Enhancing Breast Cancer Detection with Radiomics and Machine Learning: A Comprehensive Analysis Using MRI Datasets

Breast cancer is a major global health issue, driving the need for advanced early detection methods to improve treatment and prognosis. This study utilizes radiomics and machine learning to enhance breast cancer detection through MRI datasets. Radiomics converts traditional images into high-dimensional data for mining predictive biomarkers with machine learning algorithms. Our comprehensive approach includes sophisticated image segmentation to isolate regions of interest (ROIs), extracting radiomic features like shape, texture, and edge sharpness using techniques such as gray level co-occurrence matrix, gray level run length matrix, and others. These features were analyzed using support vector machines, random forests, k-nearest neighbors, decision trees, and artificial neural networks to classify the scans into normal or malignant categories. The results show a notable increase in diagnostic accuracy. The artificial neural network (ANN) model achieved the highest performance with an accuracy of 95%, precision of 93%, recall of 94%, and an F1 score of 93.5%. Conversely, the decision tree (DT) model had the lowest performance with an accuracy of 89%, precision of 87%, recall of 88%, and an F1 score of 87.5%.