Brain tumour segmentation using CNN

Abstract

The determination of tumour extent is a major challenging task in brain tumour and quantitative evaluation. Magnetic Resonance Imaging (MRI) is one of the non-invasive techniques that has emanated as a front-line diagnostic tool for brain tumour without ionizing radiation. Among brain tumours, gliomas are the most aggressive, leading to a very short life expectancy in their highest grade. In the clinical practice manual segmentation is a time-consuming task and their performance is highly depended on the operator’s experience. This project proposes fully automatic segmentation of brain tumour using CNN. It uses brain MR images together with manual FLAIR abnormality segmentation masks. Using RESNET-50 tumour detection is first done and then using RESUNET, brain tumour segmentation is done. Hence, this project detects brain tumour and if it exists, segments the brain tumour. This is an essential step in diagnosis and treatment planning, both of which is needed to be done quickly to maximise the likelihood of successful treatment.