Large language models for patient education prior to interventional radiology procedures: a comparative study

TL;DR

This study compares how well four large language models can answer patient questions about specific interventional radiology procedures, finding that some models perform well enough to potentially aid patient education.

Contribution

The study evaluates LLMs for patient education in interventional radiology and identifies performance differences across models and procedures.

Findings

DeepSeek-V3 and ChatGPT-4o outperformed OpenBioLLM-8b and BioMistral-7b in answering questions about interventional radiology procedures.

Preparation/Planning was the only category without significant differences across all models and procedures.

LLMs like DeepSeek-V3 and ChatGPT-4o show potential to enhance patient education but cannot replace clinical consultations yet.

Abstract

This study evaluates four large language models’ (LLMs) ability to answer common patient questions preceding transarterial periarticular embolization (TAPE), computed tomography (CT)-guided high-dose-rate (HDR) brachytherapy, and bleomycin electrosclerotherapy (BEST). The goal is to evaluate their potential to enhance clinical workflows and patient comprehension, while also assessing associated risks.

Thirty-five TAPE, 34 CT-HDR brachytherapy, and 36 BEST related questions were presented to ChatGPT-4o, DeepSeek-V3, OpenBioLLM-8b, and BioMistral-7b. The LLM-generated responses were independently assessed by two board-certified radiologists. Accuracy was rated on a 5-point Likert scale. Statistics compared LLM performance across question categories for patient-education suitability.

DeepSeek-V3 attained the highest mean scores for BEST [4.49 (± 0.77)] and CT-HDR [4.24 (± 0.81)] and demonstrated comparable performance to ChatGPT-4o for TAPE-related questions (DeepSeek-V3 [4.20 (± 0.77)] vs. ChatGPT-4o [4.17 (± 0.64)]; p = 1.000). In contrast, OpenBioLLM-8b (BEST 3.51 (± 1.15), CT-HDR 3.32 (± 1.13), TAPE 3.34 (± 1.16)) and BioMistral-7b (BEST 2.92 (± 1.35), CT-HDR 3.03 (± 1.06), TAPE 3.33 (± 1.28)) performed significantly worse than DeepSeek-V3 and ChatGPT-4o across all procedures. Preparation/Planning was the only category without statistically significant differences across all three procedures.

DeepSeek-V3 and ChatGPT-4o excelled on TAPE, BEST, and CT-HDR brachytherapy questions, indicating potential to enhance patient education in interventional radiology, where complex but minimally invasive procedures often are explained in brief consultations. However, OpenBioLLM-8b and BioMistral-7b exhibited more frequent inaccuracies, suggesting that LLMs cannot replace comprehensive clinical consultations yet. Patient feedback and clinical workflow implementation should validate these findings.

The online version contains supplementary material available at 10.1186/s42155-025-00609-z.

Figures