Medidata Blog

Medical Imaging is Scary without a Trusted Solution | 5 Ways CROs Can Future Proof Their Medical Imaging Strategy Series

Nov 29, 2022 - 5 min read
Medical Imaging is Scary without a Trusted Solution | 5 Ways CROs Can Future Proof Their Medical Imaging Strategy Series

This is the fourth blog in the 5 Ways CROs Can Future Proof Their Medical Imaging Strategy series. 

This blog was authored by Andrea Falkoff, Senior Director, Product Management, Rave Imaging at Medidata. Andrea is responsible for the Medidata Rave Imaging product strategy and also works directly with customers and internal stakeholders to ensure best practices are applied across imaging in clinical trials.

 

Why is Imaging Scary?

Medical imaging is an intimidating space for CROs to enter due to three obvious features—the large size of the files, the complexity of the data, which may require an understanding of medical images and scan procedures, and the importance of successfully de-identifying this data in a meaningful way. Medical image analysis data differs from all other data collected during a trial and there’s variations in types of images and file formats.

So the concept of managing, storing, and handling this data feels quite daunting. It doesn’t have to be though, and there’s a few ways for CROs to become comfortable with medical imaging. These include focusing on early-phase trials—especially those where the imaging requirements are more consistent with clinical standard of care; removing manual steps through the use of technology and leveraging automation; and uploading to a centralized, cloud-based solution rather than hosting the data.

Turning to Core Labs Unnecessarily

Core labs have the required expertise to handle imaging and the ability to remove defining, personal data. Moving the imaging responsibility to an expert might sound like the best, most efficient option, but there are many limitations to this “out of sight, out of mind” approach. Usually, contracting core labs in early-phase trials is redundant. This is especially true of early-stage oncology trials, where medical imaging is now consistent with standard of care at the site and central reads are typically not required. 

On these trials, sponsors don’t want to contract yet another vendor, preferring instead to outsource this along with all other clinical trial activities to a CRO. Contracting an imaging core lab also prevents real-time access to medical images and, perhaps most importantly, it means introducing another data silo and having to crosswalk between imaging upload and whichever solution is being used for electronic data capture (EDC). And finally, it’s an added expense that might not be necessary.

This is especially important in early-stage trials as budgets are tighter, contracting reading services and expensive radiologists are likely not needed, and training on image acquisition is not required due to the straightforward nature of the imaging requirements. Getting comfortable with medical imaging early on is optimal because the trial size is smaller and there are less images to manage. Becoming proficient at this stage then makes it easier to scale up in later-stage trials.

While medical image management has typically been outsourced, including the software, a CRO can now continue to outsource the service but choose to own the software element. In this model, a core lab can still be used for image quality control (QC) and reader services.

Demystifying Medical Imaging

Medical imaging seems scary, but it’s not so daunting with a solution that structures the steps required to collect, QC, and store images. By leveraging the features within Medidata Rave Imaging, CROs can feel confident in independently managing imaging on these early-phase trials. Below are the steps that trial sites take in a common, early-stage oncology trial to help demystify the fear of imaging. 

Electronic Data Capture Integration

This process makes sure a patient is entered into Rave Imaging as soon as the patient is enrolled in Medidata Rave EDC, saving a user from having to manually enter this information in order to upload their images. Imaging status information is also displayed in Rave EDC, saving users from having to access the imaging tool directly to check on the status of images. This removes errors and vulnerability. Rave Imaging can also be used as a stand-alone solution alongside any other EDC platform.

Edit Checks

Real-time edit checks for medical image uploads ensure minimal errors. These checks may be as simple as confirming that a scan is a CT scan, for example, or checking for specific scan parameters, like slice thickness. Medical image quality assurance checks can also verify that the user hasn’t uploaded the same scans previously to a different visit, ensuring the right scans are selected at the time of upload. These edit checks massively help reduce site burden. 

Image Filtering

This can be enabled so only the medical images relevant to the clinical trial are uploaded, significantly reducing and eliminating images with pixel PHI before they enter Rave Imaging. This feature improves the quality of the image data while reducing the burden on the site. 

Automatic De-identification

Images and scans have DICOM tags, or metadata elements associated with an image (for example, date and time the scan was performed). Both patient demographics and imaging metadata are stored in the DICOM tags of an exam. All of this needs to be anonymized and being able to automate that process saves a tremendous amount of time, not to mention minimizes risk and errors. Rave Imaging automatically modifies DICOM tags appropriately on upload, according to the trial requirements. 

Pixel Redaction

During a medical image upload and once the pixel data is reviewed, trial sites or CROs can perform the pixel redaction with cloud-based tools within Rave Imaging before the image is submitted. Rave Imaging even has automated pixel text detection tools for CT and MRI images that help identify and bulk redact any pixel PHI found on the images automatically. 

Overdue Visit Reminders

Rave Imaging has the ability to automatically alert sites when images are overdue according to the trial’s visit schedule, improving compliance with image uploads. 

eLearnings & Support Pages

Software is only as good as the training and support that goes with it. There is a plethora of accessible learning and support tools to go alongside the Rave Imaging solution. CROs can mandate these learnings or make them optional in order to familiarize a site with the solution. Additionally, Medidata’s dedicated Helpdesk delivers comprehensive support for sites, sponsors, and CROs.

This is why medical imaging doesn’t need to be scary or overwhelming. These are the only steps needed in an early-phase collect and hold study and Medidata’s image data management system simplifies all these steps. Medidata’s EDC platform is a highly structured data system and those same abilities and principles have been applied into the imaging solution. Given that a large number of trial sites are already interacting with Rave EDC and are already in the platform, using the same platform and log-ins for imaging is a relatively seamless process.

A Brighter Medical Imaging Future

While imaging might be a new undertaking, the steps outlined above demonstrate that any CRO is capable of owning this piece themselves to have more control over costs and keep trials within budget. 

As a starting point, CROs should understand how to incorporate imaging in their data management practices. And if this is currently being outsourced, Medidata can consult on how to own the software perspective. It likely makes sense to continue outsourcing imaging services, but there’s clear benefits to de-bundling services from software. By minimizing the need to engage core labs in these early-stage trials and owning the software piece, CROs can be on their way to future-proofing their medical imaging strategies.

In our next blog, learn how Medidata trains CROs to enable these features to start generating revenue and take true ownership of their imaging revenue. Catch up on previous blogs in this series here: blog 1, blog 2, and blog 3.

Subscribe to Our Blog Newsletter