Open Access Open Access  Restricted Access Subscription Access
Open Access Open Access Open Access  Restricted Access Restricted Access Subscription Access

Improved Hybrid Segmentation of Brain MRI Tissue and Tumor Using Statistical Features


Affiliations
1 Department of Electronics and Communication Engineering, PSG College of Technology, Coimbatore, India
2 Department of Bio-Medical Engineering, PSG College of Technology, Coimbatore, India
     

   Subscribe/Renew Journal


Medical image segmentation is the most essential and crucial process in order to facilitate the characterization and visualization of the structure of interest in medical images. Relevant application in neuroradiology is the segmentation of MRI data sets of the human brain into the structure classes gray matter, white matter and cerebrospinal fluid (CSF) and tumor. In this paper, brain image segmentation algorithms such as Fuzzy C means (FCM) segmentation and Kohonen means(K means) segmentation were implemented. In addition to this, new hybrid segmentation technique, namely, Fuzzy Kohonen means of image segmentation based on statistical feature clustering is proposed and implemented along with standard pixel value clustering method. The clustered segmented tissue images are compared with the Ground truth and its performance metric is also found. It is found that the feature based hybrid segmentation gives improved performance metric and improved classification accuracy rather than pixel based segmentation.

Keywords

K-Means, Fuzzy C-Means, Fuzzy Kohonen Means Clustering, Distance of Clustering, Von Dongen Index.
Subscription Login to verify subscription
User
Notifications
Font Size

Abstract Views: 255

PDF Views: 0




  • Improved Hybrid Segmentation of Brain MRI Tissue and Tumor Using Statistical Features

Abstract Views: 255  |  PDF Views: 0

Authors

S. Allin Christe
Department of Electronics and Communication Engineering, PSG College of Technology, Coimbatore, India
K. Malathy
Department of Electronics and Communication Engineering, PSG College of Technology, Coimbatore, India
A. Kandaswamy
Department of Bio-Medical Engineering, PSG College of Technology, Coimbatore, India

Abstract


Medical image segmentation is the most essential and crucial process in order to facilitate the characterization and visualization of the structure of interest in medical images. Relevant application in neuroradiology is the segmentation of MRI data sets of the human brain into the structure classes gray matter, white matter and cerebrospinal fluid (CSF) and tumor. In this paper, brain image segmentation algorithms such as Fuzzy C means (FCM) segmentation and Kohonen means(K means) segmentation were implemented. In addition to this, new hybrid segmentation technique, namely, Fuzzy Kohonen means of image segmentation based on statistical feature clustering is proposed and implemented along with standard pixel value clustering method. The clustered segmented tissue images are compared with the Ground truth and its performance metric is also found. It is found that the feature based hybrid segmentation gives improved performance metric and improved classification accuracy rather than pixel based segmentation.

Keywords


K-Means, Fuzzy C-Means, Fuzzy Kohonen Means Clustering, Distance of Clustering, Von Dongen Index.