Role of ChatGPT and Google Bard in the Diagnosis of Psychiatric Disorders: A Comparative Study

作者：Yadgar N. Abbas, Yousif M. Mahmood, Hemn Ali Hassan, Dyari Q. Hamad, Sabah Jalal Hasan, Diyar A. Omer, Suhaib H. Kakamad, Hussein Mustafa Hamasalih, Marwan Nasih Hassan, Hawbash Mohammed-Amin Rahim, Rawezh Qadir M. Salih, Fahmi H. kakamad, Berun A. Abdalla, Shvan H. Mohammed · 发表于：Barw Medical Journal · 年份：2023 · DOI：10.58742/4vd6h741 · 被引用次数：9 · 研究领域：Artificial Intelligence in Healthcare and Education、Machine Learning in Healthcare、COVID-19 diagnosis using AI

Abstract Introduction The incorporation of artificial intelligence (AI) in the medical decision-making matrix has captured interest across diverse medical domains. This study aimed to juxtapose the decision-making patterns of humans and artificial intelligence regarding psychiatric disorders. Methods A set of case stories composed of 20 questions and the ideal answers were developed by a psychiatrist (the first author) based on International Classification of Diseases or Diagnostic and Statistical Manual of Mental Disorders. The cases and replies were revised by other authors, and one by one, they were presented to ChatGPT and Google Bard. The results were presented in a table. Results Both ChatGPT and Google Bard reported a high rate of precision in the spot diagnosis of the cases. ChatGPT provided a correct diagnosis for 15 cases (75%), while Google Bard diagnosed 14 cases (70%) successfully. Conclusion ChatGPT and Google Bard's success in this study opens the door for deeper AI integration in psychiatric evaluations. As technology evolves, the boundary between human and AI decision-making may become less distinct, promising a new era in psychiatric care. Moving forward, we must approach AI in healthcare with enthusiasm, collaboration, and caution.