AbstractAs the size of the neuroimaging cohorts being increased to address key questions in the field of cognitive neuroscience, cognitive aging, and neurodegenerative diseases, the accuracy of the spatial normalization as an essential pre-processing step becomes extremely important in the neuroimaging processing pipeline. Existing spatial normalization methods have poor accuracy particularly when dealing with the highly convoluted human cerebral cortex and when brain morphology is severely altered (e.g. clinical and aging populations). To address this shortcoming, we propose to implement and evaluate a novel landmark-guided region-based spatial normalization technique that takes advantage of the existing surface-based human brain parcellation to automatically identify and match regional landmarks. To simplify the non-linear whole brain registration, the identified landmarks of each region and their counterparts are registered independently with large diffeomorphic (topology preserving) deformation via geodesic shooting. The regional diffeomorphic warping fields were combined by an inverse distance weighted interpolation technique to have a smooth global warping field for the whole brain. To ensure that the final warping field is diffeomorphic, we used simultaneously forward and reverse maps with certain symmetric constraints to yield bijectivity. We have evaluated our proposed method using both simulated and real (structural and functional) human brain images. Our evaluation shows that our method can enhance structural correspondence up to around 86%, a 67% improvement compared to the existing state-of-the-art method. Such improvement also increases the sensitivity and specificity of the functional imaging studies by about 17%, reducing the required number of subjects and subsequent costs. We conclude that our proposed method can effectively substitute existing substandard spatial normalization methods to deal with the demand of large cohorts and the need for investigating clinical and aging populations.
Spatial normalization of brain images and beyond