Enhancing Early Breast Cancer Detection Using Transfer Learning, Deep Feature Extraction and SVM Classification

In recent years, breast cancer has emerged as one of the most common types of cancer among women globally.Late diagnosis often results in a higher mortality rate, underlining the crucial need for effective early detection systems, especially those leveraging personal imaging data.In this regard, this study focuses on the integration of transfer learning and deep feature extraction techniques to adapt pretrained CNN models to breast cancer detection.Specifically, AlexNet and VGG16 are employed for feature extraction, with further tuning conducted on AlexNet.The derived features are then classified using Support Vector Machines (SVM).A comprehensive evaluation is carried out using a publicly available breast cancer dataset based on surgical and cellular pathology.Accuracy scores serve as performance metrics.The findings underscore the superior performance of the transfer learning approach compared to the combination of deep feature extraction and SVM classification.