Abstract
To accurately chart the dynamic brain developmental trajectories in infants, many longitudinal neuroimaging studies prefer having a complete dataset. Unfortunately, missing data at certain time points are unavoidable in longitudinal datasets. To better use incomplete longitudinal data, we propose a novel method to estimate the subject-specific vertex-wise cortical thickness maps at missing time points, by using a customized regression forest, Dynamically-Assembled Regression Forest (DARF). DARF ensures spatial smoothness of the estimated cortical thickness maps and also the computational efficiency. The proposed method can fully exploit the available information from the subjects both with and without missing scans. Our method has been applied to estimate the missing cortical thickness maps in a longitudinal infant dataset, which includes 31 healthy subjects, with each having up to 5 scans. The experimental results indicate that our method can accurately estimate missing cortical thickness maps, with the average vertex-wise error less than 0.23 mm.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
More From: "Medical Computer Vision: Algorithms for Big Data" : International Workshop, MCV 2015, held in conjunction with MICCAI 2015, Munich, Germany, October 9, 2015 : revised selected papers. MCV (Workshop) (5th : 2015 : Munich, Germany)
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.