3-DIMENSIONAL HIPPOCAMPAL MR MORPHOMETRY WITH HIGH-DIMENSIONAL TRANSFORMATION OF A NEUROANATOMICAL ATLAS

PURPOSE: To test automated three-dimensional magnetic resonance (MR) i maging morphometry of the human hippocampus, to determine the potentia l gain in precision compared with conventional manual morphometry. MAT ERIAL AND METHODS: A canonical three-dimensional MR image atlas was us ed as a deformable template and automatically matched to three-dimensi onal MR images of 30 individuals (five healthy and five schizophrenic subjects). A subvolume containing the hippocampus was defined by using 16 landmarks that constrained the automated search for hippocampal bo undaries. Transformation of the hippocampus template was automatically performed by using global pattern matching through a sequence of low- then high-dimensional translations, rotations, and scalings. RESULTS: The average test-retest volume difference measured with the automatic method was 3.1%, compared with the manual test-retest difference of 7 .1%. Correlation between automated and manually determined volumes dem onstrated the validity of the automated technique (intraclass correlat ion coefficient = .86). CONCLUSION: The automated method estimates hip pocampal volumes with less variability tie, lower variance) than that of manual outlining.