Quickshear defacing for neuroimages

  • Authors:
  • Nakeisha Schimke;John Hale

  • Affiliations:
  • Institute of Bioinformatics and Computational Biology, University of Tulsa;Institute of Bioinformatics and Computational Biology, University of Tulsa

  • Venue:
  • HealthSec'11 Proceedings of the 2nd USENIX conference on Health security and privacy
  • Year:
  • 2011

Quantified Score

Hi-index 0.00

Visualization

Abstract

Data sharing offers many benefits to the neuroscience research community. It encourages collaboration and interorganizational research efforts, enables reproducibility and peer review, and allows meta-analysis and data reuse. However, protecting subject privacy and implementing HIPAA compliance measures can be a burdensome task. For high resolution structural neuroimages, subject privacy is threatened by the neuroimage itself, which can contain enough facial features to re-identify an individual. To sufficiently de-identify an individual, the neuroimage pixel data must also be removed. Quickshear Defacing accomplishes this task by effectively shearing facial features while preserving desirable brain tissue.