The anatomical structures of interest that we study within the brain are complex and need to be characterized in a consistent, reliable manner. We at UCNIA are developing methods for representing and analyzing shapes from image data to facilitate the study of anatomical changes. This also includes the development of shape alignment procedures, spatial correspondence estimation, and statistical analysis of shape parameters.
Shape Analysis of Neuroanatomical Structures
We have developed a new method for constructing statistical representations of ensembles of similar shapes that uses particle systems to represent surfaces non parametrically and optimally sample surface point correspondences. We used this method to generate models for two clinical datasets: normal vs. Autistic neurological development. Hypothesis testing on these models using a non parametric permutation test of the Hotelling T-squared metric (including false-discovery-rate (FDR) correction) reveals significant group differences. Colormap indicates the magnitude and direction of the linear discriminant.
Infant MRI Head Coil Design
Improved MRI methodology for infant imaging: We study head/brain growth and create statistical models of neonates, 6mo, 1yr, 2yr and 4yr. Based on these models, the MGH group creates new parallel coils for the scanner. We then get these parallel images and combine them back with new signal processing.
We have developed a new method for constructing statistical representations of ensembles of similar shapes that uses particle systems to represent surfaces non parametrically and optimally sample surface point correspondences. We used this method to generate models for two clinical datasets: normal vs. Autistic neurological development. Hypothesis testing on these models using a non parametric permutation test of the Hotelling T-squared metric (including false-discovery-rate (FDR) correction) reveals significant group differences. Colormap indicates the magnitude and direction of the linear discriminant.
Improved MRI methodology for infant imaging: We study head/brain growth and create statistical models of neonates, 6mo, 1yr, 2yr and 4yr. Based on these models, the MGH group creates new parallel coils for the scanner. We then get these parallel images and combine them back with new signal processing.