Journal of Northeastern University Natural Science ›› 2018, Vol. 39 ›› Issue (2): 186-190.DOI: 10.12068/j.issn.1005-3026.2018.02.008

• Information & Control • Previous Articles     Next Articles

MRI Brain Tissue Automatic Extraction Algorithm Based on Modified BET

YANG Jin-zhu1,2, LU Lin1,2, CAO Peng1,2, ZHAO Da-zhe1,2   

  1. 1.School of Computer Science & Engineering, Northeastern University, Shenyang 110169, China; 2.Key Laboratory of Medical Image Computing, Ministry of Education, Northeastern University, Shenyang 110169, China.
  • Received:2016-09-04 Revised:2016-09-04 Online:2018-02-15 Published:2018-02-09
  • Contact: ZHAO Da-zhe
  • About author:-
  • Supported by:
    -

Abstract: Based on modified brain extraction tool(BET), an automatic extraction algorithm is proposed for brain MRI(magnetic resonance imaging). Firstly, the algorithm combining image gradient information can estimate the center of gravity(COG) of brain more accurately; secondly, it builds a new brain surface deformation force. Edge force is introduced into the expansion force perpendicular to the tangent of brain surface, which could suppress the brain tissue leaking and over-segmentation problem. Experiments are conducted on the algorithm proposed to automatically extract brain tissue of MRI brain images, and the results show the more accurate extraction of brain tissue, especially at brain tissue edges, compared with the BET method.

Key words: brain extraction, MRI, BET(brain extraction tool), COG(center of gravity) of brain, edge force

CLC Number: