Variational level set method with shape constraint and application to oedema cardiac magnetic resonance image

Kadir, K., Gao, H. , Payne, A., Soraghan, J. and Berry, C. (2011) Variational level set method with shape constraint and application to oedema cardiac magnetic resonance image. In: 17th International Conference on Digital Signal Processing (DSP), Corfu, 6-8 July 2011, (doi: 10.1109/ICDSP.2011.6004960)

Abstract

Quantification of oedema area after acute myocardial infarction (MI) is very important in clinical prognosis for differentiating the viable and death myocardial tissues. In order to quantify oedema region, the first step is to segment the myocardial wall accurately. This paper applies variational level set method with shape constraint to oedema cardiac magnetic resonance (CMR) images. Shape information of the myocardial wall is introduced into the variational level set formulation, and the performance of the automatic method is tested on T2 weighted CMR images from 8 patients, and compared with manual analysis from two clinical experts. Results show that the proposed automatic segmentation framework can segment left ventricle (LV) boundary with no significant difference compared to manual segmentation.

• Item Type: Conference Proceedings
• Status: Published
• Refereed: Yes
• Glasgow Author(s) Enlighten ID: Gao, Dr Hao
• Authors: Kadir, K., Gao, H., Payne, A., Soraghan, J., and Berry, C.
• College/School: College of Science and Engineering > School of Mathematics and Statistics > Mathematics