University of Heidelberg
Faculty of Medicine Mannheim
University Hospital Mannheim
These pages are still under constructions and will be available soon! Please check again later!

If you have questions concerning a specific publication please use this form with subject 'information about publications' and giving the full citation in the message body.

Collaborate Research Projects
Home > Publications > Abstract >

Wavelet-based segmentation of renal compartments in DCE-MRI of human kidney: Initial results in patients and healthy volunteers

S. Li, F. Zöllner, A. Merrem, Y. Peng, J. Roervik, A. Lundervold and L. Schad

Comput Med Imag Grap, 36 (2), pp.108-118

Renal diseases can lead to kidney failure that requires life-long dialysis or renal transplantation. Early detection and treatment can prevent progression towards end stage renal disease. MRI has evolved into a standard examination for the assessment of the renal morphology and function. We propose a wavelet-based clustering to group the voxel time courses and thereby, to segment the renal compartments. This approach comprises (1) a nonparametric, discrete wavelet transform of the voxel time course, (2) thresholding of the wavelet coefficients using Stein's Unbiased Risk estimator, and (3) k-means clustering of the wavelet coefficients to segment the kidneys. Our method was applied to 3D dynamic contrast enhanced (DCE-) MRI data sets of human kidney in four healthy volunteers and three patients. On average, the renal cortex in the healthy volunteers could be segmented at 88\%, the medulla at 91\%, and the pelvis at 98\% accuracy. In the patient data, with aberrant voxel time courses, the segmentation was also feasible with good results for the kidney compartments. In conclusion wavelet based clustering of DCE-MRI of kidney is feasible and a valuable tool towards automated perfusion and glomerular filtration rate quantification.

Contact: Dr. Frank Zöllner last modified: 21.09.2020
to top of page