SEGMENTATION OF MRI IMAGES VIA MATHEMATICAL MORPHOLOGY
Tanzila Rahman, Jianqiang Gao, Srinivas Nowduri · International Journal of Applied Mathematics and Machine Learning · 2017
Magnetic resonance imaging (MRI) has become an emergent research area in medical field in finding the exact size, shape, position, appearance, and location of tumor.In this paper, we develop and implement a new method based on mathematical morphology, which can segment brain tumor from MRI images, usable for early detection of intracranial tumors.The method consists of two stages to detect and segment a brain tumor.Firstly, quality of input image is enhanced using morphological operators.Secondly, appropriate threshold value is used to segment brain tumor from MRI images.Experimental result confirms the superiority of the proposed approach in terms of performance preservation.