Date of Publication :30th November 2021

Abstract: We have developed a web application using Artificial intelligence and Deep neural networks that studies the human brain tumor MRI images and classifies the tumor into Malignant and Benign category also highlighting the tumor portion and providing necessary information in an easier way such that it is understandable by everyone. The system will be accessible to all doctors only and not patients as knowing about the tumor will definitely be panicking for the patient and the situation will turn more problematic. Doctor will get the result and guide the patient. Tumor segmentation is an important task in the model for which we are using Fuzzy C-Means clustering and then for further classification, we are using a 4 layer Convolutional Neural Network. As compared to other traditional classifiers from the previous studies that we did, our model using CNN gave the best results. Brain tumor detection is a difficult task for doctors and specialists as well. Hence a trained and reliable model can be very useful to them to ease their job and reduce the pressure on human judgement. We aim to build a model that can prove very useful to the medical and healthcare sector and revolutionize the disease detection process in the sector

Reference :