What question did this study set out to answer?

The aim is to accurately predict individual responses to Anti-VEGF treatment in patients with diabetic macular edema.

March 7, 2026

MVTT-GMamba: A Multimodal Graph Reasoning Framework for Anti-VEGF Efficacy Prediction in Diabetic Macular Edema

SWShijing WuWenzhou Medical University YZYifan ZhengEdinburgh College TCTianyu ChenHangzhou Dianzi University

Key Points

The aim is to accurately predict individual responses to Anti-VEGF treatment in patients with diabetic macular edema.
Developed a multimodal learning framework combining OCT images and clinical indicators.
Implemented feature-wise heterogeneous graph reasoning to model patient and feature relations.
Utilized domain-specific encoders and a cross-attention module for enhanced representation learning.
Conducted experiments on both private and public datasets comparing performance against existing methods.
MVTT-GMamba outperformed state-of-the-art methods in all evaluation metrics.
The model demonstrated effective feature extraction and fusion, reducing modality redundancy.
Grad-CAM visualizations indicated the model focuses on important retinal areas, enhancing interpretability.

Abstract

Accurately predicting individual responses to Anti-Vascular Endothelial Growth Factor (Anti-VEGF) efficacy in diabetic macular edema (DME) remains a critical challenge in personalized ophthalmic care. Existing methods often rely on unimodal data or suffer from ineffective multimodal feature extraction and fusion, leading to modality redundancy and performance degradation. To address these limitations, we propose MVTT-GMamba, a novel multimodal learning framework that integrates optical coherence tomography (OCT) images and structured clinical indicators for early and accurate Anti-VEGF efficacy prediction. At the core of MVTT-GMamba is a feature-wise heterogeneous graph reasoning paradigm that explicitly models inter-patient and inter-feature relations, together with an adaptive, graph-guided prediction head that progressively anneals structural priors into the classifier. Building on this core, we adopt domain-tailored MambaVision and TabTransformer encoders and an early cross-attention fusion module to realize fine-grained multimodal representation learning. Extensive experiments on both a private clinical dataset (DMETHERA-ECSAHZU) and the public APTOS2021 benchmark demonstrate that MVTT-GMamba consistently outperforms state-of-the-art methods across all evaluation metrics. In addition, Grad-CAM visualizations reveal that the model attends to clinically relevant retinal regions, providing enhanced interpretability. Code is available at: https://github.com/DME666/DME.

AI에게 질문

Bookmark

View Full Paper