Alzheimer's disease (AD) is an irreversible neurodegenerative brain disease distinguished by progressive impairment of memory and decline in cognitive abilities. The hippocampus is widely recognized to play essential roles in forming and gradually transferring information from short-term memory into long-term memory, and it is involved in the onset of the neuropathological pathways of the brain to suffer neuron loss in the rise of AD. Thus, hippocampal information obtained from magnetic resonance imaging (MRI) scans have been established as crucial AD biomarkers. The hippocampus is composed of multiple subfields, and the neuron loss is not uniformly distributed on the whole hippocampus. However, this critical subfield information is not addressed by the existing surface-based morphometry (SBM) and voxel-based morphometry (VBM) studies. Due to the size, complexity, heterogeneity, and folding anatomy of the hippocampus, acquiring volumetric and morphometric measures of hippocampal subfields usually presents not only technical challenges in quantitative neuroimaging but also analytical challenges. To address these challenges and deeply understand the relationships between hippocampal shape changes and brain disorders, especially to examine the degeneration of hippocampal subfields, this thesis focuses on constructing a hippocampal subfield morphometric analysis framework including the following aspects: 1) hippocampal subfield segmentation; 2) 3D shape modeling; 3) feature formulation; 4) diffeomorphic surface registration; 5) surface shape reconstruction; and 6) association analytics. The goals include developing accurate hippocampal subfield guided registration methods, extracting useful features and identifying significant subfields on the hippocampus that are highly related to cognitive disabilities, and using such information to assist early detection of AD.