Quality Assessment in a DTI Multicenter Study

Abstract

Introduction: Multicenter DTI studies are becoming increasingly popular for their ability to improve the statistical power of a study by recruiting a large number of subjects [1]. Processing and analyzing data originating from various centers, however, presents unique challenges due to the intrinsic higher heterogeneity of experimental procedures compared to a single center study. Previous studies, for example, have highlighted scanner and site contributions to variability in multicenter studies [1,2]. Here we report the quality assessment and artifact remediation strategy that we implemented on the DTI data of the NIH MRI study of normal brain development. This unique study scanned unsedated healthy children in the age range 0-18 years with the purpose of creating a database of normative MRI and neuropsychological data (www.NIH-pediatricMRI.org). The emphasis of the study was on structural MRI data and the DTI acquisition was added as an “ancillary” component without strict quality control requirements for data acquisition. The challenge during data processing has been to produce good quality tensor–derived quantities that would be suitable for inclusion into a database, despite the highly heterogeneous nature of the incoming raw data.

1 Figure or Table

Cite this paper

@inproceedings{Nayak2010QualityAI, title={Quality Assessment in a DTI Multicenter Study}, author={Arjun Nayak and Lindsay Walker and Carlo Pierpaoli}, year={2010} }