: As advancements in artificial intelligence (AI) continue, large language models (LLMs) have emerged as promising tools for generating medical information. Their rapid adaptation and potential benefits in health care require rigorous assessment in terms of the quality, accuracy, and safety of the generated information across diverse medical specialties. Objective: This study aimed to evaluate the performance of 4 prominent LLMs, namely, Claude-instant-v1.0, GPT-3.5-Turbo, Command-xlarge-nightly, and Bloomz, in generating medical content spanning the clinical specialties of ophthalmology, orthopedics, and dermatology. Methods: Three domain-specific physicians evaluated the AI-generated therapeutic recommendations for a diverse set of 60 diseases. The evaluation criteria involved the mDISCERN score, correctness, and potential harmfulness of the recommendations. ANOVA and pairwise t tests were used to explore discrepancies in content quality and safety across models and specialties. Additionally, using the capabilities of OpenAI’s most advanced model, GPT-4, an automated evaluation of each model’s responses to the diseases was performed using the same criteria and compared to the physicians’ assessments through Pearson correlation analysis. Results: Claude-instant-v1.0 emerged with the highest mean mDISCERN score (3.35, 95% CI 3.23-3.46). In contrast, Bloomz lagged with the lowest score (1.07, 95% CI 1.03-1.10). Our analysis revealed significant differences among the models in terms of quality (P<.001). Evaluating their reliability, the models displayed strong contrasts in their falseness ratings, with variations both across models (P<.001) and specialties (P<.001). Distinct error patterns emerged, such as confusing diagnoses; providing vague, ambiguous advice; or omitting critical treatments, such as antibiotics for infectious diseases. Regarding potential harm, GPT-3.5-Turbo was found to be the safest, with the lowest harmfulness rating. All models lagged in detailing the risks associated with treatment procedures, explaining the effects of therapies on quality of life, and offering additional sources of information. Pearson correlation analysis underscored a substantial alignment between physician assessments and GPT-4’s evaluations across all established criteria (P<.01). Conclusions: This study, while comprehensive, was limited by the involvement of a select number of specialties and physician evaluators. The straightforward prompting strategy (“How to treat…”) and the assessment benchmarks, initially conceptualized for human-authored content, might have potential gaps in capturing the nuances of AI-driven information. The LLMs evaluated showed a notable capability in generating valuable medical content; however, evident lapses in content quality and potential harm signal the need for further refinements. Given the dynamic landscape of LLMs, this study’s findings emphasize the need for regular and methodical assessments, oversight, and fine-tuning of these AI tools to ensure they produce consistently trustworthy and clinically safe medical advice. Notably, the introduction of an auto-evaluation mechanism using GPT-4, as detailed in this study, provides a scalable, transferable method for domain-agnostic evaluations, extending beyond therapy recommendation assessments.

Background: Large language models (LLMs) have demonstrated significant potential in diverse domains, including medicine. Nonetheless, there is a scarcity of studies examining their performance in medical examinations, especially those conducted in languages other than English, and in direct comparison with medical students. Analyzing the performance of LLMs in state medical examinations can provide insights into their capabilities and limitations and evaluate their potential role in medical education and examination preparation. Objective: This study aimed to assess and compare the performance of 3 LLMs, GPT-4, Bing, and GPT-3.5-Turbo, in the German Medical State Examinations of 2022 and to evaluate their performance relative to that of medical students. Methods: The LLMs were assessed on a total of 630 questions from the spring and fall German Medical State Examinations of 2022. The performance was evaluated with and without media-related questions. Statistical analyses included 1-way ANOVA and independent samples t tests for pairwise comparisons. The relative strength of the LLMs in comparison with that of the students was also evaluated. Results: GPT-4 achieved the highest overall performance, correctly answering 88.1% of questions, closely followed by Bing (86.0%) and GPT-3.5-Turbo (65.7%). The students had an average correct answer rate of 74.6%. Both GPT-4 and Bing significantly outperformed the students in both examinations. When media questions were excluded, Bing achieved the highest performance of 90.7%, closely followed by GPT-4 (90.4%), while GPT-3.5-Turbo lagged (68.2%). There was a significant decline in the performance of GPT-4 and Bing in the fall 2022 examination, which was attributed to a higher proportion of media-related questions and a potential increase in question difficulty. Conclusions: LLMs, particularly GPT-4 and Bing, demonstrate potential as valuable tools in medical education and for pretesting examination questions. Their high performance, even relative to that of medical students, indicates promising avenues for further development and integration into the educational and clinical landscape.

