Breast Cancer Classification with ANN and DBN

As per the reports of the National Cancer Institution, breast cancer is impacting 13% of women’s lives. Breast cancer is ranked as the second major cause of death and requires an urgent diagnosis. This research on breast cancer combines machine learning and medical informatics to precisely diagnose breast cancer. A hybrid ensemble model is proposed in this research by utilizing characteristics of Artificial Neural Network and Deep Belief Network. The breast cancer Wisconsin dataset consisting of digital breast mass aspirate images is employed in this research to address the challenge of differentiating benign cases from malignant cases. The main aim is to manage the mutual prowess of individual models within the ensemble thus increasing the classification accuracy. Artificial Neural Network catches and shows the important features of the input data whereas Deep Belief Network is utilized for unsupervised pre-training to learn the conceptual features from raw data. The proposed model outperformed various machine learning techniques and shows an accuracy rate of 98.14%. So, the proposed hybrid ensemble framework shows effectiveness in the diagnosis of breast cancer, which is a significant advancement in the merging of machine learning and medical informatics. The view of magnifying classification accuracy holds enough potential to increase survival rates and patient care.