DAY 1, May 22, 2023 (10:00 am to 2:00 pm EDT)
Welcome and Opening Comments
Keyvan Farahani, PhD, Center for Biomedical Informatics and Information Technology, National Cancer Institute, National Institutes of Health
Session 1: Report of the MIDI Task Group
Report of the Medical Image De-Identification (MIDI) Task Group - Best Practices and Recommendations
Session Chair: David Clunie, MBBS, PixelMed, Inc.
In this session, David Clunie, chair of the MIDI Task Group, will summarize the best practices and recommendations included in the task group's report, recently available in pre-print, followed by a question and answer period.
Session 2: Tools for Conventional Approaches to De-Identification
Session Chair: Fred Prior, PhD, University of Arkansas for Medical Sciences
In this session, speakers will share methods currently in use for medical de-identification.
Break
Estimated at 11:55-12:05 p.m.
Session 3: International Approaches to De-Identification
Session chair: Bill Parker, MD, DABR, FRCPC, University of British Columbia
This session will focus on international approaches to de-identification.
Session 3: Industry Approaches to De-Identification
Session chair: Keyvan Farahani, PhD, Center for Biomedical Informatics and Information Technology, National Cancer Institute, National Institutes of Health
The priorities of and approaches by industry when it comes to protecting human identity in medical images can differ from those of governments. This session will examine those priorities and approaches with presentations by various industry groups.
Closing Remarks
David Clunie, MBBS, PixelMed, Inc.
DAY 2, May 23, 2023 (10:00 am to 2:00 pm EDT)
Welcome and Recap
David Clunie, MBBS, PixelMed, Inc.
Session 4: Statistical Risk Analysis of Indirect Identifiers in an Image Context
Session chair: Mark Elliot, PhD, University of Manchester
In this session, researchers will discuss how to analyze the re-identification risk of indirect identifiers in medical images.
Session 5: De-facing
Session chair: Ying Xiao, Ph.D. Hospital of the University of Pennsylvania
This session will focus on balancing the risks of removing potentially reconstructable facial information in head and neck cross-sectional images, also called de-facing, with the diminished utility of these images caused by restricted access to them.
Break
Estimated at 11:50 a.m. - 12:00 p.m.
Session 6: The Role of Artificial Intelligence in De-Identification
Session chair: Judy Wawira Gichoya, MD, MS, Emory University
In this session, researchers discuss the utility and risk of using algorithms from machine learning (ML) and artificial intelligence (AI) to de-identify medical images.
Session 7: Panel Discussion
Session chair: Jürgen Klenk, PhD, Deloitte Consulting LLP
Discussion of the approaches and challenges identified during the workshop and opportunities for the future.
Closing Remarks
David Clunie, MBBS, PixelMed, Inc.
