convoHER2: A Deep Neural Network for Multi-Stage Classification of HER2 Breast Cancer

TL;DR

This study introduces convoHER2, a CNN-based model that classifies HER2 breast cancer stages from high-resolution histopathological images with high accuracy, potentially improving diagnostic efficiency and reducing costs.

Contribution

The paper presents a novel deep learning model, convoHER2, for multi-stage HER2 breast cancer classification using histopathological images, achieving high accuracy and demonstrating clinical potential.

Findings

ConvoHER2 achieved 85% accuracy with H&E images.

ConvoHER2 achieved 88% accuracy with IHC images.

The model can classify cancer stages effectively from high-resolution images.

Abstract

Generally, human epidermal growth factor 2 (HER2) breast cancer is more aggressive than other kinds of breast cancer. Currently, HER2 breast cancer is detected using expensive medical tests are most expensive. Therefore, the aim of this study was to develop a computational model named convoHER2 for detecting HER2 breast cancer with image data using convolution neural network (CNN). Hematoxylin and eosin (H&E) and immunohistochemical (IHC) stained images has been used as raw data from the Bayesian information criterion (BIC) benchmark dataset. This dataset consists of 4873 images of H&E and IHC. Among all images of the dataset, 3896 and 977 images are applied to train and test the convoHER2 model, respectively. As all the images are in high resolution, we resize them so that we can feed them in our convoHER2 model. The cancerous samples images are classified into four classes based on the stage of the cancer (0+, 1+, 2+, 3+). The convoHER2 model is able to detect HER2 cancer and its grade with accuracy 85% and 88% using H&E images and IHC images, respectively. The outcomes of this study determined that the HER2 cancer detecting rates of the convoHER2 model are much enough to provide better diagnosis to the patient for recovering their HER2 breast cancer in future.