Publication

Abstract : Breast cancer is the cause of most frequent cancer-related death in women around the globe, with a five-year survival rate that is significantly lower than that of many other cancers. Much research has been done on identifying and spotting breast cancer using various image processing and classification approaches. Despite this, the disease remains one of the most lethal. Because the root of breast cancer remains unknown, avoidance is impossible. Breast cancer can only be cured if tumours in the breast are detected early. We can detect breast cancer using bag of deep visual words on histopathological images. The rate of mortality can be effectively reduced if real discovery is made. Our model can aid in the early identification of breast cancer. Here, we quantitatively depict the study methods used for feature detection of cancer by using VGG19, K-means, and histogram generation. These features are compared and processed via an SVM classifier to better comprehend the cancer pattern in the histopathology image.