Acknowledging the current limitations of breast cancer detection using machine learning techniques, with a detection rate at 50%, this application aims to provide users with access to early detection capabilities. We utilized mammogram images from datasets such as INbreast and CBIS-DDSM, containing a diverse range of breast mammography images including masses, benign, normal, and cancer volumes. Our model builds upon the ResNet-50 architecture, a model pre-trained on the ImageNet dataset, providing robust capabilities in deciphering images with low signal-to-noise ratios. We further enhanced the ResNet-50 by chaining it with other models, such as the CNN, to improve its performance in breast cancer detection.