ConvNet Based Detection and Segmentation of Brain Tumor from MR Images

One of the diseases that affects humans is brain tumor. It is a type of malignancy disease. A brain tumor is aberrant brain cells that has grown out of control in the brain. This sickness affects many people, and it might be difficult to survive in large groups. When allowing people for early detection of brain tumor, it will help to survive and reduce the death rate of people. Detection of aberrant cells formation in brain is very difficult in medical imaging. The Detection is done by using magnetic resonance imaging (MRI). In this paper, ConvNet architecture is proposed with transfer learning to detect tumor and it aims to differentiate the tumor area by using ROI and non-ROI. The data set is taken from open source Kaggle repository. This model obtained 98.1% accuracy on test data set. This model performed state of the art work.