Skip to main content
. Author manuscript; available in PMC: 2020 Aug 2.
Published in final edited form as: IEEE Trans Med Imaging. 2015 Mar 20;34(9):1901–13. doi: 10.1109/TMI.2015.2415453

Table II.

Left: Example Foreground (Yellow) and Background (Red) Constraints for the Segmentation of the Ventricles [42]. Right: Evaluating the Influence of Different PSFs on the Dice Coefficient for Semi-Automatic Segmentation Compared to a Ground Truth. We Evaluate the Accuracy of Ventricular, White-Matter, and Cortical Segmentation of the BrainWeb Dataset after Applying Simulated Motion Corruption and Reconstruction using each PSF.

graphic file with name EMS87724-i001.jpg
PSF ventricles white-matter cortex
PSFtrunc 0.912 0.845 0.829
PSFGauss 0.916 0.853 0.840
PSFMRI 0.918 0.867 0.851