Generative AI Models for Education in Hand and Wrist Surgery: A Narrative Review

Abstract

Generative artificial intelligence (AI) models are designed to process a wide variety of inputs and produce responses that closely resemble natural human communication. There is a growing body of research evaluating the ability of these systems to answer common patient medical questions. The current evidence on AI in patient education for hand and wrist conditions is limited.

This narrative review evaluates the literature on generative AI models for patient education in hand and wrist surgery. By characterizing the current evidence, the paper summarizes the utility of these systems for patient education and highlights gaps that can guide future research, helping optimize the use of generative AI in patient care.

Studies evaluating generative AI models for patient education in hand and wrist surgery were identified through targeted literature review and reference screening. Relevant original studies were reviewed and synthesized narratively.

Eight studies were included, all of which evaluated ChatGPT (Chat Generative Pre-trained Transformer). There was variability in question prompts and the method of evaluation of AI models. The studies suggested that ChatGPT performed well, and the conversational nature of the platform was identified as a strength. Limitations included a lack of comprehensive responses and concerns regarding readability. Clinical recommendations advised the use of ChatGPT as an adjunct, but not as a replacement for clinical counseling.

Collectively, the studies suggest ChatGPT may serve as a supplementary tool for patient education in hand and wrist surgery, while highlighting the need for cautious integration and clinician oversight.