Development of MRI template library for nucleus accumbens

Segmentation of medical images is one of the most critical steps in many clinical applications. In brain MRI analysis, image segmentation is commonly used for measuring and visualizing the brain’s anatomical structures, for analyzing brain changes. In this project a manual template library of Nucleus Accumbens is created by manually segmenting the MRI images. Nucleus Accumbens is situated in the basal ganglia part of brain and is major component of ventral striatum. A protocol is generated in order to recognize the structure and boundaries of Nucleus Accumbens in the MRI images. The FreeSurfer images are used and manually segmented. The FS-LDDMM algorithm is then used for mapping the segmentations over the rest of the subjects. This template library is a heterogeneous combination of healthy and diseased Nucleus Accumbens structures which were used to train FS+LDDMM and therefore, enable FS+LDDMM to identify Nucleus Accumbens over a diverse range of structural variation which would be present for different neurodegenerative diseases. Cross-validations are done in order to compare the efficiency of the FreeSurfer segmented images and the segmentations performed by FS-LDDMM method. The manual template library generated will be used with a multi- atlas segmentation algorithm like FS-LDDMM to segment large number of images acquired from patients with neurodegenerative disorders so as to get better understanding of the diseases.