Esha Datta1, Nico Papinutto1, Regina Schlaeger1, Julio Carballido-Gamio1, Alyssa Zhu1, and Roland G Henry1
This study demonstrates the accuracy and reliability
of a new method for automatic segmentation of spinal cord gray matter in 2D
PSIR images at 3T of the C2-C3 spinal cord level in healthy controls. This method
deforms an initial contour, based on registration from a template, using an
active contours algorithm to ultimately obtain the final gray matter
segmentation. When comparing the
automatic segmentations with manual segmentations in 12 subjects, the Dice
coefficients ranged from .82 to .93, with an average of .88. In 8 additional subjects that were scanned
twice, the percent changes ranged from 1% to 6%.
