An Improved Diagnostic Process to Identify Breast Cancer in Earlier Stages Using Deep Learning Mechanism

Breast cancer remains one of the leading causes of mortality among women globally, with early diagnosis being vital for effective treatment and better prognosis. Early detection through non-invasive imaging techniques such as ultrasound imaging has gained significant attention. However, traditional diagnostic methods often rely on manual evaluations, which can be subjective, time-consuming, and prone to inaccuracies. This study introduces a novel diagnostic system leveraging a Cascaded Artificial Neural Network (ANN) with Support Vector Machine (SVM) to analyze ultrasound images for the early detection of breast cancer. The proposed hybrid model integrates the feature extraction capabilities of ANN and the decision-making strengths of SVM to achieve superior performance in classifying early-stage breast cancer. The Breast Ultrasound Images Dataset from Kaggle utilized, and preprocessing steps included normalization, feature extraction, and adaptive contrast enhancement techniques to prepare the images for model training. The proposed Cascaded ANN with SVM system compared with nine other conventional machine learning (ML) models, including Logistic Regression (LR), Random Forest (RF), CNN, and SVM. The system achieved an accuracy of 97.75%, a precision of 96.88%, a recall of 98.25%, and an F1-Score of 97.56<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">%</sup>, demonstrating a substantial improvement over existing models.