Brain Tumor Detection and Marker-Based Watershed Segmentation for Size Estimation using CNN

Brain tumors are a common and severe malignant tumor condition that can reduce one's life if not diagnosed early enough. The categorization of brain tumors is a difficult task. Furthermore, when there is a huge amount of data to be assisted, it is a challenging work. Because brain tumors vary in appearance and share similarities with normal tissues, it is challenging to isolate tumor regions. In this algorithm, we suggested a model that first uses a convolutional neural network and classical classifiers, then uses the Fuzzy C-Means clustering approach to remove brain tumors from 2D MRI brain scans. The structure of available data sets, brain tumors, component extraction, augmentation techniques, Transfer Learning (TL), Deep Learning (DL) classification, and reinforcement learning (RL) are also examined in this work.