Adaptive Registration of Varying Contrast-Weighted Images for Improved Tissue Characterization (ARCTIC): Application to T1 Mapping

PurposeTo propose and evaluate a novel nonrigid image registration approach for improved myocardial T-1 mapping. MethodsMyocardial motion is estimated as global affine motion refined by a novel local nonrigid motion estimation algorithm. A variational framework is proposed, which simultaneously estimates motion field and intensity variations, and uses an additional regularization term to constrain the deformation field using automatic feature tracking. The method was evaluated in 29 patients by measuring the DICE similarity coefficient and the myocardial boundary error in short axis and four chamber data. Each image series was visually assessed as no motion or with motion. Overall T-1 map quality and motion artifacts were assessed in the 85 T-1 maps acquired in short axis view using a 4-point scale (1-nondiagnostic/severe motion artifact, 4-excellent/no motion artifact). ResultsIncreased DICE similarity coefficient (0.78 0.14 to 0.87 +/- 0.03, P < 0.001), reduced myocardial boundary error (1.29 +/- 0.72 mm to 0.84 +/- 0.20 mm, P < 0.001), improved overall T-1 map quality (2.86 +/- 1.04 to 3.49 +/- 0.77, P < 0.001), and reduced T-1 map motion artifacts (2.51 +/- 0.84 to 3.61 +/- 0.64, P < 0.001) were obtained after motion correction of with motion data (approximate to 56% of data). ConclusionsThe proposed nonrigid registration approach reduces the respiratory-induced motion that occurs during breath-hold T-1 mapping, and significantly improves T-1 map quality. Magn Reson Med 73:1469-1482, 2015. (c) 2014 Wiley Periodicals, Inc.