Background: The integration of artificial intelligence (AI) in medicine, particularly through AI-based language models like ChatGPT, offers a promising avenue for enhancing patient education and healthcare delivery. This study aims to evaluate the quality of medical information provided by Chat Generative Pre-trained Transformer (ChatGPT) regarding common orthopedic and trauma surgical procedures, assess its limitations, and explore its potential as a supplementary source for patient education. Methods: Using the GPT-3.5-Turbo version of ChatGPT, simulated patient information was generated for 20 orthopedic and trauma surgical procedures. The study utilized standardized information forms as a reference for evaluating ChatGPT's responses. The accuracy and quality of the provided information were assessed using a modified DISCERN instrument, and a global medical assessment was conducted to categorize the information's usefulness and reliability. Results: ChatGPT mentioned an average of 47% of relevant keywords across procedures, with a variance in the mention rate between 30.5% and 68.6%. The average modified DISCERN (mDISCERN) score was 2.4 out of 5, indicating a moderate to low quality of information. None of the ChatGPT-generated fact sheets were rated as "very useful," with 45% deemed "somewhat useful," 35% "not useful," and 20% classified as "dangerous." A positive correlation was found between higher mDISCERN scores and better physician ratings, suggesting that information quality directly impacts perceived utility. Conclusion: While AI-based language models like ChatGPT hold significant promise for medical education and patient care, the current quality of information provided in the field of orthopedics and trauma surgery is suboptimal. Further development and refinement of AI sources and algorithms are necessary to improve the accuracy and reliability of medical information. This study underscores the need for ongoing research and development in AI applications in healthcare, emphasizing the critical role of accurate, high-quality information in patient education and informed consent processes.

Background: The rapid development of large language models (LLMs) such as OpenAI’s ChatGPT has significantly impacted medical research and education. These models have shown potential in fields ranging from radiological imaging interpretation to medical licensing examination assistance. Recently, LLMs have been enhanced with image recognition capabilities. Objective: This study aims to critically examine the effectiveness of these LLMs in medical diagnostics and training by assessing their accuracy and utility in answering image-based questions from medical licensing examinations. Methods: This study analyzed 1070 image-based multiple-choice questions from the AMBOSS learning platform, divided into 605 in English and 465 in German. Customized prompts in both languages directed the models to interpret medical images and provide the most likely diagnosis. Student performance data were obtained from AMBOSS, including metrics such as the “student passed mean” and “majority vote.” Statistical analysis was conducted using Python (Python Software Foundation), with key libraries for data manipulation and visualization. Results: GPT-4 1106 Vision Preview (OpenAI) outperformed Bard Gemini Pro (Google), correctly answering 56.9% (609/1070) of questions compared to Bard’s 44.6% (477/1070), a statistically significant difference (χ²₁=32.1, P<.001). However, GPT-4 1106 left 16.1% (172/1070) of questions unanswered, significantly higher than Bard’s 4.1% (44/1070; χ²₁=83.1, P<.001). When considering only answered questions, GPT-4 1106’s accuracy increased to 67.8% (609/898), surpassing both Bard (477/1026, 46.5%; χ²₁=87.7, P<.001) and the student passed mean of (674/1070, SE 1.48%; χ²₁=4.8, P=.03). Language-specific analysis revealed both models performed better in German than English, with GPT-4 1106 showing greater accuracy in German (282/465, 60.65% vs 327/605, 54.1%; χ²₁=4.4, P=.04) and Bard Gemini Pro exhibiting a similar trend (255/465, 54.8% vs 222/605, 36.7%; χ²₁=34.3, P<.001). The student majority vote achieved an overall accuracy of 94.5% (1011/1070), significantly outperforming both artificial intelligence models (GPT-4 1106: χ²₁=408.5, P<.001; Bard Gemini Pro: χ²₁=626.6, P<.001). Conclusions: Our study shows that GPT-4 1106 Vision Preview and Bard Gemini Pro have potential in medical visual question-answering tasks and to serve as a support for students. However, their performance varies depending on the language used, with a preference for German. They also have limitations in responding to non-English content. The accuracy rates, particularly when compared to student responses, highlight the potential of these models in medical education, yet the need for further optimization and understanding of their limitations in diverse linguistic contexts remains critical.

