Brain Tumour Classification Based on MRI Scans Using Deep Learning Approaches

Wong, Jia Ying (2023) Brain Tumour Classification Based on MRI Scans Using Deep Learning Approaches. Final Year Project (Bachelor), Tunku Abdul Rahman University of Management and Technology.

Text Full Text - WongJiaYing.pdf Restricted to Registered users only Download (5MB)

Abstract

Brain Tumour is known as a lump or growth of abnormal cells in the brain. There are numerous varieties of brain tumours. Both benign or can be known as noncancerous and malignant, or can be known as cancerous brain tumours can occur (malignant). Primary brain tumours are those that start in the brain; secondary (metastatic) brain tumours are those that start in other regions of the body and spread to the brain. A subset of Machine Learning, known as Deep Learning, is simply a neural network with three or more layers These neural networks make an effort to mimic how the human brain functions, however they fall far short of being able to match it, enabling it to "learn" from vast volumes of data. Many artificial intelligence (AI) applications and services are fueled by deep learning, which enhances automation by carrying out mental and physical tasks without the need for human intervention. Deep learning is the technology that powers both established and emerging technologies, like voice-activated TV remote controls, digital assistants, and etc. In this paper, Deep Learning is utilized, specifically Convolutional Neural Network (CNN)-based model is proposed, to detect and classify Brain Tumour, into four classes, which is either Healthy Brain with no Tumour, Glioma Tumour, Meningioma Tumour or Pituitary Tumour using the dataset from Kaggle, a public online community platform. As current system and process of diagnosis of Brain Tumour, is through manual inspection of MRI scans and detection of abnormalities on MRI by Radiologists and Doctors, which has lower efficiency and prone to make mistakes. Artificial Intelligence is utilized, specifically Deep Learning algorithm was implemented to boost efficiency of diagnosis process by reducing the scope of other possible diseases.