Saved in:
Bibliographic Details
Main Authors: Chopra, Shivang, Sanchez-Rodriguez, Gabriela, Mao, Lingchao, Feola, Andrew J, Li, Jing, Kira, Zsolt
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2506.08356
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915336396210176
author Chopra, Shivang
Sanchez-Rodriguez, Gabriela
Mao, Lingchao
Feola, Andrew J
Li, Jing
Kira, Zsolt
author_facet Chopra, Shivang
Sanchez-Rodriguez, Gabriela
Mao, Lingchao
Feola, Andrew J
Li, Jing
Kira, Zsolt
contents Different medical imaging modalities capture diagnostic information at varying spatial resolutions, from coarse global patterns to fine-grained localized structures. However, most existing vision-language frameworks in the medical domain apply a uniform strategy for local feature extraction, overlooking the modality-specific demands. In this work, we present MedMoE, a modular and extensible vision-language processing framework that dynamically adapts visual representation based on the diagnostic context. MedMoE incorporates a Mixture-of-Experts (MoE) module conditioned on the report type, which routes multi-scale image features through specialized expert branches trained to capture modality-specific visual semantics. These experts operate over feature pyramids derived from a Swin Transformer backbone, enabling spatially adaptive attention to clinically relevant regions. This framework produces localized visual representations aligned with textual descriptions, without requiring modality-specific supervision at inference. Empirical results on diverse medical benchmarks demonstrate that MedMoE improves alignment and retrieval performance across imaging modalities, underscoring the value of modality-specialized visual representations in clinical vision-language systems.
format Preprint
id arxiv_https___arxiv_org_abs_2506_08356
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle MedMoE: Modality-Specialized Mixture of Experts for Medical Vision-Language Understanding
Chopra, Shivang
Sanchez-Rodriguez, Gabriela
Mao, Lingchao
Feola, Andrew J
Li, Jing
Kira, Zsolt
Computer Vision and Pattern Recognition
Different medical imaging modalities capture diagnostic information at varying spatial resolutions, from coarse global patterns to fine-grained localized structures. However, most existing vision-language frameworks in the medical domain apply a uniform strategy for local feature extraction, overlooking the modality-specific demands. In this work, we present MedMoE, a modular and extensible vision-language processing framework that dynamically adapts visual representation based on the diagnostic context. MedMoE incorporates a Mixture-of-Experts (MoE) module conditioned on the report type, which routes multi-scale image features through specialized expert branches trained to capture modality-specific visual semantics. These experts operate over feature pyramids derived from a Swin Transformer backbone, enabling spatially adaptive attention to clinically relevant regions. This framework produces localized visual representations aligned with textual descriptions, without requiring modality-specific supervision at inference. Empirical results on diverse medical benchmarks demonstrate that MedMoE improves alignment and retrieval performance across imaging modalities, underscoring the value of modality-specialized visual representations in clinical vision-language systems.
title MedMoE: Modality-Specialized Mixture of Experts for Medical Vision-Language Understanding
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2506.08356