After lung cancer, breast cancer is the most common kind of the illness. In terms of prevalence, lung cancer is far and by the winner. When comparing males and women of reproductive age, breast cancer disproportionately affects women. Since the root cause of breast cancer remains unknown, early diagnosis is crucial for lowering mortality rates. Cancer survival rates might improve by as much as 8% if diagnosed and treated early. X-rays, mammograms, and even MRIs may fall under this category. Exactly what is the problem Extremely small masses and lumps may be difficult for even the most experienced radiologists to detect, which can lead to a high rate of false positives and negatives. To put it mildly, this is a very worrying development. Many individuals are working to improve breast cancer detection applications so that the disease may be detected at an earlier stage. New technology allows for the analysis of photographs, which may then be used to teach itself. In order to distinguish between calcifications, masses, asymmetry, and carcinomas, we used a Deep Convolutional Neural Network (CNN) in this study. Studies that came before us often used simple methods to achieve this end. Cancer was classified as benign or malignant, allowing for more targeted therapy. Previous training has been completed on a model. We first use this strategy for completing transfer learning effectively. ResNet50. Our model was similarly improved to better accommodate deep learning. The significance of a neural network's learning rate during its training phase cannot be emphasised. Using the technique we provide, the learning pace may be adjusted to fit new circumstances. An individual learning anything new is certain to make a few blunders along the way.

Convolutional Neural Network ResNet50 X-rays Cancer DCT.