Brain MRI Segmentation and Tumor Detection: Challenges, Techniques and Applications

Abstract

To detect a brain tumor and its growth from MRI of brain image is one of the most common techniques in medical research field. From the internal structure of human brain, the scanning process of brain image gives more detailed information about the growth of brain tumor. Despite many years of research and substantial contributions, brain MRI segmentation is still a very challenging task to suit for range of diagnosis. It has been generalized to detect the brain tumor manually from medical image, which is a complex and tedious task. Even though several methods and encouraging results are obtained for brain tumor segmentation, sensitive and accurate detection are still a thought-provoking task due to the different shapes, locations and image intensities of different types of tumor. In this paper, a comprehensive survey on the brain tumor detection from MRI images is presented. The paper draws an attention towards the background of brain tumor and its distinction, discussion on brain tumor segmentation and algorithms classification. Finally, state-of-the-art-techniques for brain tumor detection are discussed based on the recent works and the review is concluded with open issues and challenges. This paper not only reviews, compares and consolidates the recent related works, but also admires the author's findings, solutions and discusses its usefulness towards tumor detection in medical applications. The uniqueness of this paper lies in the review of brain tumor segmentation and detection using MRI images.