Evaluation of ChatGPT-Generated Differential Diagnosis for Common Diseases With Atypical Presentation: Descriptive Research.
Kiyoshi ShikinoTaro ShimizuYuki OtsukaMasaki TagoHiromizu TakahashiYasuharu TokudaYosuke SasakiGenmei IizukaHiroki TamuraKoichi NakashimaKotaro KunitomoMorika SuzukiSayaka AoyamaShintaro KosakaTeiko KawahigashiTomohiro MatsumotoFumina OriharaToru MorikawaToshinori NishizawaYoji HoshinaYu YamamotoYuichiro MatsuoYuto UnokiHirofumi KimuraMidori TokushimaSatoshi WatanukiTakuma SaitoFumio OtuskaYasuharu TokudaPublished in: JMIR medical education (2024)
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.