Automated brain structure segmentation based on atlas registration and appearance models
Research output: Contribution to journal › Journal article › Research › peer-review
Standard
Automated brain structure segmentation based on atlas registration and appearance models. / van der Lijn, Fedde; de Bruijne, Marleen; Klein, Stefan; den Heijer, Tom; Hoogendam, Yoo. Y.; van der Lugt, Aad; Breteler, Monique M. B.; Niessen, Wiro J.
In: IEEE Transactions on Medical Imaging, Vol. 31, No. 2, 2012, p. 276-286.Research output: Contribution to journal › Journal article › Research › peer-review
Harvard
APA
Vancouver
Author
Bibtex
}
RIS
TY - JOUR
T1 - Automated brain structure segmentation based on atlas registration and appearance models
AU - van der Lijn, Fedde
AU - de Bruijne, Marleen
AU - Klein, Stefan
AU - den Heijer, Tom
AU - Hoogendam, Yoo. Y.
AU - van der Lugt, Aad
AU - Breteler, Monique M. B.
AU - Niessen, Wiro J.
PY - 2012
Y1 - 2012
N2 - Accurate automated brain structure segmentationmethods facilitate the analysis of large-scale neuroimaging studies.This work describes a novel method for brain structure segmentation in magnetic resonance images that combines informationabout a structure’s location and appearance. The spatial modelis implemented by registering multiple atlas images to the targetimage and creating a spatial probability map. The structure’s appearance is modeled by a classi¿er based on Gaussian scale-spacefeatures. These components are combined with a regularizationterm in a Bayesian framework that is globally optimized usinggraph cuts. The incorporation of the appearance model enables themethod to segment structures with complex intensity distributionsand increases its robustness against errors in the spatial model.The method is tested in cross-validation experiments on twodatasets acquired with different magnetic resonance sequences,in which the hippocampus and cerebellum were segmented byan expert. Furthermore, the method is compared to two othersegmentation techniques that were applied to the same data. Results show that the atlas- and appearance-based method producesaccurate results with mean Dice similarity indices of 0.95 for thecerebellum, and 0.87 for the hippocampus. This was comparable toor better than the other methods, whereas the proposed techniqueis more widely applicable and robust.
AB - Accurate automated brain structure segmentationmethods facilitate the analysis of large-scale neuroimaging studies.This work describes a novel method for brain structure segmentation in magnetic resonance images that combines informationabout a structure’s location and appearance. The spatial modelis implemented by registering multiple atlas images to the targetimage and creating a spatial probability map. The structure’s appearance is modeled by a classi¿er based on Gaussian scale-spacefeatures. These components are combined with a regularizationterm in a Bayesian framework that is globally optimized usinggraph cuts. The incorporation of the appearance model enables themethod to segment structures with complex intensity distributionsand increases its robustness against errors in the spatial model.The method is tested in cross-validation experiments on twodatasets acquired with different magnetic resonance sequences,in which the hippocampus and cerebellum were segmented byan expert. Furthermore, the method is compared to two othersegmentation techniques that were applied to the same data. Results show that the atlas- and appearance-based method producesaccurate results with mean Dice similarity indices of 0.95 for thecerebellum, and 0.87 for the hippocampus. This was comparable toor better than the other methods, whereas the proposed techniqueis more widely applicable and robust.
U2 - 10.1109/TMI.2011.2168420
DO - 10.1109/TMI.2011.2168420
M3 - Journal article
C2 - 21937346
VL - 31
SP - 276
EP - 286
JO - I E E E Transactions on Medical Imaging
JF - I E E E Transactions on Medical Imaging
SN - 0278-0062
IS - 2
ER -
ID: 33950222