山本 祐

BACKGROUND: As healthcare professional trainees, resident physicians are expected to help with COVID-19 care in various ways. Many resident physicians worldwide have cared for COVID-19 patients despite the increased risk of burnout. However, few studies have examined the experience with COVID-19 care among resident physicians and its effects on competency achievement regarding clinical basics and COVID-19 patient care. METHOD: This nationwide, cross-sectional Japanese study used a clinical training environment questionnaire for resident physicians (PGY-1 and - 2) in 593 teaching hospitals during the General Medicine In-Training Examination in January 2021. The General Medicine In-Training Examination questions comprised four categories (medical interviews and professionalism; symptomatology and clinical reasoning; physical examination and clinical procedures; and disease knowledge) and a COVID-19-related question. We examined the COVID-19 care experience and its relationship with the General Medicine In-Training Examination score, adjusting for resident and hospital variables. RESULTS: Of the 6,049 resident physicians, 2,841 (47.0%) had no experience caring for patients with COVID-19 during 2020. Total and categorical General Medicine In-Training Examination scores were not different irrespective of the experience with COVID-19 patient care. For the COVID-19-related question, residents with experience in COVID-19 care showed a significant increase in correct response by 2.6% (95% confidence interval, 0.3-4.9%; p = 0.028). CONCLUSIONS: The resident physicians' COVID-19 care experience was associated with better achievement of COVID-19-related competency without reducing clinical basics. However, approximately half of the residents missed the critical experience of caring for patients during this unparalleled pandemic in Japan.

To assess the utility of wearable cameras in medical examinations, we created a physician-view video-based examination question and explanation, and the survey results indicated that these cameras can enhance the evaluation and educational capabilities of medical examinations.

BACKGROUND: The persistence of diagnostic errors, despite advances in medical knowledge and diagnostics, highlights the importance of understanding atypical disease presentations and their contribution to mortality and morbidity. Artificial intelligence (AI), particularly generative pre-trained transformers like GPT-4, holds promise for improving diagnostic accuracy, but requires further exploration in handling atypical presentations. OBJECTIVE: This study aimed to assess the diagnostic accuracy of ChatGPT in generating differential diagnoses for atypical presentations of common diseases, with a focus on the model's reliance on patient history during the diagnostic process. METHODS: We used 25 clinical vignettes from the Journal of Generalist Medicine characterizing atypical manifestations of common diseases. Two general medicine physicians categorized the cases based on atypicality. ChatGPT was then used to generate differential diagnoses based on the clinical information provided. The concordance between AI-generated and final diagnoses was measured, with a focus on the top-ranked disease (top 1) and the top 5 differential diagnoses (top 5). RESULTS: ChatGPT's diagnostic accuracy decreased with an increase in atypical presentation. For category 1 (C1) cases, the concordance rates were 17% (n=1) for the top 1 and 67% (n=4) for the top 5. Categories 3 (C3) and 4 (C4) showed a 0% concordance for top 1 and markedly lower rates for the top 5, indicating difficulties in handling highly atypical cases. The χ2 test revealed no significant difference in the top 1 differential diagnosis accuracy between less atypical (C1+C2) and more atypical (C3+C4) groups (χ²1=2.07; n=25; P=.13). However, a significant difference was found in the top 5 analyses, with less atypical cases showing higher accuracy (χ²1=4.01; n=25; P=.048). CONCLUSIONS: ChatGPT-4 demonstrates potential as an auxiliary tool for diagnosing typical and mildly atypical presentations of common diseases. However, its performance declines with greater atypicality. The study findings underscore the need for AI systems to encompass a broader range of linguistic capabilities, cultural understanding, and diverse clinical scenarios to improve diagnostic utility in real-world settings.