Brain tumor classification for MR imaging using support vector machine

Abstract

Nowadays, brain tumor segmentation is most challenging task in the field of medical image processing. Manual segmentation of these images by the domain experts is a time-consuming process. There is numerous automatic algorithm for MRI image segmentation and classification but still, they need to develop an efficient and fast algorithm. Accurate segmentation of tumor helps in early diagnosis of the tumor. This paper presents an efficient approach for brain tumor for MRI image using support vector machine (SVM) and Otsu thresholding. We tested the performance of fuzzy c-means clustering, k-means, and KIFCM (integration of k-means and fuzzy c-means). Our proposed method outperforms the existed algorithm in terms of accuracy and execution time.