Main Article Content
Cancer, undeniably, is one of the most dangerous threats to public health worldwide. Breast Cancer in specific accounts for about 12.3% of all cancer cases, being the second most common type of cancer among women. Recent statistics reveal that Breast Cancer cases among Indian women are on the rise. With a new case coming up every 4 minutes, it's now prominently present even among younger age groups. Breast Cancer Detection requires a histopathologist to carry out a tissue biopsy & determine the presence of abnormal cells. Even the slightest slip in precision can result in False Positive or False Negative results. Over the years, medical advancements have made this process simpler. Nowadays, Machine Learning approaches such as K-Nearest Neighbour, Support Vector Machine, etc., are employed. These house effective diagnostic capabilities but aren't 100% accurate. Further, these achieve high accuracy only for Binary classification (Benign/Malignant) and support low accuracies for Multi-Class classification & early-stage detection, posing a need to improve diagnosis quality. In this paper, we represent the application of DenseNet Classifier and Random Forest Algorithm for Breast Cancer Detection, Classification & Prediction. We propose a robust integrated system capable of detecting Breast Cancer along its specific stage. Further, if a suspected patient comes out to be healthy, it will be able to; predict their chances of developing Breast Cancer in future. Results of this implementation show that the Random Forest Algorithm gives high accuracy of 97.07% for prediction. Similarly, the DenseNet Classifier achieves an exceptional accuracy of 97.59% for detection.