摘要
MRI brings information regarding disease effect on brain structures in several pathologies. The hippocampus has been shown to be damaged in early Alzheimer's disease. Computing hippocampal volume from MRI datasets has led to a wide range of methods, with various automation levels. We proposed a region deformation method for the simultaneous segmentation of the hippocampus and the amygdala. This method is driven by prior information from two sources: anatomical priors. derived from automatically retrieved landmarks, and probabilistic priors. derived from in atlas built from the manual segmentations from 16 young controls. This method was evaluated on these 16 controls, as well as on patients with epilepsy and Alzheimers disease. It has shown to be fast and robust. Moreover, this method has been proven to be useful in classifying patients with Alzheimer's disease, using two kinds of features: either the hippocampal volume or features derived from a model built with spherical harmonics from the hippocampal shape retrieved by the segmentation. In this paper, we will also introduce the perspective for hippocampal segmentation brought by ultra-high field MRI.
学科与关键词
- Atlas-based hippocampus segmentation in Alzheimer's disease and mild cognitive impairmentCarmichael, OT*; Aizenstein, HA; Davis, SW; Becker, JT; Thompson, PM; Meltzer, CC; Liu, YXNeuroImage, 2005, 27(4): 979-990.被引 163
- Automated Methods for Hippocampus Segmentation: the Evolution and a Review of the State of the ArtDill, Vanderson*; Franco, Alexandre Rosa; Pinho, Marcio SarrogliaNEUROINFORMATICS, 2015, 13(2): 133-150.被引 65
- Multidimensional classification of hippocampal shape features discriminates Alzheimer's disease and mild cognitive impairment from normal agingGerardin, Emilie; Chetelat, Gael; Chupin, Marie; Cuingnet, Remi; Desgranges, Beatrice; Kim, Ho-Sung; Niethammer, Marc; Dubois, Bruno; Lehericy, Stephane; Garnero, Line; Eustache, Francis; Colliot, Olivier*NeuroImage, 2009, 47(4): 1476-1486.被引 312
