Medical Image De-Identification Automation and Validation
The theme of June’s NCI Imaging and Informatics Community Webinar will focus on DICOM medical image de-identification (MIDI). Patient privacy rules require the removal of protected health information (PHI) prior to sharing medical images publicly. With the growing volume of imaging data, manually removing PHI is inefficient. In this presentation, speakers will disclose how their teams assessed the accuracy of automated de-identification algorithms and cloud services in removing PHI.
“A DICOM Data Set for Evaluation of Medical Image De-Identification” |1:00–1:20|
Dr. Fred Prior, the Chair of the Department of Biomedical Informatics at the University of Arkansas for Medical Sciences, will share the Cancer Imaging Archive (TCIA) team’s efforts to develop a DICOM data set that can be used to evaluate the performance of de-identification algorithms.
“Medical Image De-Identification Using Cloud Services” |1:20–1:40|
Dr. Benjamin Kopchick and Ms. Dina Mikdadi from the Deloitte Biomedical Data Science team will present on the use of a cloud service for automated MIDI. Their team tested the Google Cloud Platform’s Healthcare API service leveraging data sets with synthetic PHI generated by the TCIA team.
This event is free and open to the public.
Dr. Prior is the Professor and Inaugural Chair of the Department of Biomedical Informatics and Professor of Radiology at the University of Arkansas for Medical Sciences (UAMS).
Dr. Kopchick is a member of the Biomedical Data Science team at Deloitte. His current focus is on machine learning and artificial intelligence using medical images.
Ms. Mikdadi is a data scientist at Deloitte focused on clinical, imaging, and bioinformatics serving clients in the life science and healthcare space. Much of Ms. Mikdadi’s work lies at the intersection of science and technology, testing and validating cloud-native, cutting-edge solutions (AI/ML) for biomedicine.
- NCI Imaging Data Commons Hosts 2022 RSNA Session: Towards Transparency and Reproducibility in Imaging AINovember 27, 2022 - November 30, 2022DataViz + Cancer Micro Lab: Using Data Visualization and Interactive Experiences to Advance Immuno-OncologyDecember 01, 2022