Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2020 Sep;17(9):1159-1162.
doi: 10.1016/j.jacr.2020.04.007. Epub 2020 Apr 28.

Medical Imaging and Privacy in the Era of Artificial Intelligence: Myth, Fallacy, and the Future

Eyal Lotan  1 , Charlotte Tschider  2 , Daniel K Sodickson  3 , Arthur L Caplan  4 , Mary Bruno  5 , Ben Zhang  6 , Yvonne W Lui  7
Affiliations

Medical Imaging and Privacy in the Era of Artificial Intelligence: Myth, Fallacy, and the Future

Eyal Lotan et al. J Am Coll Radiol. 2020 Sep.
No abstract available

PubMed Disclaimer

Conflict of interest statement

The authors state that they have no conflict of interest related to the material discussed in this article..

Figures

Fig 1.
Fig 1.
Whereas in the past, dedicated medical imaging software was required, surface rendering of high-resolution (1-mm) medical imaging data is now easily achievable using freely available software such as Fast STL Viewer, an app intended for 3-D printing, shown here in use on an Android mobile device (A). Advances in image postprocessing methods such as slice interpolation allow the generation of a facial likeness even from 2-D, thick-section images (5 mm), albeit with noticeable stairstep artifacts (B). A digital photograph of the same individual (C) shows that identifying people using reconstructed images alone may not be trivial; however, advances in facial recognition algorithms are likely to resolve current shortcomings. Images used with permission.
Fig 2.
Fig 2.
Implementation of one of the top-performing segmentation algorithms from the 2017 Medical Image Computing and Computer Assisted Intervention Brain Tumor Segmentation Challenge shows excellent results (A) when data are preprocessed using skull-stripping and registration algorithms before inputting into a segmentation neural network; however, when the algorithm is deployed on a clinical MRI data set, the resulting segmentation contains many errors (B). Here, image segments identified as enhancing tumor are outlined in red, regions of enhancing tumor with any central nonenhancing components are shown in green, and regions of tumor with surrounding edema are shown in blue.
See this image and copyright information in PMC

References

    1. Price WN, Cohen IG. Privacy in the age of medical big data. Nat Med 2019;25:37–43. - PMC - PubMed
    1. Tschider C The consent myth: improving choice for patients of the future. Wash L Rev 2018;96:1505–36.
    1. The Health Insurance Portability and Accountability Act of 1996 (HIPAA). Pub L No 104–191, August 21, 1996; Security and Privacy 45 CFR Part 160.103 (2002).
    1. Decker SJ, Ford JM. Forensic personal identification utilizing part-to-part comparison of CT-derived 3D lumbar models. Forensic Sci Int 2019;294:21–6. - PubMed
    1. Finn ES, Shen X, Scheinost D, et al. Functional connectome fingerprinting: identifying individuals using patterns of brain connectivity. Nat Neurosci 2015;18:1664–71. - PMC - PubMed