• C.Sushma
• Dr. C.Subhas

Tumor is a rapid uncontrolled growth of cells. Now-a-days number of people suffering from brain tumor has increased rapidly. There are several types of tumors in which some are cancerous and some are non-cancerous. Tumor stage generally depends on the size and its growth rate. If the features of tumor in the brain are properly detected at an early stage, it can be treated or removed by the doctor effectively. Thus the life of the patient can be saved by proper detection of tumor cells in the brain. There are several fundamental segmentation approaches for the tumor cells detection like K-Means clustering, Threshold based segmentation, Morphological operations, Watershed segmentation, Edge detection methods etc. These methods are used in various sequences in the literature for segmenting out the tumor region. In this paper, we propose novel segmentation method by sequencing fundamental segmentation approaches effectively that segments out the tumor with better PSNR than the existing methods. Further, this method can segment tumors having low contrast in the MRI images. We superimpose this segmented region over the original MRI image for better identification of its structure. Finally we find various region parameters of the tumor like its Area, Diameter, Perimeter and Centroid.

Clustering, K-Means, Morphological, MRI Image, Segmentation, Watershed Transform.

"Novel Segmentation Method for Detection of Tumors in Human Brain", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.5, Issue 7, page no.882-891, July-2018, Available :http://www.jetir.org/papers/JETIR1807143.pdf

"Novel Segmentation Method for Detection of Tumors in Human Brain", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.5, Issue 7, page no. pp882-891, July-2018, Available at : http://www.jetir.org/papers/JETIR1807143.pdf