Regresar

Cutting-Edge Advanced Machine Learning Model for Enhanced Breast Cancer Diagnostics

Abstract:

Breast cancer is one of the most severe and prevalent cancers. It occurs in all of the countries around the world in women at any age. Nowadays it is increasing its morbidity and mortality rates. This paper presents a novel segmentation model to increase the quality measures applied to the detection of benign and malignant pathologies in the context of breast cancer through the application of convolutional operations, segmentation procedures, and bagging between support vector classifiers (SVC) to optimize the diagnostic process. It analyses the Chinese Mammography Database [1] (mammography images of breast carcinoma), aiming to evaluate the performance of these Machine Learning algorithms. It processes and analyses pixels that these techniques have learned as the most relevant, giving insights into the decision-making process of the SVC accuracy, precision, specificity, and sensitivity. The algorithm focuses on breast carcinoma, obtaining results that evidence the promising performance, increasing the quality of the results obtained in contrast with other methodologies, allowing a more efficient procedure to enhance the application of the model and its behavior.