Abstract
AbstractThis technical report presents the Automatic Temporal Registration Algorithm (ATRA) for symmetric rigid-body affine registration of longitudinal T1-weighted three-dimensional magnetic resonance imaging (MRI) scans of the human brain. This is a fundamental processing step in computational neuroimaging. The notion of leave-one-out consistent (LOOC) landmarks with respect to a supervised landmark detection algorithm is introduced. An automatic algorithm is presented for identification of LOOC landmarks on MRI scans. Using this technique, multiple sets of LOOC landmarks (around 150) are identified on each of the volumes being registered. Then, a Generalized Orthogonal Procrustes Analysis of the identified landmarks is used to find a rigid-body transformation of each volume into a common space where the transformed volumes are precisely aligned. In addition, a new approach is introduced for quantitative assessment of registration accuracy in the absence of a gold standard. Qualitative and quantitative evaluations of ATRA registration accuracy are performed using 2012 volumes from 503 subjects (4 longitudinal volumes/subject) from the Alzheimer’s Disease Neuroimaging Initiative database, and on a further 120 volumes acquired from 3 normal subjects (40 longitudinal volumes/subject). The algorithm is symmetric, in the sense that any permutation of the input volumes does not change the resulting transformation matrices, and unbiased, in the sense that all volumes undergo one and only one interpolation operation, which precisely aligns them in a common space. There is no interpolation bias and no reference volume. All volumes are treated exactly the same. The algorithm is fast and highly accurate. The software is publicly available.
Publisher
Cold Spring Harbor Laboratory
Cited by
5 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献