Devising a Breast Cancer Diagnosis Protocol through Machine Learning

Abstract Breast cancer is a life-threatening disease and has serious health implications. It is categorized based on receptors, including the estrogen receptor (ER) and human epidermal growth factor receptor 2 (HER2), which are the focus of the present research We analyzed gene expression from data obtained from a functional genomics repository called Array Express. The accession numbers are E-GEOD-52194, E-GEOD-75367, and E-GEOD-58135, and the molecular details of these subsets of cancer receptors. Upon following a predefined computational pipeline, we identified 369 genes that had distinct patterns of gene expression profiles in cases of ER-positive (ER + ) and HER2-negative (HER2-) breast cancer. The support vector machine (SVM) and decision tree models of machine learning were used to evaluate the prognostic and diagnostic significance. Accuracy, sensitivity, and specificity were examined to gauge the effectiveness of these models. Then, a network analysis was performed to assess the significant biological process and signaling pathways of HER2- and ER+ breast cancer development. The present study facilitates an enhanced approach to these subcategories of breast cancer so that precise diagnoses can be made, and better and more focused treatment plans can be provided. The current research provides valuable information on the molecular and genetic basis of ER+ and HER2- breast cancer and has great potential for improving patients' treatment.