Devising a Set of Compact and Explainable Spoken Language Feature for Screening Alzheimer's Disease

TL;DR

This paper introduces a new, explainable set of spoken language features for Alzheimer's disease screening, leveraging large language models and TF-IDF, which outperform traditional features and enhance interpretability.

Contribution

The study presents a novel, interpretable feature set for AD detection based on spoken language, combining LLM and TF-IDF, with superior performance over traditional features.

Findings

New features outperform traditional linguistic features

Features are highly dimension-efficient

Enhanced interpretability of AD screening process

Abstract

Alzheimer's disease (AD) has become one of the most significant health challenges in an aging society. The use of spoken language-based AD detection methods has gained prevalence due to their scalability due to their scalability. Based on the Cookie Theft picture description task, we devised an explainable and effective feature set that leverages the visual capabilities of a large language model (LLM) and the Term Frequency-Inverse Document Frequency (TF-IDF) model. Our experimental results show that the newly proposed features consistently outperform traditional linguistic features across two different classifiers with high dimension efficiency. Our new features can be well explained and interpreted step by step which enhance the interpretability of automatic AD screening.