Beyond Self-Consistency: Ensemble Reasoning Boosts Consistency and Accuracy of LLMs in Cancer Staging

TL;DR

This paper introduces an ensemble reasoning method for large language models that enhances the consistency and accuracy of cancer staging from clinical reports, demonstrating improved reliability in medical decision-making.

Contribution

It presents a novel ensemble reasoning approach that improves both the consistency and performance of LLMs in extracting cancer stages from unstructured clinical text.

Findings

Ensemble reasoning improves model consistency.

Enhanced accuracy in cancer staging from reports.

Potential for reliable clinical application.

Abstract

Advances in large language models (LLMs) have encouraged their adoption in the healthcare domain where vital clinical information is often contained in unstructured notes. Cancer staging status is available in clinical reports, but it requires natural language processing to extract the status from the unstructured text. With the advance in clinical-oriented LLMs, it is promising to extract such status without extensive efforts in training the algorithms. Prompting approaches of the pre-trained LLMs that elicit a model's reasoning process, such as chain-of-thought, may help to improve the trustworthiness of the generated responses. Using self-consistency further improves model performance, but often results in inconsistent generations across the multiple reasoning paths. In this study, we propose an ensemble reasoning approach with the aim of improving the consistency of the model generations. Using an open access clinical large language model to determine the pathologic cancer stage from real-world pathology reports, we show that the ensemble reasoning approach is able to improve both the consistency and performance of the LLM in determining cancer stage, thereby demonstrating the potential to use these models in clinical or other domains where reliability and trustworthiness are critical.