Attention-Map Augmentation for Hypercomplex Breast Cancer Classification

TL;DR

This paper introduces a novel hypercomplex neural network framework with attention maps for improved breast cancer classification, effectively focusing on regions of interest and modeling local relations in mammography and histopathology images.

Contribution

The paper proposes a parameterized hypercomplex attention map framework that enhances breast cancer classification by focusing on ROIs and exploiting local relations in the data.

Findings

Outperforms state-of-the-art attention-based networks.

Surpasses real-valued counterparts in classification accuracy.

Effective on both mammography and histopathological images.

Abstract

Breast cancer is the most widespread neoplasm among women and early detection of this disease is critical. Deep learning techniques have become of great interest to improve diagnostic performance. However, distinguishing between malignant and benign masses in whole mammograms poses a challenge, as they appear nearly identical to an untrained eye, and the region of interest (ROI) constitutes only a small fraction of the entire image. In this paper, we propose a framework, parameterized hypercomplex attention maps (PHAM), to overcome these problems. Specifically, we deploy an augmentation step based on computing attention maps. Then, the attention maps are used to condition the classification step by constructing a multi-dimensional input comprised of the original breast cancer image and the corresponding attention map. In this step, a parameterized hypercomplex neural network (PHNN) is employed to perform breast cancer classification. The framework offers two main advantages. First, attention maps provide critical information regarding the ROI and allow the neural model to concentrate on it. Second, the hypercomplex architecture has the ability to model local relations between input dimensions thanks to hypercomplex algebra rules, thus properly exploiting the information provided by the attention map. We demonstrate the efficacy of the proposed framework on both mammography images as well as histopathological ones. We surpass attention-based state-of-the-art networks and the real-valued counterpart of our approach. The code of our work is available at https://github.com/ispamm/AttentionBCS.