Brain Tumor Detection and Classification is an application build to classify an MRI scan of the brain into malignant types of Glioblastoma Multiforma and High-Grade Glioma and benign types of Meningioma and Pituitary Adenoma.
In today’s world, the brain tumor is a very rare but a deadly disease which you would not wish on anyone. It is caused by abnormal growth of cells in the brain. The norm right now for tumor classification is manually inspecting the magnetic resonance images which is an invasive method. Once the MR images are inspected, a biopsy is performed by the surgeon. The biopsy is performed to remove small piece of brain tissue which is tested for abnormalities. MRI scans play an integral role in the diagnosis procedure. Human inspection may lead to errors which may be deadly in cases of dealing with the brain tumor. It is tedious as well because a tumor is of many kinds and the treatment may differ based on the kind. The objective of the project is to provide a non-invasive method to segment and classify the tumor into benign or malignant category. The MRI scans of the brain will be segmented and the tumor part will be separated for further analysis. After removing noise and segmentation, the segmented image will be sent for feature extraction which will then classify the brain tumor to benign or malignant. Further sub-classification will take place such as Glioblastoma Multiforme(GBM) or glioma under malignant type and meningioma or pituitary adenoma under benign type tumor.
The project is completely on MATLAB. Please read the User Manual, Segmentation implementation and Classification implementation to get better idea about how to operate the application.