Early prediction of breast cancer based on the classification of HER‐2 and ER biomarkers using deep neural network

Abstract

AbstractBackgroundOwing to the highly coarse chromatin, multi‐dimensionality of the histo image, irregularity of shape and size, texture and appearance, nuclei extraction is challenging. To address these complexities, a deep‐learning algorithm called a stacked sparse autoencoder was considered as a research factor in this study.Methods and MaterialsThis study focuses on detecting the epithelial regions and extracting high‐level features to segment the patches based on the nuclei and classify the biomarkers concerning the nuclei patches. We used 6,53,400 microscopic image patches of 363 patients sourced from the BreakHis database, of which 4,90,050 prominent image patches containing only nuclei were utilized for biomarker classification (essentially eliminating the non‐nuclei patches from 363 whole‐slide images (WSI)). Non‐nuclei patches were eliminated owing to an imbalanced class distribution. From the nuclei patches the biomarkers oestrogen receptor (ER) and human epidermal growth factor receptor‐2 (HER‐2) that are crucial for the early diagnosis of breast cancer are predicted based on the nuclei count and Nuclei intensity. These indicators aid in predicting the risk of breast cancer as well as the severity and potential effectiveness of various treatments. About 20%–25% of breast cancer cases have an overexpression of the protein HER‐2, which makes the tumour more aggressive and on the other hand, the majority of breast cancer cases express the protein ER.ResultsThe classifier finally classified if the nuclei detected based on the features were benign, malignant, or normal with an accuracy of 99.73%, using which early prediction is performed by extracting and classifying the biomarkers HER‐2 and ER. The overall classification rate for classifying HER‐2 and ER was 97.52%.ConclusionHER‐2 +ve was classified with an intensity above 23%, and total nuclei in the range of 150–1000 were termed ER‐positive. Based on these 40 HER‐2 +ve patients, 25 ER +ve patients were detected in 363 patients. From this observation, it is concluded that 25–40 patients are at risk of breast cancer in the next 5 years due to the cell proliferation rate of 7000.