Mammo-Mamba: A Hybrid State-Space and Transformer Architecture with Sequential Mixture of Experts for Multi-View Mammography

TL;DR

Mammo-Mamba is a novel hybrid architecture combining state-space models, transformers, and mixture of experts to improve multi-view mammogram classification efficiency and accuracy.

Contribution

It introduces the SecMamba block with Sequential Mixture of Experts, enhancing high-resolution mammogram feature learning within a unified framework.

Findings

Achieves superior classification performance on CBIS-DDSM dataset.

Maintains computational efficiency compared to traditional Transformer models.

Effectively mitigates Transformer limitations with dynamic expert gating.

Abstract

Breast cancer (BC) remains one of the leading causes of cancer-related mortality among women, despite recent advances in Computer-Aided Diagnosis (CAD) systems. Accurate and efficient interpretation of multi-view mammograms is essential for early detection, driving a surge of interest in Artificial Intelligence (AI)-powered CAD models. While state-of-the-art multi-view mammogram classification models are largely based on Transformer architectures, their computational complexity scales quadratically with the number of image patches, highlighting the need for more efficient alternatives. To address this challenge, we propose Mammo-Mamba, a novel framework that integrates Selective State-Space Models (SSMs), transformer-based attention, and expert-driven feature refinement into a unified architecture. Mammo-Mamba extends the MambaVision backbone by introducing the Sequential Mixture of Experts (SeqMoE) mechanism through its customized SecMamba block. The SecMamba is a modified MambaVision block that enhances representation learning in high-resolution mammographic images by enabling content-adaptive feature refinement. These blocks are integrated into the deeper stages of MambaVision, allowing the model to progressively adjust feature emphasis through dynamic expert gating, effectively mitigating the limitations of traditional Transformer models. Evaluated on the CBIS-DDSM benchmark dataset, Mammo-Mamba achieves superior classification performance across all key metrics while maintaining computational efficiency.