Chain-of-Thought Reasoning with Large Language Models for Clinical Alzheimer's Disease Assessment and Diagnosis

TL;DR

This paper introduces a novel approach using chain-of-thought reasoning with large language models to improve Alzheimer's disease diagnosis from electronic health records, enhancing interpretability and diagnostic accuracy.

Contribution

It presents a new CoT reasoning framework that leverages LLMs for AD assessment, providing explicit diagnostic rationale and improved performance over baseline methods.

Findings

Up to 15% improvement in F1 score over zero-shot baseline

Enhanced interpretability of AD diagnosis process

Improved stability across multiple CDR grading tasks

Abstract

Alzheimer's disease (AD) has become a prevalent neurodegenerative disease worldwide. Traditional diagnosis still relies heavily on medical imaging and clinical assessment by physicians, which is often time-consuming and resource-intensive in terms of both human expertise and healthcare resources. In recent years, large language models (LLMs) have been increasingly applied to the medical field using electronic health records (EHRs), yet their application in Alzheimer's disease assessment remains limited, particularly given that AD involves complex multifactorial etiologies that are difficult to observe directly through imaging modalities. In this work, we propose leveraging LLMs to perform Chain-of-Thought (CoT) reasoning on patients' clinical EHRs. Unlike direct fine-tuning of LLMs on EHR data for AD classification, our approach utilizes LLM-generated CoT reasoning paths to provide the model with explicit diagnostic rationale for AD assessment, followed by structured CoT-based predictions. This pipeline not only enhances the model's ability to diagnose intrinsically complex factors but also improves the interpretability of the prediction process across different stages of AD progression. Experimental results demonstrate that the proposed CoT-based diagnostic framework significantly enhances stability and diagnostic performance across multiple CDR grading tasks, achieving up to a 15% improvement in F1 score compared to the zero-shot baseline method.