Saved in:
Bibliographic Details
Main Authors: Hasan, Md. Najib, Ahmad, Imran, Shuvo, Sourav Basak, Ankon, Md. Mahadi Hasan, Das, Sunanda, Siddique, Nazmul, Wang, Hui
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2512.13742
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914343071776768
author Hasan, Md. Najib
Ahmad, Imran
Shuvo, Sourav Basak
Ankon, Md. Mahadi Hasan
Das, Sunanda
Siddique, Nazmul
Wang, Hui
author_facet Hasan, Md. Najib
Ahmad, Imran
Shuvo, Sourav Basak
Ankon, Md. Mahadi Hasan
Das, Sunanda
Siddique, Nazmul
Wang, Hui
contents Medical image classifiers detect gastrointestinal diseases well, but they do not explain their decisions. Large language models can generate clinical text, yet they struggle with visual reasoning and often produce unstable or incorrect explanations. This leaves a gap between what a model sees and the type of reasoning a clinician expects. We introduce a framework that links image classification with structured clinical reasoning. A new hybrid model, MobileCoAtNet, is designed for endoscopic images and achieves high accuracy across eight stomach-related classes. Its outputs are then used to drive reasoning by several LLMs. To judge this reasoning, we build two expert-verified benchmarks covering causes, symptoms, treatment, lifestyle, and follow-up care. Thirty-two LLMs are evaluated against these gold standards. Strong classification improves the quality of their explanations, but none of the models reach human-level stability. Even the best LLMs change their reasoning when prompts vary. Our study shows that combining DL with LLMs can produce useful clinical narratives, but current LLMs remain unreliable for high-stakes medical decisions. The framework provides a clearer view of their limits and a path for building safer reasoning systems. The complete source code and datasets used in this study are available at https://github.com/souravbasakshuvo/DL3M.
format Preprint
id arxiv_https___arxiv_org_abs_2512_13742
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle DL$^3$M: A Vision-to-Language Framework for Expert-Level Medical Reasoning through Deep Learning and Large Language Models
Hasan, Md. Najib
Ahmad, Imran
Shuvo, Sourav Basak
Ankon, Md. Mahadi Hasan
Das, Sunanda
Siddique, Nazmul
Wang, Hui
Computer Vision and Pattern Recognition
Artificial Intelligence
Medical image classifiers detect gastrointestinal diseases well, but they do not explain their decisions. Large language models can generate clinical text, yet they struggle with visual reasoning and often produce unstable or incorrect explanations. This leaves a gap between what a model sees and the type of reasoning a clinician expects. We introduce a framework that links image classification with structured clinical reasoning. A new hybrid model, MobileCoAtNet, is designed for endoscopic images and achieves high accuracy across eight stomach-related classes. Its outputs are then used to drive reasoning by several LLMs. To judge this reasoning, we build two expert-verified benchmarks covering causes, symptoms, treatment, lifestyle, and follow-up care. Thirty-two LLMs are evaluated against these gold standards. Strong classification improves the quality of their explanations, but none of the models reach human-level stability. Even the best LLMs change their reasoning when prompts vary. Our study shows that combining DL with LLMs can produce useful clinical narratives, but current LLMs remain unreliable for high-stakes medical decisions. The framework provides a clearer view of their limits and a path for building safer reasoning systems. The complete source code and datasets used in this study are available at https://github.com/souravbasakshuvo/DL3M.
title DL$^3$M: A Vision-to-Language Framework for Expert-Level Medical Reasoning through Deep Learning and Large Language Models
topic Computer Vision and Pattern Recognition
Artificial Intelligence
url https://arxiv.org/abs/2512.13742