Data Quality

The Brain Connectivity and Cognition Laboratory at the University of Miami has performed a quality check on a subset of the Rockland Sample data. Four trained undergraduates supervised by Dina Dajani and Paola Odriozola checked 176 subjects between the ages of 8 and 50.

MPRAGE and 1400 ms TR resting-state scans from the first visit were categorized as passing, qualified passing (QP), or failing. Passing data does not have any issues. QP data for structural images is defined as data usable for coregistration but not for structural studies. QP data for resting-state data can mean that one of several issues (e.g., motion slice, superior coverage issues) were noted both as one-word and detailed comments. Failing data is unusable. For any resting-state images with motion slice issues that are not considered failing, the offending slices are noted and a second rater has double-checked that volumes are correct. More detailed information about how these categories are defined can be found here.

The results of this quality check may be downloaded as a CSV here.

Credits

• Lucina Uddin
• Dina Dajani
• Paola Odriozola
• Melanie Winters
• Laura Molina
• Michael Ortega
• Hannah Long