Journal of Northeastern University:Natural Science ›› 2013, Vol. 34 ›› Issue (1): 52-56.DOI: -

• 论著 • Previous Articles     Next Articles

Effective Automated Kidney Locating and Segmentation Method Based on Multi-phase MR Images

LI Hong1, PAN Xiao-guang2, SUN Hao-ran3, KANG Yan1   

  1. 1. School of Sino-Dutch Biomedical & Information Engineering, Northeastern University, Shenyang 110819, China; 2. School of Information Science & Engineering, Northeastern University, Shenyang 110819, China; 3. Department of Radiology, Tianjin Medical University General Hospital, Tianjin 300052, China.
  • Received:2012-07-10 Revised:2012-07-10 Online:2013-01-15 Published:2013-01-26
  • Contact: KANG Yan
  • About author:-
  • Supported by:
    -

Abstract: Automatically locating the kidney region and segmenting the kidneys in MR images are the first and very important steps for the whole renal function analysis flow. A novel method of automatically locating the kidney based on multi-phase MR images was proposed. First, 3D registration and subtraction were employed; then max intensity projection along different directions was performed to get projection images; afterwards, the MID(maximum intensity distribution) curve analysis was used to obtain the kidney region boundary. Inside the kidney region, the level set method was employed to segment each kidney on the parenchymal phase images. Finally, the segmentation result was mapped onto all the other phase images. Experimental results showed that the location results are very good, and the segmentation performed within the location region has higher efficiency and better accuracy.

Key words: MRI(magnetic resonance imaging), kidney segmentation, maximum intensity distribution curve, level set, glomerular filtration rate (GFR)

CLC Number: