Evaluation of 14 nonlinear deformation algorithms applied to human brain MRI registration

• Published in: NeuroImage (Orlando, Fla.) Vol. 46; no. 3; pp. 786 - 802
• Main Authors: Klein, Arno, Andersson, Jesper, Ardekani, Babak A., Ashburner, John, Avants, Brian, Chiang, Ming-Chang, Christensen, Gary E., Collins, D. Louis, Gee, James, Hellier, Pierre, Song, Joo Hyun, Jenkinson, Mark, Lepage, Claude, Rueckert, Daniel, Thompson, Paul, Vercauteren, Tom, Woods, Roger P., Mann, J. John, Parsey, Ramin V.
• Format: Journal Article
• Language: English
• Published: United States Elsevier Inc 01-07-2009; Elsevier Limited; Elsevier
• Subjects: Adult; Algorithms; Anatomy & physiology; Architecture; Artificial Intelligence; Automation; Bioengineering; Brain; Brain - anatomy & histology; C plus plus; Computer Science; Female; Humans; Image Enhancement; Image Enhancement - methods; Image Interpretation, Computer-Assisted; Image Interpretation, Computer-Assisted - methods; Imaging; Labeling; Life Sciences; Linux; Magnetic Resonance Imaging; Magnetic Resonance Imaging - methods; Male; Medical Imaging; NMR; Nonlinear Dynamics; Nuclear magnetic resonance; Nuclear medicine; Operating systems; Pattern Recognition, Automated; Pattern Recognition, Automated - methods; Reproducibility of Results; Sensitivity and Specificity; Servers; Subtraction Technique
• ISSN: 1053-8119; 1095-9572
• DOI: 10.1016/j.neuroimage.2008.12.037

All fields of neuroscience that employ brain imaging need to communicate their results with reference to anatomical regions. In particular, comparative morphometry and group analysis of functional and physiological data require coregistration of brains to establish correspondences across brain structures. It is well established that linear registration of one brain to another is inadequate for aligning brain structures, so numerous algorithms have emerged to nonlinearly register brains to one another. This study is the largest evaluation of nonlinear deformation algorithms applied to brain image registration ever conducted. Fourteen algorithms from laboratories around the world are evaluated using 8 different error measures. More than 45,000 registrations between 80 manually labeled brains were performed by algorithms including: AIR, ANIMAL, ART, Diffeomorphic Demons, FNIRT, IRTK, JRD-fluid, ROMEO, SICLE, SyN, and four different SPM5 algorithms (“SPM2-type” and regular Normalization, Unified Segmentation, and the DARTEL Toolbox). All of these registrations were preceded by linear registration between the same image pairs using FLIRT. One of the most significant findings of this study is that the relative performances of the registration methods under comparison appear to be little affected by the choice of subject population, labeling protocol, and type of overlap measure. This is important because it suggests that the findings are generalizable to new subject populations that are labeled or evaluated using different labeling protocols. Furthermore, we ranked the 14 methods according to three completely independent analyses (permutation tests, one-way ANOVA tests, and indifference-zone ranking) and derived three almost identical top rankings of the methods. ART, SyN, IRTK, and SPM's DARTEL Toolbox gave the best results according to overlap and distance measures, with ART and SyN delivering the most consistently high accuracy across subjects and label sets. Updates will be published on the http://www.mindboggle.info/papers/ website.