Breast cancer detection based on thermographic images using machine learning and deep learning algorithms

According to the latest data, breast carcinoma is the most prevalent kind of cancer in the world, and it is responsible for the deaths of almost 900 thousand people each year. If the disease is detected at the early stage and diagnosed properly, it can improve the chance of positive outcomes, thus reducing the fatality rate. An early diagnosis in fact can help in preventing it to spread and saves the premature victims from obtaining it. When trying to distinguish among benign and malignant tumors, as well as when trying to draw conclusions about mild and advanced breast cancer, researchers who study cancer encounter a number of challenges. The identification of all tumors is accomplished through the application of machine learning, which makes use of algorithms that are able to locate and recognize patterns. All of them, however, revolve around the concept of "binary grouping," as was mentioned earlier (malignant and benign; no-cancer and cancer). In this study, we propose a Computer-aided Diagnosis (CAD) method for the identification and diagnosis of patients into 3 classes (cancer, no cancer, and non-cancerous) under the management of a database. CAD is an abbreviation for computer-aided diagnosis. The Convolutional Neural Network (CNN), the Support Vector Machine (SVM), and Random Forest are all remarkable classifiers (RF). Convolution Networks, Support Vector Machines (SVM), and Random Forest are the three effective classifiers that we look into and analyses for the classification stage (RF). In addition to this, we investigate the impact of the mammography pictures being pre-processed in advance, which allows for a higher success rate in categorization.