… High on the agenda? Difficult? Important? Necessary?
These are some examples of the unpublished words of the question “Why is interoperability so…?” of my previous blog. The answer to all these? Because it brings value to whom needs it!
Yes, unfortunately it is also the answer to the question why interoperability is so difficult. As industry, as government, as health care provider, as human being, we all want the best for our people, we all want to bring value to whom needs it. We also want to ensure the benefits outweigh the risks. We all are aware there could be risks when two or more devices, each with their own intended use, will be combined and used for a different purpose than intended. Therefore, interoperability can be regarded as difficult, I rather phrase it as ‘can be perceived as becoming complex’.
Nevertheless, it might not be as complex as we think. During the pandemic, the Centers for Medicare & Medicaid Services (CMS), who keeps oversight of practice of medicine, temporarily waived the requirement for remote locations to have separate Clinical Laboratory Improvement Amendments (CLIA) licenses, provided that the primary lab has a CLIA certificate, see also College of American Pathologist (CAP) website. Also, the Food and Drug Administration (FDA), who keeps oversight of medical devices marketed by vendors, allowed through the Emergency Use Authorization (EUA) remote use of WSI system devices. Even more, under certain conditions the agency temporarily allowed non-510(k) cleared WSI devices to be used, including ‘mixing and matching’ the different components of a WSI system device.
Now the pandemic is declared over, CMS has not revoked their remote use waiver . I have been informed they are working towards updating the rules, allowing remote use permanently. FDA however, has a transition plan in place to returning to the ‘normal’ policies, i.e., pre-pandemic.
Even though I understand this, it does not make 100% sense in my brain. During the pandemic, as far as I am aware of, no serious incidents were reported while health care providers were using components from different vendors interchangeably in the WSI system device. This shows, that both vendors and health care providers, including pathologists who are well trained medical professionals, privileged to sign out cases, trust this interoperability between components.
This leads me to the next reason why it doesn’t make 100% sense in my brain. That is, there should be ways to leverage this real world evidence to allow interoperability of WSI devices.
As you might have sensed from previous blogs, I am a strong believer in teamship, partnership and collaborations. I believe that we could and should use these learnings, this real world evidence and related data to speed up the pace of collaboration between policy makers, vendors, and health care providers . This to define 1) minimal interoperability requirements for the different WSI components and 2) general principles to safeguard interoperability – for the sake of patients. This speeding up the pace is essential to ensure the momentum of digital pathology evolution will not get lost.
- A software company developing Software as a Medical Device (SaMD) could distort images, and use this to develop their algorithm to make its algorithm robust. That is, the algorithm safety and effectiveness performance will not change when a scanner has been modified or another scanner with same minimal requirements will be used to produce WSIs for the algorithm to ingest.
- Or they could have included internal controls in the algorithm to measure changes when scanning a calibration slide. When the changes are outside boundaries an alarm is triggered.
- Or a phantom can be used to measure the algorithm performance
- Or the use of by FDA accepted training, tuning, test and validated datasets managed and under control by a neutral third party.
There are multiple options which perfectly address the FDA guidance ”Design considerations and pre-market submission recommendations for interoperable medical devices”.
I am fully aware the devil is in the details, and I do not know all the technical details because I am not an engineer. And this is my point exactly. This is why I am a strong believer in collaboration. I am invested to speed up the pace of collaboration. That is why I am working with the Digital Pathology Association (DPA) and Pathology Innovation Collaborative Community (PIcc) who are collaborating with all the stakeholders, including the FDA to make interoperability happen.
Let’s work in teamship to getting the big question “when will components of WSI systems become truly interoperable, enabling and allowing mixing and matching, plugging and diagnosing for the sake of patients” answered.