3MDBench: Medical Multimodal Multi-agent Dialogue Benchmark

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Sviridov, Ivan, Miftakhova, Amina, Tereshchenko, Artemiy, Zubkova, Galina, Blinov, Pavel, Savchenko, Andrey
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866918195001032704
author Sviridov, Ivan
Miftakhova, Amina
Tereshchenko, Artemiy
Zubkova, Galina
Blinov, Pavel
Savchenko, Andrey
author_facet Sviridov, Ivan
Miftakhova, Amina
Tereshchenko, Artemiy
Zubkova, Galina
Blinov, Pavel
Savchenko, Andrey
contents Though Large Vision-Language Models (LVLMs) are being actively explored in medicine, their ability to conduct complex real-world telemedicine consultations combining accurate diagnosis with professional dialogue remains underexplored. This paper presents 3MDBench (Medical Multimodal Multi-agent Dialogue Benchmark), an open-source framework for simulating and evaluating LVLM-driven telemedical consultations. 3MDBench simulates patient variability through temperament-based Patient Agent and evaluates diagnostic accuracy and dialogue quality via Assessor Agent. It includes 2996 cases across 34 diagnoses from real-world telemedicine interactions, combining textual and image-based data. The experimental study compares diagnostic strategies for widely used open and closed-source LVLMs. We demonstrate that multimodal dialogue with internal reasoning improves F1 score by 6.5% over non-dialogue settings, highlighting the importance of context-aware, information-seeking questioning. Moreover, injecting predictions from a diagnostic convolutional neural network into the LVLM's context boosts F1 by up to 20%. Source code is available at https://github.com/univanxx/3mdbench.
format Preprint
id arxiv_https___arxiv_org_abs_2504_13861
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle 3MDBench: Medical Multimodal Multi-agent Dialogue Benchmark
Sviridov, Ivan
Miftakhova, Amina
Tereshchenko, Artemiy
Zubkova, Galina
Blinov, Pavel
Savchenko, Andrey
Human-Computer Interaction
Computation and Language
Multiagent Systems
68T42
I.2.1
Though Large Vision-Language Models (LVLMs) are being actively explored in medicine, their ability to conduct complex real-world telemedicine consultations combining accurate diagnosis with professional dialogue remains underexplored. This paper presents 3MDBench (Medical Multimodal Multi-agent Dialogue Benchmark), an open-source framework for simulating and evaluating LVLM-driven telemedical consultations. 3MDBench simulates patient variability through temperament-based Patient Agent and evaluates diagnostic accuracy and dialogue quality via Assessor Agent. It includes 2996 cases across 34 diagnoses from real-world telemedicine interactions, combining textual and image-based data. The experimental study compares diagnostic strategies for widely used open and closed-source LVLMs. We demonstrate that multimodal dialogue with internal reasoning improves F1 score by 6.5% over non-dialogue settings, highlighting the importance of context-aware, information-seeking questioning. Moreover, injecting predictions from a diagnostic convolutional neural network into the LVLM's context boosts F1 by up to 20%. Source code is available at https://github.com/univanxx/3mdbench.
title 3MDBench: Medical Multimodal Multi-agent Dialogue Benchmark
topic Human-Computer Interaction
Computation and Language
Multiagent Systems
68T42
I.2.1
url https://arxiv.org/abs/2504.13861