Automatic Threshold Level Set Model Applied on MRI Image Segmentation of Brain Tissue

Authors

Ming Zhao, Department of computer science and technology, Harbin Institute of Technology Shenzhen Graduate School, Shenzhen , P.R.ChinaFollow
Hsiao-Yu Lin, Department of Applied Mathematics, National Chung Hsing University, Taichung, TaiwanFollow
Hsiao-Yu Lin, Department of Applied Mathematics, National Chung Hsing University, Taichung, TaiwanFollow
Chih-Hung Yang, Department of Applied Mathematics, National Chung Hsing University, Taichung, TaiwanFollow
Chih-Yu Hsu, Department of Information & Communication Engineering, ChaoYang University of Technology, Taichung, TaiwanFollow
Jeng-Shyang Pan, Department of computer science and technology, Harbin Institute of Technology Shenzhen Graduate School, Shenzhen , P.R.ChinaFollow

Author Country (or Countries)

P.R.China

Abstract

In the paper, a mathematical proof is given for a different mean of Chan and Vese model. Based on the proof, an image segmentation method called Automatic Threshold Level Set Without Edge was developed for the extraction of tissues in brain MR images. Thresholds are defined to find the boundary of tissues in the brain and they can be automatically obtained by Fuzzy C Mean algorithm. A similarity index (SI) is used for quantitative evaluation of the segmentation results. By testing MRI brain slice images and comparing to the ground truth of tissue segmentation, the mean and the variance of SI are 0.90311 and 0.042049. The experimental results demonstrate our method can automatically and accurately segment the regions of tissues in brain.

Recommended Citation

Zhao, Ming; Lin, Hsiao-Yu; Lin, Hsiao-Yu; Yang, Chih-Hung; Hsu, Chih-Yu; and Pan, Jeng-Shyang (2015) "Automatic Threshold Level Set Model Applied on MRI Image Segmentation of Brain Tissue," Applied Mathematics & Information Sciences: Vol. 09: Iss. 4, Article 37.
Available at: https://digitalcommons.aaru.edu.jo/amis/vol09/iss4/37