SemioLLM: Evaluating Large Language Models for Diagnostic Reasoning from Unstructured Clinical Narratives in Epilepsy

1. 1Task LLMs with unstructured clinical narratives to evaluate diagnostic reasoning
2. 2Compare performance of medical and non-medical LLMs on epilepsy diagnosis
3. 3Analyze the ability of LLMs to encode clinical knowledge from unstructured text
4. 4Investigate the challenges of interpreting and reasoning about unstructured clinical narratives

AI engineers and researchers in the healthcare industry can benefit from this study to improve diagnostic accuracy, while data scientists can apply the findings to develop more effective natural language processing models

💡 Large Language Models can encode clinical knowledge and perform diagnostic reasoning from unstructured clinical narratives

💡 LLMs can diagnose epilepsy from unstructured clinical narratives!