Enhancing Large Language Models with Human Expertise for Disease Detection in Electronic Health Records

Electronic health records (EHR) are widely available to complement administrative data-based disease surveillance and healthcare performance evaluation. Defining conditions from EHR is labour-intensive, requiring advanced medical informatics knowledge. We linked a cardiac registry cohort in 2015 with an EHR system in a city in Canada. We developed a throughput pipeline that leverages a generative large language model (LLM) to analyze, understand, and interpret EHR notes through clinical experts' designed prompts and rules. The pipeline was applied to detect diabetes, hypertension, and acute myocardial infarction from the notes. The performance was compared against clinician-validated diagnoses as the reference standard.