Methods to Detect Breast Cancer Using Traditional Machine Learning Technique

In this research paper, a computer-aided diagnostic system based on image processing techniques and deep learning is proposed for the classification of breast cancer in mammograms. In the preprocessing sets for image processing pectoral muscle removal, An improved mammography is created by applying contrast-limited adaptive histogram equalization algorithms for image advancement.. Finally, the region of interest was generated using gradient field coverage features based thresholding technique. Next, the system uses a reformed ResNet-50 model for feature extraction. The gleaned features are then classified using three classifiers and their performance is compared. The classifiers belonging to different classification families namely AdaBoost, support vector machine (SVM), and random forest. The overall framework achieved high accuracy, sensitivity, and specificity, demonstrating its potential as a powerful tool for breast cancer diagnosis. Furthermore, the proposed image processing and feature extraction technique shows superior performance compared to individual feature sets, illustrating the significance of the suggested approach.