An Intelligent Deep Learning Strategy for Breast Cancer prediction using feature ensemble learning

One in six women will be diagnosed with breast cancer (BC) by the time they reach their forties, making it the leading cause of death among this demographic. It's the leading killer in the world, yet it's hard to spot and diagnose in its early stages. Among women, BC is a foremost origin of death in third world countries. The prevention and treatment of these tumors can be greatly aided by screening and early diagnosis. At the moment, BC is the leading malignancy in females, even more so than ovarian cancer. When researchers get access to patient medical records, they can uncover hidden patterns in healthcare. The National-Cancer-Institute (NCI) reports that mortality rates from breast cancer can be reduced with early detection. In order to accurately identify patients with cancer, this work presents a voting classifier based on ensemble learning that integrates the logistic-regression(LR) & stochastic-gradient-descent(SGD) with deep-convoluted features. The ensemble voting classifier is fed deeply-convoluted features mined from the microscopic characteristics. This concept offers a refined framework for differentiating amid benign and malignant tumors. The results show that the highest level of classification accuracy, 100%, may be attained when utilizing voting classifiers with complex features. When compared to the traditional methods, the proposed method showed an improvement in accuracy. Thus, the research aids doctors in diagnosing and avoiding breast cancer.