In my roles as both a technology executive and a medical imaging core director, I spend much of my time with research teams at some of the most respected healthcare institutions in the world that enroll patients in clinical trials. That perspective gives me a direct view into the operational realities that clinical trial study staff face every day.
I have heard clinical trial site study staff describe de-identification and export for central review as “the bane of my existence.” That reaction is not surprising. The workflow is still manual, time-intensive, and far too easy to get wrong. Thousands of healthcare sites enroll trial participants in studies that include medical imaging as part of the protocol.
To prepare images for central review, site staff typically have to retrieve the correct images for each trial visit from PACS or request them from an outside facility, de-identify the files by removing PHI from headers and addressing any burned-in PHI from the images, perform manual quality checks and cleanup, and then upload the de-identified images to a sponsor or its delegated partner for archival and analysis. Every step is manual. Every step takes time. And every step introduces another opportunity for delay, inconsistency, or error.
Sites may be managing dozens to hundreds of clinical trials across thousands of patients, and each trial may have different requirements, including rules for which image header information should be preserved or removed. Those trials may also involve a wide range of sponsors working with various delegated core labs, CROs, and iCROs, each with its own portal, upload process, interface, instructions, and support model for site staff to navigate. It’s exhausting and frustrating for study staff. I’m tired just thinking about it!
When the process is this manual, mistakes are inevitable. Sites can upload the wrong information to the central archive and then have to retrieve, de-identify, and re-upload the correct scans. As one iCRO noted, they once even received uploaded images of someone’s child’s birthday party! Those errors create extra work for sites, sponsors, and imaging delegates alike. Even when the same patient returns across multiple visits, the de-identification process is often treated as a one-off task each time, with staff repeatedly looking up and reapplying the same anonymization coding. That is not a connected workflow. It is repeated manual labor.
The pressure on site staff is even greater because many protocols impose limited time windows for de-identification and upload. In some cases, the individual responsible for retrieving scans from PACS and preparing them for submission has little or no backup. That creates a real risk of delayed image uploads and missed protocol windows at exactly the moment when speed and accuracy matter most.
Sites also often need to track and record how many scans were uploaded for which trials and which central organizations, adding another layer of manual administrative work. That burden consumes valuable staff time and limited resources that should be devoted to higher-value work, including study startup and patient enrollment.
What sites need is a simpler, more connected workflow that reduces the burden of anonymization and export instead of adding to it. They need a system that helps them retrieve the right images, apply the right de-identification rules, preserve traceability, and export sponsor-ready trial data through a workflow they can actually manage. These are the types of challenges that have lingered in research for far too long, and they are the kinds of problems Yunu was built to solve.
With De-ID & Export, Yunu now gives sites an in-workflow way to anonymize and export trial images and results directly from Yunu. It is straightforward, embedded in the site workflow, and designed to reduce the manual, error-prone processes sites have been left to manage for years. Preparing protocol-compliant data for transfer is now far more controlled, consistent, and practical for the teams doing the work. Sites needed a simpler way to tackle these tasks. Yunu delivered. As we like to say… “You knew there had to be a better way. Now there is.”
