A vital publication has recently been issued in Nature Communications, describing the privacy risks associated with sharing whole-slide images in the field of digital pathology.
This work was carried out with support from EOSC-Life.
Providing access to ‘whole-slide images’—high-resolution scans of complete pathological slides—is necessary for the development novel diagnostic AI methods in pathology, education/training of pathologists, and research.
Risk analysis methods that could be used to evaluate the privacy risks associated with sharing these imaging data were, however, until now lacking.
This landmark paper describes a model developed for the privacy risk analysis for whole-slide images, and especially to protect the identity of the sample donor.
The developed risk assessment model and resulting taxonomy were applied in experiments to demonstrate the risks of using real-world imaging data. Based on these results, guidelines for risk assessment and recommendations for low-risk sharing of whole-slide image data could be developed.
This publication represents a crucial step forward in data privacy protection in the field of digital pathology!
