2019
DOI: 10.1056/nejmc1908881
|View full text |Cite
|
Sign up to set email alerts
|

Identification of Anonymous MRI Research Participants with Face-Recognition Software

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

1
119
0
1

Year Published

2020
2020
2024
2024

Publication Types

Select...
8
1

Relationship

0
9

Authors

Journals

citations
Cited by 162 publications
(121 citation statements)
references
References 4 publications
1
119
0
1
Order By: Relevance
“…7 University of Pennsylvania (UPenn), Philadelphia, PA, USA. 8 Owkin, Paris, France. 9 Vanderbilt University, Nashville, TN, USA.…”
Section: Data-driven Medicine Requires Federated Effortsmentioning
confidence: 99%
See 1 more Smart Citation
“…7 University of Pennsylvania (UPenn), Philadelphia, PA, USA. 8 Owkin, Paris, France. 9 Vanderbilt University, Nashville, TN, USA.…”
Section: Data-driven Medicine Requires Federated Effortsmentioning
confidence: 99%
“…Even if data anonymisation could bypass these limitations, it is now well understood that removing metadata such as patient name or date of birth is often not enough to preserve privacy 7 . It is, for example, possible to reconstruct a patient's face from computed tomography (CT) or magnetic resonance imaging (MRI) data 8 . Another reason why data sharing is not systematic in healthcare is that collecting, curating, and maintaining a high-quality data set takes considerable time, effort, and expense.…”
Section: Introductionmentioning
confidence: 99%
“…Re-identification of participants based on scan data had been considered a remote possibility, which motivated the creation of defacing algorithms (Alfaro-Almagro et al, 2018;Bischoff-Grethe et al, 2007;Milchenko & Marcus, 2013;Schimke, Kuehler, & Hale, 2011). Such re-identification, however, has recently been demonstrated to be feasible (Schwarz et al, 2019), which now renders defacing mandatory for publicly available data. Moreover, two recent developments further complicate matters.…”
Section: De-identificationmentioning
confidence: 99%
“…Several researchers have shown that it is possible to reconstruct biometric facial features from CT and MRI examinations of the head and reidentify patients when searched across limited databases using facial recognition technology [8][9][10]. Although it may be difficult for an unfamiliar observer to match these images to a specific individual [8], it is not hard to imagine that a company such as Google or Facebook could use these data with facial recognition software against a pool of patients identified from geotags at a particular institution during a defined time period to reidentify these patients.…”
Section: Challenges With Anonymization Of Radiologic Imagesmentioning
confidence: 99%