From Knowledge Generation to Knowledge Verification: Examining the BioMedical Generative Capabilities of ChatGPT

TL;DR

This paper presents a computational method to evaluate the factual accuracy of biomedical knowledge generated by ChatGPT, combining disease association generation with ontology-based verification to assess reliability.

Contribution

It introduces a novel approach that integrates prompt-engineering and semantic verification to assess the factual correctness of biomedical information from LLMs.

Findings

High accuracy in disease, drug, and gene identification (88%-98%)

Lower accuracy in symptom identification (49%-61%)

Verification shows 89%-91% literature coverage for disease-drug and disease-gene pairs

Abstract

The generative capabilities of LLM models offer opportunities for accelerating tasks but raise concerns about the authenticity of the knowledge they produce. To address these concerns, we present a computational approach that evaluates the factual accuracy of biomedical knowledge generated by an LLM. Our approach consists of two processes: generating disease-centric associations and verifying these associations using the semantic framework of biomedical ontologies. Using ChatGPT as the selected LLM, we designed prompt-engineering processes to establish linkages between diseases and related drugs, symptoms, and genes, and assessed consistency across multiple ChatGPT models (e.g., GPT-turbo, GPT-4, etc.). Experimental results demonstrate high accuracy in identifying disease terms (88%-97%), drug names (90%-91%), and genetic information (88%-98%). However, symptom term identification was notably lower (49%-61%), due to the informal and verbose nature of symptom descriptions, which hindered effective semantic matching with the formal language of specialized ontologies. Verification of associations reveals literature coverage rates of 89%-91% for disease-drug and disease-gene pairs, while symptom-related associations exhibit lower coverage (49%-62%).