UMass-BioNLP at MEDIQA-M3G 2024: DermPrompt -- A Systematic Exploration of Prompt Engineering with GPT-4V for Dermatological Diagnosis

TL;DR

This paper explores prompt engineering with GPT-4V for dermatological diagnosis, demonstrating high retrieval accuracy, the effectiveness of different reasoning strategies, and the superiority of a multi-agent conversation framework for accurate diagnosis.

Contribution

It introduces a novel multi-agent conversation framework and empirically evaluates prompt strategies for GPT-4V in dermatology diagnosis, advancing multimodal diagnostic methods.

Findings

GPT-4V retrieves correct skin conditions 85% of the time

Naive CoT works well for retrieval tasks

Multi-Agent Conversation outperforms other strategies

Abstract

This paper presents our team's participation in the MEDIQA-ClinicalNLP2024 shared task B. We present a novel approach to diagnosing clinical dermatology cases by integrating large multimodal models, specifically leveraging the capabilities of GPT-4V under a retriever and a re-ranker framework. Our investigation reveals that GPT-4V, when used as a retrieval agent, can accurately retrieve the correct skin condition 85% of the time using dermatological images and brief patient histories. Additionally, we empirically show that Naive Chain-of-Thought (CoT) works well for retrieval while Medical Guidelines Grounded CoT is required for accurate dermatological diagnosis. Further, we introduce a Multi-Agent Conversation (MAC) framework and show its superior performance and potential over the best CoT strategy. The experiments suggest that using naive CoT for retrieval and multi-agent conversation for critique-based diagnosis, GPT-4V can lead to an early and accurate diagnosis of dermatological conditions. The implications of this work extend to improving diagnostic workflows, supporting dermatological education, and enhancing patient care by providing a scalable, accessible, and accurate diagnostic tool.