Detection of Female Breast Cancer Based Digitized Image using Machine Learning Techniques

Abstract

Early detection of breast cancer dramatically increases the odds of treatment and strategies are needed that allow fast, accurate and inexpensive early detection and detection. Vibrational spectroscopy is a forward-looking approach with all these features. A characterization and visualization strategy focused on statistical evidence for automated diagnosis needed to take the next step toward transforming this technology into a clinical instrument. In this paper, a diagnostic model was developed from the axillary lymph node tissue of patients suffering from breast cancer. Various classification methods have been examined for this reason. The best choice of classification system was a support vector machine (SVM) as it accurately categorized 100 per cent of the unseen test sample. The resultant diagnostic frameworks have been carefully checked for their strength against spectral corruption predicted during routine clinical testing. It shows that a potential diagnostic routine has adequate robustness. Strategies for image processing have been built in this work towards a fully automatic identification of vibrational experimental observations. These technologies may identify interesting characteristics and foresee tissue pathology.