An Efficient Identification of Breast Cancer using Linear Discriminant Analysis on Histopathological Images
G. Sajiv, G. Ramkumar · 2022
There is a significant amount of interest in machine learning, and scientists generally examine its application in every possible setting. Breast cancer is one of the diseases that have the highest mortality rate among women. As a result of a delay in the detection of malignant tissues, it is also known as the second biggest cause of mortality in the world. Postmenopausal symptoms, obesity, genes passed down from families, hormonal imbalance, and genetic abnormalities that are not expressed are the most prevalent risk factors for breast cancer. Even though breast cancer is being discovered at a later stage in many underdeveloped nations, death rates remain excessively high. Early detection is essential if we are to bring the mortality toll down and save more lives. Histopathological pictures are the types of data that are taken into consideration by the suggested method. In this article, we will concentrate mostly on distinguishing between benign and malignant forms of cancer. In this investigation, the diagnosis of breast cancer was accomplished by the application of linear discriminant analysis (LDA). The model is evaluated with the assistance of efficiency indicators such as accuracy, sensitivity, and specificity, as well as the F1-score.