Aspect-Oriented Summarization for Psychiatric Short-Term Readmission Prediction
This work addresses a high-impact healthcare prediction task for psychiatric patients, but it is incremental as it builds on existing summarization and fine-tuning approaches.
The authors tackled the problem of predicting 30-day psychiatric readmissions from discharge documents by using aspect-oriented LLM summaries to capture diverse information signals, and their method improved prediction performance on real-world data from four hospitals.
Recent progress in large language models (LLMs) has enabled the automated processing of lengthy documents even without supervised training on a task-specific dataset. Yet, their zero-shot performance in complex tasks as opposed to straightforward information extraction tasks remains suboptimal. One feasible approach for tasks with lengthy, complex input is to first summarize the document and then apply supervised fine-tuning to the summary. However, the summarization process inevitably results in some loss of information. In this study we present a method for processing the summaries of long documents aimed to capture different important aspects of the original document. We hypothesize that LLM summaries generated with different aspect-oriented prompts contain different information signals, and we propose methods to measure these differences. We introduce approaches to effectively integrate signals from these different summaries for supervised training of transformer models. We validate our hypotheses on a high-impact task -- 30-day readmission prediction from a psychiatric discharge -- using real-world data from four hospitals, and show that our proposed method increases the prediction performance for the complex task of predicting patient outcome.