a. arjmand

Breast cancer is the most common type of cancer that affects the female population. Early detection of cancer can increase the chance of treatment and is also the most effective way to fight the disease. The development of automated methods for the detection of cancer or tumor mass has been of interest to researchers. In this paper, a method based on deep convolutional neural networks for detecting tumor area from MRI images is introduced. The proposed method is to collect MRI images along with GT images from their tumor area and expand the data to train and test the neural network. The type of learning method used in this paper is supervised learning. The algorithm is tested on the RIDER breast dataset and the results show that the proposed method performs better than other image detection methods such as clustering methods. Benefits include high quality in tumor detection and acceptable speed at runtime.