The mirror method of assessing segmentation quality in atlas label propagation

Atlas-based brain image segmentation quality is difficult to assess in the absence of reference target segmentations. We propose a measure of segmentation success based on transforming the atlas label twice: once by registering the atlas to the target and a second time by registering the target to t...

Full description

Saved in:
Bibliographic Details
Main Authors: Heckemann, R.A., Hammers, A., Aljabar, P., Rueckert, D., Hajnal, J.V.
Format: Conference Proceeding
Language:English
Subjects:
Alzheimer's disease
Brain
Data mining
Educational institutions
Engines
Image segmentation
Magnetic resonance imaging
Mirrors
Neuroscience
Time measurement
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Atlas-based brain image segmentation quality is difficult to assess in the absence of reference target segmentations. We propose a measure of segmentation success based on transforming the atlas label twice: once by registering the atlas to the target and a second time by registering the target to the atlas. Each registration represents transformations by free-form deformations. The overlap between the twice-transformed label and the original (dasiamirror overlappsila) correlates with the forward overlap (between the once-transformed label and a target reference), especially for subcortical structures. Using mirror overlap as an atlas selection criterion results in improved segmentations versus random selection.
ISSN:1945-7928
1945-8452
DOI:10.1109/ISBI.2009.5193272