Deep Learning-Based Intelligent Diagnostics Framework For The Labeling Of Mammogram Image

Abstract

Around the world, women deal with breast cancer as a major health concern. The survival rate for breast cancer may be greatly improved by detecting anomalies at an earlier stage. When it comes to screening for and detecting breast cancer, a mammography is considered a reliable and popular model. Potentially useful for detecting breast cancer from other types of cancers, especially those with smaller tumor cells. When it comes to early detection, radiological screening is paramount. In order to categorise breast lesions, digitalized diagnostic systems have lately made heavy use of mammography screening models. The slight variation in X-ray permeability between normal and abnormal areas makes cancer identification challenging, despite mammography being recognized as the most effective radiological screening procedure for breast inquiry and diagnosis. This problem becomes worse as the breast tissue gets thicker. Consequently, in order to raise the detection rate and lower the mortality rate, a CAD model is necessary. In most cases, CAD models rely on ML approaches to identify tumors in digital mammography images. Radiology professionals have been able to improve diagnostic accuracy and prediction accuracy with the use of deep learning (DL) models during the last several decades. Histopathology pictures and tissue categorization are only two examples of the many clinical imaging applications that make use of DL approaches.