SHISRCNet: Super-resolution And Classification Network For Low-resolution Breast Cancer Histopathology Image

TL;DR

SHISRCNet is a novel deep learning model that enhances low-resolution breast cancer histopathology images through super-resolution and multi-scale feature classification, improving diagnostic accuracy with limited hardware.

Contribution

The paper introduces SHISRCNet, combining super-resolution and multi-scale feature fusion for low-resolution images, addressing hardware limitations and fixed receptive field issues in prior methods.

Findings

Achieves near state-of-the-art accuracy with low-resolution inputs.

Effectively reconstructs high-quality images from LR inputs.

Enhances classification performance by multi-scale feature fusion.

Abstract

The rapid identification and accurate diagnosis of breast cancer, known as the killer of women, have become greatly significant for those patients. Numerous breast cancer histopathological image classification methods have been proposed. But they still suffer from two problems. (1) These methods can only hand high-resolution (HR) images. However, the low-resolution (LR) images are often collected by the digital slide scanner with limited hardware conditions. Compared with HR images, LR images often lose some key features like texture, which deeply affects the accuracy of diagnosis. (2) The existing methods have fixed receptive fields, so they can not extract and fuse multi-scale features well for images with different magnification factors. To fill these gaps, we present a \textbf{S}ingle \textbf{H}istopathological \textbf{I}mage \textbf{S}uper-\textbf{R}esolution \textbf{C}lassification network (SHISRCNet), which consists of two modules: Super-Resolution (SR) and Classification (CF) modules. SR module reconstructs LR images into SR ones. CF module extracts and fuses the multi-scale features of SR images for classification. In the training stage, we introduce HR images into the CF module to enhance SHISRCNet's performance. Finally, through the joint training of these two modules, super-resolution and classified of LR images are integrated into our model. The experimental results demonstrate that the effects of our method are close to the SOTA methods with taking HR images as inputs.