CoD，基于诊断链实现可解释医疗代理程序

CoD, Towards an Interpretable Medical Agent using Chain of Diagnosis

July 18, 2024

作者: Junying Chen, Chi Gui, Anningzhe Gao, Ke Ji, Xidong Wang, Xiang Wan, Benyou Wang

cs.AI

摘要

随着大型语言模型（LLMs）的出现，医学诊断领域发生了重大变革，然而这些模型内部的可解释性挑战仍然未得到很好的解决。本研究引入了诊断链（CoD）来增强基于LLM的医学诊断的可解释性。CoD将诊断过程转化为一种诊断链，模拟了医生的思维过程，提供了透明的推理路径。此外，CoD输出疾病置信度分布，以确保决策过程的透明性。这种可解释性使模型诊断可控，并有助于通过减少置信度的熵来识别需要进一步调查的关键症状。借助CoD，我们开发了DiagnosisGPT，能够诊断9604种疾病。实验结果表明，DiagnosisGPT在诊断基准上优于其他LLMs。此外，DiagnosisGPT提供了可解释性，同时确保了诊断严谨性的可控性。

English

The field of medical diagnosis has undergone a significant transformation with the advent of large language models (LLMs), yet the challenges of interpretability within these models remain largely unaddressed. This study introduces Chain-of-Diagnosis (CoD) to enhance the interpretability of LLM-based medical diagnostics. CoD transforms the diagnostic process into a diagnostic chain that mirrors a physician's thought process, providing a transparent reasoning pathway. Additionally, CoD outputs the disease confidence distribution to ensure transparency in decision-making. This interpretability makes model diagnostics controllable and aids in identifying critical symptoms for inquiry through the entropy reduction of confidences. With CoD, we developed DiagnosisGPT, capable of diagnosing 9604 diseases. Experimental results demonstrate that DiagnosisGPT outperforms other LLMs on diagnostic benchmarks. Moreover, DiagnosisGPT provides interpretability while ensuring controllability in diagnostic rigor.