Assessing the performance of ChatGPT in answering questions regarding cirrhosis and hepatocellular carcinoma

作者：Yee Hui Yeo, Jamil S. Samaan, Wee Han Ng, Peng–Sheng Ting, Hirsh D. Trivedi, Aarshi Vipani, Walid S. Ayoub, Ju Dong Yang, Omer Liran, Brennan Spiegel, Alexander Kuo · 发表于：Clinical and Molecular Hepatology · 年份：2023 · DOI：10.3350/cmh.2023.0089 · 被引用次数：653 · 研究领域：Artificial Intelligence in Healthcare and Education、Clinical Reasoning and Diagnostic Skills、Digital Mental Health Interventions

BACKGROUND/AIMS: Patients with cirrhosis and hepatocellular carcinoma (HCC) require extensive and personalized care to improve outcomes. ChatGPT (Generative Pre-trained Transformer), a large language model, holds the potential to provide professional yet patient-friendly support. We aimed to examine the accuracy and reproducibility of ChatGPT in answering questions regarding knowledge, management, and emotional support for cirrhosis and HCC. METHODS: ChatGPT's responses to 164 questions were independently graded by two transplant hepatologists and resolved by a third reviewer. The performance of ChatGPT was also assessed using two published questionnaires and 26 questions formulated from the quality measures of cirrhosis management. Finally, its emotional support capacity was tested. RESULTS: We showed that ChatGPT regurgitated extensive knowledge of cirrhosis (79.1% correct) and HCC (74.0% correct), but only small proportions (47.3% in cirrhosis, 41.1% in HCC) were labeled as comprehensive. The performance was better in basic knowledge, lifestyle, and treatment than in the domains of diagnosis and preventive medicine. For the quality measures, the model answered 76.9% of questions correctly but failed to specify decision-making cut-offs and treatment durations. ChatGPT lacked knowledge of regional guidelines variations, such as HCC screening criteria. However, it provided practical and multifaceted advice to patients and caregivers regarding the next steps and adjusting to a...