Vol. 4, Issue 12, Part E (2018)

Abstract
Breast cancer is a major public health concern that impacts millions of women worldwide. Effective treatment depends on an early diagnosis, and mammography is the primary screening technique. mammography has limitations such as low sensitivity and the possibility of false positives in women with dense breasts. We created a deep learning-based breast cancer diagnosis in this work and evaluated its efficacy against other models. Our approach includes data collection, prioritization, design, evaluation and interpretation. We collected mammogram images and their corresponding diagnostic information from the Mammogram Screening Digital Database (DDSM) and used a convolutional neural network (CNN) to build our model. We evaluate the performance of our model using metrics such as accuracy, sensitivity and specificity and compare it with existing models. Our results show that our model outperforms existing methods in accuracy and sensitivity, demonstrating the potential of deep learning in tumor diagnosis.