Language Models for Longitudinal Clinical Prediction

TL;DR

This paper presents a lightweight method that adapts large language models to analyze longitudinal clinical data, enabling accurate predictions in neuropsychological assessments without fine-tuning, especially useful for early Alzheimer's detection.

Contribution

It introduces a novel framework that leverages frozen language models for longitudinal clinical prediction, avoiding the need for extensive model fine-tuning.

Findings

Achieves high accuracy in neuropsychological assessment predictions.

Performs reliably with minimal training data.

Shows potential for early Alzheimer's disease monitoring.

Abstract

We explore a lightweight framework that adapts frozen large language models to analyze longitudinal clinical data. The approach integrates patient history and context within the language model space to generate accurate forecasts without model fine-tuning. Applied to neuropsychological assessments, it achieves accurate and reliable performance even with minimal training data, showing promise for early-stage Alzheimer's monitoring.