A method to investigate the size and shape variation of the lateral ventricles with age

Various structures within the brain are associated with aging. The ventricles are known to enlarge with age as the cerebrospinal fluid (CSF) increases. The objective of this study is to develop a semi-automated method for acquiring landmark data representing brain structures, such as the lateral ventricles, to map size and shape variation with age. The methodology developed is a multiple-step processing of magnetic resonance images (MRI) in a brain atlas. The MRI data was collected for various ages (20-100 years old). Group-specific templates were created by decade. Each template was normalized to an atlas in Montreal Neurological Institute (MNI) space using a high-dimensional non-linear transformation procedure. A manually segmented label map of the lateral ventricles in MNI space was applied to segment the structure and establish homologous landmarks surrounding the external axial surface of the structure. The homologous landmarks were transformed back to patient space. A Generalized Procrustes Analysis (GPA) was completed for the homologous landmarks to determine the size and shape differences. The coordinate locations were regressed onto age, which generated a model predicting the landmark location for each age. A significant non-linear increase was observed in the anterior/posterior, medial/lateral, and cranial/caudal directions along the lateral ventricle, with an exponential increase after age 40 (P<0.0001). This novel method of semi-automated landmark identification will allow for the creation of mapping functions to describe age-associated changes of the brain. This work is important and may lead to a better understanding of how brain morphology is correlated to the biomechanics of traumatic brain injury.