_version_ 1866914063892611072
author Bai, Yang
Cheng, Haoran
Zhou, Yang
Zhou, Jun
Thirunavukarasu, Arun
Ke, Yuhe
Yao, Jie
Fukutsu, Kanae
Quek, Chrystie Wan Ning
Hong, Ashley
Gutierrez, Laura
Teo, Zhen Ling
Ting, Darren Shu Jeng
Soetikno, Brian T.
Nielsen, Christopher S.
Elze, Tobias
Li, Zengxiang
Dinh, Linh Le
Chan, Hiok Hong
Koh, Victor
Tan, Marcus
Li, Kelvin Z.
Yip, Leonard
Cheng, Ching Yu
Tham, Yih Chung
Tan, Gavin Siew Wei
Schmetterer, Leopold
Ang, Marcus
Hussain, Rahat
Mehta, Jod
Aung, Tin
Cheng, Lionel Tim-Ee
Anh, Tran Nguyen Tuan
Cheng, Chee Leong
Wong, Tien Yin
Liu, Nan
Tan, Iain Beehuat
Lim, Soon Thye
Klang, Eyal
Lim, Tony Kiat Hon
Goh, Rick Siow Mong
Liu, Yong
Ting, Daniel Shu Wei
author_facet Bai, Yang
Cheng, Haoran
Zhou, Yang
Zhou, Jun
Thirunavukarasu, Arun
Ke, Yuhe
Yao, Jie
Fukutsu, Kanae
Quek, Chrystie Wan Ning
Hong, Ashley
Gutierrez, Laura
Teo, Zhen Ling
Ting, Darren Shu Jeng
Soetikno, Brian T.
Nielsen, Christopher S.
Elze, Tobias
Li, Zengxiang
Dinh, Linh Le
Chan, Hiok Hong
Koh, Victor
Tan, Marcus
Li, Kelvin Z.
Yip, Leonard
Cheng, Ching Yu
Tham, Yih Chung
Tan, Gavin Siew Wei
Schmetterer, Leopold
Ang, Marcus
Hussain, Rahat
Mehta, Jod
Aung, Tin
Cheng, Lionel Tim-Ee
Anh, Tran Nguyen Tuan
Cheng, Chee Leong
Wong, Tien Yin
Liu, Nan
Tan, Iain Beehuat
Lim, Soon Thye
Klang, Eyal
Lim, Tony Kiat Hon
Goh, Rick Siow Mong
Liu, Yong
Ting, Daniel Shu Wei
contents Despite the promise of foundation models in medical AI, current systems remain limited - they are modality-specific and lack transparent reasoning processes, hindering clinical adoption. To address this gap, we present EVLF-FM, a multimodal vision-language foundation model (VLM) designed to unify broad diagnostic capability with fine-grain explainability. The development and testing of EVLF-FM encompassed over 1.3 million total samples from 23 global datasets across eleven imaging modalities related to six clinical specialties: dermatology, hepatology, ophthalmology, pathology, pulmonology, and radiology. External validation employed 8,884 independent test samples from 10 additional datasets across five imaging modalities. Technically, EVLF-FM is developed to assist with multiple disease diagnosis and visual question answering with pixel-level visual grounding and reasoning capabilities. In internal validation for disease diagnostics, EVLF-FM achieved the highest average accuracy (0.858) and F1-score (0.797), outperforming leading generalist and specialist models. In medical visual grounding, EVLF-FM also achieved stellar performance across nine modalities with average mIOU of 0.743 and Acc@0.5 of 0.837. External validations further confirmed strong zero-shot and few-shot performance, with competitive F1-scores despite a smaller model size. Through a hybrid training strategy combining supervised and visual reinforcement fine-tuning, EVLF-FM not only achieves state-of-the-art accuracy but also exhibits step-by-step reasoning, aligning outputs with visual evidence. EVLF-FM is an early multi-disease VLM model with explainability and reasoning capabilities that could advance adoption of and trust in foundation models for real-world clinical deployment.
format Preprint
id arxiv_https___arxiv_org_abs_2509_24231
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle EVLF-FM: Explainable Vision Language Foundation Model for Medicine
Bai, Yang
Cheng, Haoran
Zhou, Yang
Zhou, Jun
Thirunavukarasu, Arun
Ke, Yuhe
Yao, Jie
Fukutsu, Kanae
Quek, Chrystie Wan Ning
Hong, Ashley
Gutierrez, Laura
Teo, Zhen Ling
Ting, Darren Shu Jeng
Soetikno, Brian T.
Nielsen, Christopher S.
Elze, Tobias
Li, Zengxiang
Dinh, Linh Le
Chan, Hiok Hong
Koh, Victor
Tan, Marcus
Li, Kelvin Z.
Yip, Leonard
Cheng, Ching Yu
Tham, Yih Chung
Tan, Gavin Siew Wei
Schmetterer, Leopold
Ang, Marcus
Hussain, Rahat
Mehta, Jod
Aung, Tin
Cheng, Lionel Tim-Ee
Anh, Tran Nguyen Tuan
Cheng, Chee Leong
Wong, Tien Yin
Liu, Nan
Tan, Iain Beehuat
Lim, Soon Thye
Klang, Eyal
Lim, Tony Kiat Hon
Goh, Rick Siow Mong
Liu, Yong
Ting, Daniel Shu Wei
Computer Vision and Pattern Recognition
Despite the promise of foundation models in medical AI, current systems remain limited - they are modality-specific and lack transparent reasoning processes, hindering clinical adoption. To address this gap, we present EVLF-FM, a multimodal vision-language foundation model (VLM) designed to unify broad diagnostic capability with fine-grain explainability. The development and testing of EVLF-FM encompassed over 1.3 million total samples from 23 global datasets across eleven imaging modalities related to six clinical specialties: dermatology, hepatology, ophthalmology, pathology, pulmonology, and radiology. External validation employed 8,884 independent test samples from 10 additional datasets across five imaging modalities. Technically, EVLF-FM is developed to assist with multiple disease diagnosis and visual question answering with pixel-level visual grounding and reasoning capabilities. In internal validation for disease diagnostics, EVLF-FM achieved the highest average accuracy (0.858) and F1-score (0.797), outperforming leading generalist and specialist models. In medical visual grounding, EVLF-FM also achieved stellar performance across nine modalities with average mIOU of 0.743 and Acc@0.5 of 0.837. External validations further confirmed strong zero-shot and few-shot performance, with competitive F1-scores despite a smaller model size. Through a hybrid training strategy combining supervised and visual reinforcement fine-tuning, EVLF-FM not only achieves state-of-the-art accuracy but also exhibits step-by-step reasoning, aligning outputs with visual evidence. EVLF-FM is an early multi-disease VLM model with explainability and reasoning capabilities that could advance adoption of and trust in foundation models for real-world clinical deployment.
title EVLF-FM: Explainable Vision Language Foundation Model for Medicine
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2509.24231