AShortSurveyonMRIBrainDetection_v3.docxVIP

A Short Survey on MRI Brain DetectionSi-Yuan Lu1*, Xing-Xing Zhou1, Guang-Shuai Zhang1, Ling Wei2,(1 School of Computer Science and Technology, Nanjing Normal University, Nanjing, Jiangsu, 210023, China)(2 School of Electronic Information Electrical Engineering, Shanghai Jiaotong University, Shanghai 200030, China)Abstract. Automated and accurate classification of MR brain images is of importance for the analysis. Recently researchers have proposed many different and innovative methods to improve upon this technology. This review is focused on discussing the strength and limitations of earlier proposed classification techniques. Besides analyzing these literature, the paper also provides a critical evaluation of the surveyed literature which reveals new hotspots of research.Keywords: magnetic resonance imaging; classification; computer-aided diagnosisIntroductionMagnetic resonance imaging (MRI) is capable of producing high quality images of the anatomical structures of the human body (especially the brain), while it is a low-risk, fast, non-invasive imaging technique. The rich information it provides is of great help for clinical diagnosis and biomedical research [1]. However, because of this huge information repository associated with MRIs, employing manual methods to interpret each image is tedious and time consuming, and it is of necessity to develop automated diagnosis tools to draw quicker and easier inferences from the MR images. These automated tools can benefit the medical personnel in diagnosis, prognosis, pre-surgical and postsurgical procedures, etc. [2].Sophisticated signal/image processing techniques and some computational intelligent techniques are applied to develop diagnostic tools. In one possible methodology, the problem of automatic classification of normal/pathological subjects, based on brain MRIs, can be configured as a three-stage problem: 1) feature extraction; 2) feature reduction; 3) classification [3].The main contribution of this study i