Canadian Association of Radiologists White Paper on De-identification of Medical Imaging: Part 2, Practical Considerations

Abstract

The application of big data, radiomics, machine learning, and artificial intelligence (AI) algorithms in radiology requires access to large data sets containing personal health information. Because machine learning projects often require collaboration between different sites or data transfer to a third party, precautions are required to safeguard patient privacy. Safety measures are required to prevent inadvertent access to and transfer of identifiable information. The Canadian Association of Radiologists (CAR) is the national voice of radiology committed to promoting the highest standards in patient-centered imaging, lifelong learning, and research. The CAR has created an AI Ethical and Legal standing committee with the mandate to guide the medical imaging community in terms of best practices in data management, access to health care data, de-identification, and accountability practices. Part 2 of this article will inform CAR members on the practical aspects of medical imaging de-identification, strengths and limitations of de-identification approaches, list of de-identification software and tools available, and perspectives on future directions.

2021 Year in Review

“…Following another white paper dedicated to ethical and legal issues in 2019, 5 the AI Working Group again published in our journal in 2021 two white papers discussing the general principles and practical considerations of de-identification of medical imaging. 6,7 We anticipate the AI Working Group will continue to produce influential guiding pieces under the new leadership of Dr Jaron Chong.…”

How to Improve the Management of Acute Ischemic Stroke by Modern Technologies, Artificial Intelligence, and New Treatment Methods

“…Radiology is a prime candidate for the early adoption and implementation of AI [141]. The best practices in data management and personal de-identification are necessary to prevent unacceptable risks of patient reidentification [142,143].…”

CARJ 2021: Year in Review

“…The 8 recommendations are summarized as follows: 1) Any custodian of patient data should be comfortable with a small inherent risk of data reidentification; 2) Public data sets should be encouraged if the data can be de-identified to low reidentification risk; 3) Medical AI algorithm validation is important prior to their use in an institution; 4) Commercialization of AI adds complexity, although is important, in developing AI applications; 5) Reidentification competitions have value highlighting vulnerabilities in deidentification and encryption processes; 6) Analogous to the ALARA principle, de-identification should aim to have ''as little retained medical data as reasonably acceptable''; 7) Up-to-date storage and encryption help maintain confidentiality in the event of a data breach; and 8) De-identification is a right. 14,15 The use of these recommendations as well as educating ourselves on the background and methods of de-identification can allow us to safely perform research and advance our field further into the 21st century while ensuring a safe and respectful environment for our patients.…”

“…Part 1 focused on the general principles of de-identification and part 2 on practical considerations. 14,15 In total, 8 recommendations were made for the de-identification of data in artificial intelligence research. The first 4 address background considerations and the last 4 are considerations for the specific application of de-identification.…”