Comparative Analysis of Fuzzy C-Mean and Level Set Segmentation Method for Brain Tumor Classes
Image analysis is very important for many methodologies to process and identify the information of biomedical images. The segmentation technique is used to detect and characterize the biomedical images. The understanding of patient disease problem is an important task for radiologists. In this paper, we have presented the Fuzzy C-Mean (FCM) method for an objective function with fitting value and the Level Set Method (LSM) used for initial seeds contour for initialization. In this work, we have compared the abilities of both the algorithms with respect to the time complexity and iterations required for successful segmentation. The experimental work consisted of total 13 magnetic resonance (MR) brain tumor images: (Astrocytoma (3), ganglioglioma (2), glioblastoma (3), glioma (1) and meningioma (4)). The experimental result shows that the Level Set Method (LSM) required on average 56% less time than Fuzzy C-Mean (FCM) however required twice as much. The iterations of complete tumor segmentation in the level set method (LSM) on an average were 10–20% lower than a Fuzzy C-Mean (FCM).
Keywords: Magnetic Resonance Imaging (MRI), brain tumour, segmentation, fuzzy c-mean (FCM), level set method (LSM)
Cite this Article
Nikita Singh, Naveen Choudhary. Comparative Analysis of Fuzzy C-Mean and Level Set Segmentation Method for Brain Tumor Classes. Journal of Open Source Developments. 2017; 4(3): 1–8p.
- There are currently no refbacks.