/Brain_cancer_classification

In this project, we will implement and compare ResNet, AlexNet and MLP on brain cancer T2-weighted MRI image.

Primary LanguageJupyter NotebookMIT LicenseMIT

Brain Cancer MRI Images Classification

Cancer is the second leading cause of death after cardiovascular diseases. Out of all types of cancer, brain cancer has the lowest survival rate. Brain tumors can have different types depending on several factors such as the shape, texture, and location of the tumor (e.g., Acoustic Neuroma, Meningioma, Pituitary, Glioma, CNS Lymphoma . . . etc). In clinical practice, the incident rates of Glioma, Meningioma, and Pituitary tumors are approximately 45%, 15%, and 15%, respectively, among all brain tumors. Proper diagnosis of the tumor type enables the doctor to make the correct treatment choice and help save the patient’s life.

Currently, anomaly detection through MRI is manual mostly and clinicians have to spend a lot of time to detect and segment the tumor for treatment and surgical purpose. This manual technique is also prone to errors and can compromise life. Also, diversity of Tumor types, makes the detection more difficult due to the complex structure of the brain. In order to resolve these issues, studies have started to focus on various machine learning and Deep Learning techniques for computer-based tumor detection and segmentation.

This paper reviews different papers that tried to do the task of brain cancer MRI image classification. In this project, we will implement and compare some of these methods.