Preserving Privacy in Structural Neuroimages

Evolving technology has enabled large-scale collaboration for neuroimaging data. For high resolution structural neuroimages, these data are inherently identifiable and must be given the same privacy considerations as facial photographs. To preserve privacy, identifiable metadata should be removed or replaced, and the voxel data de-identified to remove facial features by applying skull stripping or a defacing algorithm. The Quickshear Defacing method uses a convex hull to identify a plane that divides the volume into two parts, one containing facial features and another the brain volume, and removes the voxels on the facial features side. This method is an effective alternative to existing solutions and can provide reductions in running time.