BREAST-CAD: A Computer-Aided Diagnosis System for Breast Cancer Detection Using Machine Learning

Abstract

This research presents a novel Computer-Aided Diagnosis (CAD) system called BREAST-CAD, developed to support clinicians in breast cancer detection. Our approach follows a three-phase methodology: Initially, a comprehensive literature review between 2000 and 2024 informed the choice of a suitable dataset and the selection of Naive Bayes (NB), K-Nearest Neighbors (KNN), Support Vector Machines (SVM), and Decision Trees (DT) Machine Learning (ML) algorithms. Subsequently, the dataset was preprocessed and the four ML models were trained and validated, with the DT model achieving superior accuracy. We developed a novel, integrated client–server architecture for real-time diagnostic support, an aspect often underexplored in the current CAD literature. In the final phase, the DT model was embedded within a user-friendly client application, empowering clinicians to input patient diagnostic data directly and receive immediate, AI-driven predictions of cancer probability, with results securely transmitted and managed by a dedicated server, facilitating remote access and centralized data storage and ensuring data integrity.