Causality extraction from medical text using Large Language Models (LLMs)

TL;DR

This paper investigates the effectiveness of various language models, including LLMs, in extracting causal relations from medical texts, specifically Clinical Practice Guidelines for gestational diabetes, and compares their performance.

Contribution

It is the first to evaluate LLMs and BERT variants for causality extraction in medical guidelines, providing a new annotated corpus and code for future research.

Findings

BioBERT outperformed LLMs with an F1-score of 0.72

GPT-4 and LLAMA2 showed similar but less consistent results

The study releases an annotated corpus and code for causality extraction in medical texts

Abstract

This study explores the potential of natural language models, including large language models, to extract causal relations from medical texts, specifically from Clinical Practice Guidelines (CPGs). The outcomes causality extraction from Clinical Practice Guidelines for gestational diabetes are presented, marking a first in the field. We report on a set of experiments using variants of BERT (BioBERT, DistilBERT, and BERT) and using Large Language Models (LLMs), namely GPT-4 and LLAMA2. Our experiments show that BioBERT performed better than other models, including the Large Language Models, with an average F1-score of 0.72. GPT-4 and LLAMA2 results show similar performance but less consistency. We also release the code and an annotated a corpus of causal statements within the Clinical Practice Guidelines for gestational diabetes.