Narrative Feature or Structured Feature? A Study of Large Language Models to Identify Cancer Patients at Risk of Heart Failure
This work addresses the critical need to improve cancer treatment outcomes and safety for patients by predicting heart failure risk, representing an incremental advance in applying LLMs to medical data.
This study tackled the problem of identifying cancer patients at risk of heart failure using electronic health records, finding that a large language model (GatorTron-3.9B) with narrative features achieved the best F1 scores, outperforming traditional support vector machines by 39%, a T-LSTM model by 7%, and BERT by 5.6%.
Cancer treatments are known to introduce cardiotoxicity, negatively impacting outcomes and survivorship. Identifying cancer patients at risk of heart failure (HF) is critical to improving cancer treatment outcomes and safety. This study examined machine learning (ML) models to identify cancer patients at risk of HF using electronic health records (EHRs), including traditional ML, Time-Aware long short-term memory (T-LSTM), and large language models (LLMs) using novel narrative features derived from the structured medical codes. We identified a cancer cohort of 12,806 patients from the University of Florida Health, diagnosed with lung, breast, and colorectal cancers, among which 1,602 individuals developed HF after cancer. The LLM, GatorTron-3.9B, achieved the best F1 scores, outperforming the traditional support vector machines by 39%, the T-LSTM deep learning model by 7%, and a widely used transformer model, BERT, by 5.6%. The analysis shows that the proposed narrative features remarkably increased feature density and improved performance.