Content area
Due to their high resolution, anonymized CT scans can be reidentified using face recognition tools. However, little is known regarding PET deanonymization because of its lower resolution. In this study, we analysed PET/CT scans of 853 patients from a TCIA-restricted dataset (AutoPET). First, we built denoised 2D morphological reconstructions of both PET and CT scans, and then we determined how frequently a PET reconstruction could be matched to the correct CT reconstruction with no other metadata. Using the CT morphological reconstructions as ground truth allows us to frame the problem as a face recognition problem and to quantify our performance using traditional metrics (top k accuracies) without any use of patient pictures. Using our denoised PET 2D reconstructions, we achieved 72% top 10 accuracy after the realignment of all CTs in the same reference frame, and 71% top 10 accuracy after realignment and mixing within a larger face dataset of 10, 168 pictures. This highlights the need to consider face identification issues when dealing with PET imaging data.
Details
; Berenbaum, Arnaud 2 ; Allassonnière, Stéphanie 3 ; Besson, Florent L. 4 1 ENS Paris-Saclay, Paris, France (GRID:grid.5607.4) (ISNI:0000 0001 2353 2622)
2 BioMaps, Université Paris-Saclay, Commissariat à l’énergie atomique et aux énergies alternatives (CEA), Centre National de la Recherche Scientifique (CNRS), Inserm, Orsay, France (GRID:grid.5607.4)
3 Sorbonne Université, CRC, HeKA, Parisanté Campus, Université Paris Cité, Inria, Inserm, Paris, France (GRID:grid.462844.8) (ISNI:0000 0001 2308 1657)
4 BioMaps, Université Paris-Saclay, Commissariat à l’énergie atomique et aux énergies alternatives (CEA), Centre National de la Recherche Scientifique (CNRS), Inserm, Orsay, France (GRID:grid.462844.8); Assistance Publique-Hôpitaux de Paris, DMU SMART IMAGING, CHU Bicêtre, Department of Nuclear Medicine-Molecular Imaging, Hôpitaux Universitaires Paris-Saclay, Le Kremlin-Bicêtre, France (GRID:grid.50550.35) (ISNI:0000 0001 2175 4109); School of Medicine, Université Paris-Saclay, Le Kremlin-Bicêtre, France (GRID:grid.460789.4) (ISNI:0000 0004 4910 6535)