GlioSeg 3D: Advancing Glioma Brain Tumor Classification & Segmentation with 3D MRI Data

One of the biggest problems in today's world is human health. The most important organ of the human body is the brain. However, many people overlook the well-being of this vital organ. Brain diseases like tumors or cancer often progress without noticeable symptoms until they reach a critical stage. Therefore, it's crucial for people to include regular brain check-ups as part of their health routine, just as they would for other organs. There are various problems or diseases are created in the brain; one of them is a brain glioma tumor. A glioma brain tumor is a type of malignant tumor that is highly cancerous and fast-growing. Its symptoms are not properly identified, so this tumor turns into cancer before people know it. Although there are different types of tumors in the human brain, one of the most common is the glioma tumor. The main objective of our work was to detect this tumor and create a 3D model for it. We have used different methods, like convolutional neural networks (CNNs), MobileNetV2, and MobileNetV3, and also worked with some machine learning models to find the most efficient solution. The CNN model was trained and tested on a dataset of brain MRI images, achieving a validation accuracy of 97.14%. Comparatively, the accuracy of the model using MobileNetV2 and MobileNetV3 was 97.80% and 82.20% respectively.