A Novel Technique for identification of tumor region in MR Brain Image

• Published in: 2019 3rd International conference on Electronics, Communication and Aerospace Technology (ICECA) pp. 1061 - 1066
• Main Authors: Alagarsamy, Saravanan, Kamatchi, Kartheeban, Govindaraj, Vishnuvarthanan
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.06.2019
• Subjects: Aerospace electronics; Brain; Clustering algorithms; Conferences; Fruit fly Algorithm; Image segmentation; Interval Type-2 Fuzzy C-Means Clustering (IT2FCM); Magnetic Resonance Imaging; Mathematical model; Tumor identification; Tumors
• DOI: 10.1109/ICECA.2019.8822188

Detection of brain tumor is a vital role in field of medical. Brain tumor with low grade and high grade will threaten the life of human being, Diagnosis of brain tumors are identified in the early stage can increase the survival rate of patients. Due to the multifarious structure in human brain there is a need of novel technique for better identification of tumors in the early stage. The suggested fruit fly based Interval type fuzzy c-means clustering techniques performs outstanding segmentation result for tumors with low grade and high grade. The competence of proposed techniques is tested with BRATS-SICAS (2015) dataset. Demarcated result obtained by the fruit fly based Interval type fuzzy c-means clustering algorithm deliberates better identification of tumors and compared with the ground truth obtained from BRATS-SICAS (2015) dataset. The suggested techniques deliberate an impressive sensitivity value of 98.47 % and dice score value of 96.23 % respectively which is much better than existing techniques used for segmentation. These novel techniques will aid the clinicians for accurate identification of tumors in the field of medical.