Segmentation And Shape Analysis of Corpus Callosum (CC) in Alzheimer Brain MR Images Using Improved Variational Level Set Method and Phase Congruency Map
-
1
Anna University, Electronics and Communication Engineering, India
-
2
INDIAN INSTITUTE OF TECHNOLOGY MADRAS, DEPARTMENT OF APPLIED MECHANICS, India
Alzheimer’s Disease (AD) is a neurological disorder and common form dementia that causes memory impairment problems in patients. AD leads to atrophy of gray and white matter structures and results in tissue loss. World Alzheimer report informs that an estimated population of 35.6 million people around the world suffers from this neurological disorder. Magnetic Resonance (MR) imaging is an useful non-invasive imaging modality which is highly useful in reflecting the brain pathology and different stages of AD. Corpus Callosum (CC) is the largest white matter structure that connects the left and right cerebral hemispheres. It is also responsible for integrating the sensory, motor and cognitive functions of brain. Manifestation of AD also results in the shrinkage of CC.
Segmentation and feature extraction are essential in the shape analysis of CC. Level set methods are dynamic curves or surfaces that undergo iterative evolution to track the complex topological structures. The diffusion rate equation used in the level set function avoids the complex re-initialization problem and provides numerical stability. Local phase information extracted form an image provides maximum information in representing the lines and edges than the intensity values. Phase Congruency (PC) map obtained from an image can be used as an edge indicator in the level set evolution.
In this work, T1-weighted sagittal view MR images are obtained from open access series of imaging studies, a public domain database. Image acquisition has been done using 1.5-T Vision scanner in both the men and women subjects. Images (Normal=20 and AD=20) are subjected to Phase Based Improved Variational Level Set Method (PBIVLSM) to segment CC. PC map extracted from the image is used as edge stopping criterion in the level set function. Geometric features are extracted from the segmented CC and analyzed graphically.
Results show that PBIVLSM is able to segment CC in both the normal and AD subjects. The edge map obtained using PC is found to have continuous and high contrast edges in all the images. The extracted geometric features such as major axis and minor axis shows appreciable demarcation between the normal and AD images with the percentage difference of 4.85% and 11.19% respectively. Since shape analysis of CC is essential in the diagnosis of AD, this study seems to be clinically useful.
Keywords:
Alzheimer Disease,
segmentation,
Level set method,
Corpus Callosum,
image processing
Conference:
Neuroinformatics 2014, Leiden, Netherlands, 25 Aug - 27 Aug, 2014.
Presentation Type:
Poster, to be considered for oral presentation
Topic:
Neuroimaging
Citation:
Ramaniharan
AK,
Manoharan
SC and
Swaminathan
R
(2014). Segmentation And Shape Analysis of Corpus Callosum (CC) in Alzheimer Brain MR Images Using Improved Variational Level Set Method and Phase Congruency Map.
Front. Neuroinform.
Conference Abstract:
Neuroinformatics 2014.
doi: 10.3389/conf.fninf.2014.18.00065
Copyright:
The abstracts in this collection have not been subject to any Frontiers peer review or checks, and are not endorsed by Frontiers.
They are made available through the Frontiers publishing platform as a service to conference organizers and presenters.
The copyright in the individual abstracts is owned by the author of each abstract or his/her employer unless otherwise stated.
Each abstract, as well as the collection of abstracts, are published under a Creative Commons CC-BY 4.0 (attribution) licence (https://creativecommons.org/licenses/by/4.0/) and may thus be reproduced, translated, adapted and be the subject of derivative works provided the authors and Frontiers are attributed.
For Frontiers’ terms and conditions please see https://www.frontiersin.org/legal/terms-and-conditions.
Received:
27 Apr 2014;
Published Online:
04 Jun 2014.
*
Correspondence:
Mr. Anandh K Ramaniharan, Anna University, Electronics and Communication Engineering, Chennai, Tamilnadu, 600025, India, anandhmurali@gmail.com