Brain Tumor Detection using Deep Learning

Abstract

Brain tumor is a group of unfamiliar cells present in the cerebrum that may lead to cancer. Brain tumor can be diagnosed by several easy ways but among them MRI imaging is the best way to discover a tumor in the brain. To form images of a brain it uses radio waves and magnetic field so that the information regarding the abnormal tissue growth in the brain can be identified. The detection of brain tumor is done by using deep learning algorithm which somewhat a branch or a subset of machine learning and in this proposed system we will use the CNN(Convolution neural network) model to determine whether the brain has a tumor or not and all of this can only be done by using MRI(Magnetic resonance imaging) scans. CNN is a kind of network architecture for deep learning algorithms and is used specifically for image recognition and processing the pixel data. Brain tumor detection is done quickly with higher accuracy when these algorithms were used on the MRI scans Images which also helps in providing treatment to the diagnosed person. These predictions also help the radiologists in making the decisions quickly. The steps for creating this project is as follows: A collection of data containing images from MRI scans from various sources. In addition to that the data will be pre- processed and augmented. Next, a neural network model for tumor detection will be developed and in the end model training and testing will be performed. In generally, CNN offers a fairly high training accuracy and the error rate is also very low as compared to other techniques