TY - GEN
T1 - Registration of brain CT images to an MRI template for the purpose of lesion-symptom mapping
AU - Kuijf, Hugo J.
AU - Biesbroek, J. Matthijs
AU - Viergever, Max A.
AU - Biessels, Geert Jan
AU - Vincken, Koen L.
PY - 2013/9/5
Y1 - 2013/9/5
N2 - Lesion-symptom mapping is a valuable tool for exploring the relation between brain structure and function. In order to perform lesion-symptom mapping, lesion delineations made on different brain CT images need to be transformed to a standardized coordinate system. The preferred choice for this is the MNI152 template image that is based on T1-weighted MR images. This requires a multi-modal registration procedure to transform lesion delineations for each CT image to the MNI152 template image. A two-step registration procedure was implemented, using lesion-masking and contrast stretching to correctly align the soft tissue of the CT image to the MNI152 template image. The results were used to transform the lesion delineations to the template. The quality of the registration was assessed by an expert human observer. Of the 86 CT images, the registration was highly successful in 71 cases (83%). Slight manual adjustments of the lesion delineations in the standard coordinate system were required to make unsuccessful cases suitable for a lesion-symptom mapping study.
AB - Lesion-symptom mapping is a valuable tool for exploring the relation between brain structure and function. In order to perform lesion-symptom mapping, lesion delineations made on different brain CT images need to be transformed to a standardized coordinate system. The preferred choice for this is the MNI152 template image that is based on T1-weighted MR images. This requires a multi-modal registration procedure to transform lesion delineations for each CT image to the MNI152 template image. A two-step registration procedure was implemented, using lesion-masking and contrast stretching to correctly align the soft tissue of the CT image to the MNI152 template image. The results were used to transform the lesion delineations to the template. The quality of the registration was assessed by an expert human observer. Of the 86 CT images, the registration was highly successful in 71 cases (83%). Slight manual adjustments of the lesion delineations in the standard coordinate system were required to make unsuccessful cases suitable for a lesion-symptom mapping study.
UR - https://www.scopus.com/pages/publications/84883304057
U2 - 10.1007/978-3-319-02126-3_12
DO - 10.1007/978-3-319-02126-3_12
M3 - Conference contribution
AN - SCOPUS:84883304057
SN - 9783319021256
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 119
EP - 128
BT - Multimodal Brain Image Analysis - Third International Workshop, MBIA 2013, Held in Conjunction with MICCAI 2013, Proceedings
T2 - 3rd International Workshop on Multimodal Brain Image Analysis, MBIA 2013, Held in Conjunction with the 16th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2013
Y2 - 22 September 2013 through 22 September 2013
ER -