Abstract #0822

k-Means Segmentation of Kidney Cortex & Medulla for BOLD Images

Yin Huang¹, Nathan Hanson¹, Elizabeth Sadowski², David Niles¹, Nathan Artz¹, Arjang Djamali³, Thomas Grist^1,2, Sean Fain^1,2

¹Medical Physics, University of Wisconsin Madison, Madison, WI, United States; ²Radiology, University of Wisconsin Madison, Madison, WI, United States; ³Nephrology, University of Wisconsin Madison, Madison, WI, United States

The k-means segmentation method was implemented to semi-automatically segment kidney cortex and medulla for MR BOLD images of 6 subjects. By acquiring an extra T1 weighted image, k-means segmentation was performed based on two kidney feature values -- T1 and T2* weighted signal intensities. Manual segmentation results on the same subjects were used as reference and sensitivity and specificity measures were calculated to evaluate the quality of the k-means segmentation.