Background: As the digital health landscape continues to evolve, analyzing the progress and direction of the field can yield valuable insights. The Journal of Medical Internet Research (JMIR) has been at the forefront of disseminating digital health research since 1999. A comprehensive network analysis of JMIR publications can help illuminate the evolution and trends in digital medicine over the past 25 years. Objective: This study aims to conduct a detailed network analysis of JMIR’s publications to uncover the growth patterns, dominant themes, and potential future trajectories in digital health research. Methods: We retrieved 8068 JMIR papers from PubMed using the Biopython library. Keyword metrics were assessed using accuracy, recall, and F1-scores to evaluate the effectiveness of keyword identification from Claude 3 Opus and Gemini 1.5 Pro in addition to 2 conventional natural language processing methods using key bidirectional encoder representations from transformers. Future trends for 2024-2026 were predicted using Claude 3 Opus, Google’s Time Series Foundation Model, autoregressive integrated moving average, exponential smoothing, and Prophet. Network visualization techniques were used to represent and analyze the complex relationships between collaborating countries, paper types, and keyword co-occurrence. Results: JMIR’s publication volume showed consistent growth, with a peak in 2020. The United States dominated country contributions, with China showing a notable increase in recent years. Keyword analysis from 1999 to 2023 showed significant thematic shifts, from an early internet and digital health focus to the dominance of COVID-19 and advanced technologies such as machine learning. Predictions for 2024-2026 suggest an increased focus on artificial intelligence, digital health, and mental health. Conclusions: Network analysis of JMIR publications provides a macroscopic view of the evolution of the digital health field. The journal’s trajectory reflects broader technological advances and shifting research priorities, including the impact of the COVID-19 pandemic. The predicted trends underscore the growing importance of computational technology in future health care research and practice. The findings from JMIR provide a glimpse into the future of digital medicine, suggesting a robust integration of artificial intelligence and continued emphasis on mental health in the postpandemic era.

Background: Large language models (LLMs) like GPT-3.5-Turbo and GPT-4 show potential to transform medical diagnostics through their linguistic and analytical capabilities. This study evaluates their diagnostic proficiency using English and German medical examination datasets. Methods: We analyzed 452 English and 637 German medical examination questions using GPT models. Performance metrics included broad and exact accuracy rates for primary and three-model generated guesses, with an analysis of performance against varying question difficulties based on student accuracy rates. Results: GPT-4 demonstrated superior performance, achieving up to 95.4% accuracy when considering approximate similarity in English datasets. While GPT-3.5-Turbo showed better results in English, GPT-4 maintained consistent performance across both languages. Question difficulty was correlated with diagnostic accuracy, particularly in German datasets. Conclusions: The study demonstrates GPT-4’s significant diagnostic capabilities and cross-linguistic flexibility, suggesting potential for clinical applications. However, further validation and ethical consideration are necessary before widespread implementation.