Multi-View Hypercomplex Learning for Breast Cancer Screening

TL;DR

This paper introduces multi-view hypercomplex neural networks for breast cancer screening, capturing view correlations efficiently and accurately, outperforming existing models across multiple radiographic tasks.

Contribution

It presents a novel hypercomplex learning paradigm with specialized architectures that improve multi-view breast cancer classification and generalize across different medical imaging tasks.

Findings

Outperforms state-of-the-art multi-view models in accuracy.

Efficiently captures intra- and inter-view relations.

Generalizes across radiographic modalities and tasks.

Abstract

Radiologists interpret mammography exams by jointly analyzing all four views, as correlations among them are crucial for accurate diagnosis. Recent methods employ dedicated fusion blocks to capture such dependencies, but these are often hindered by view dominance, training instability, and computational overhead. To address these challenges, we introduce multi-view hypercomplex learning, a novel learning paradigm for multi-view breast cancer classification based on parameterized hypercomplex neural networks (PHNNs). Thanks to hypercomplex algebra, our models intrinsically capture both intra- and inter-view relations. We propose PHResNets for two-view exams and two complementary four-view architectures: PHYBOnet, optimized for efficiency, and PHYSEnet, optimized for accuracy. Extensive experiments demonstrate that our approach consistently outperforms state-of-the-art multi-view models, while also generalizing across radiographic modalities and tasks such as disease classification from chest X-rays and multimodal brain tumor segmentation. Full code and pretrained models are available at https://github.com/ispamm/PHBreast.