As healthcare organizations see their troves of imaging data grow exponentially, many are concerned about how to leverage that data to their best advantage. The restructuring of processes and image metadata practices are a must as healthcare providers redefine their enterprise imaging informatics strategy—however, understanding the importance of having access to vendor-neutral data is the key to getting the most out of imaging data sets. Mark Watts, Associate Vice President of Enterprise Imaging at Novarad and a top 25 innovator in health imaging and IT, talks with Healthcare Informatics about the power of imaging informatics, the challenge of having truly vendor-neutral data, and why the right IT partnerships are crucial to thriving in a value-based care world.
Tell us more about imaging informatics. How does it differ from traditional informatics approaches?
Mark Watts: Consider all the images that are used throughout the hospital. Traditionally, imaging informatics was used in the Digital Imaging and Communications in Medicine (DICOM) world for radiology and cardiology to assist with productivity. But when we talk about imaging informatics, we’re talking about expanding the ownership and governance of that data beyond the walls of the radiology and cardiology departments. We’re talking about taking every single image that a healthcare organization may acquire and taking the metadata from those images so it can be presented, stored, and analyzed in a logical manner.
We know that a picture is worth a thousand words. We capture an image and then bring it in to our system so we can involve it in patient interactions, in order to show them the problem and show them why a particular course of treatment is necessary. This helps that patient become more engaged and helps them understand the treatment path they are on. Imaging informatics can facilitate that story so the physician can really engage with the patient and bring understanding to the bedside.
However, there is also an analytics side to this. With advanced computer analytics and artificial intelligence systems, we now have the ability to do something with all this data we’ve been storing. It allows us to ask how we can unleash the value proposition, get into all that image data, and use it for research. We are learning how to use these data sets going forward. It may be for research projects to help with diagnosis and treatment. It might also assist radiologists if we can put those images into a clinical decision support system that might point out regions of interest that require a further look, but we can’t do that unless those data sets are normalized—and we can extract the data and make it available to others.
You emphasize the importance of vendor-neutral data. How can it help the entire enterprise—beyond the IT department?
Mark Watts: We’ve been storing all these data sets in individual silos throughout the organization. As we acquire new organizations, as so many hospitals are doing, you realize that for your organization to know where the data is and to make sure it meets all the compliance rules, it has to be standardized and brought under a single umbrella.
We’ve been under ransom to individual vendors who have privatized these data sets for years by denying us access to information that we own. For instance, if you leave a particular vendor, you then have to pay to get to that information. This is the same as paying a ransom to get your own data back from them. It’s a bad thing both for the healthcare organization and the patient. Each patient should have the right to their own data because we must be able to make decisions on our own healthcare. If our physician no longer has access to that data because of a merger or acquisition, how can we ensure that patients can exercise that right?
How does this benefit providers as we move to value-based care environments?
Mark Watts: Accountable care organizations are concerned about controlling the cost for their patient population. They’re looking for ways to move and free up data—to normalize it so that wherever it’s acquired, we don’t have this duplication of services. It’s estimated that seven percent of all images are reacquired because organizations don’t have access to the results or images themselves. If you ask any radiologist about the most important part of seeing a mammography study, they will tell you it’s being able to compare it to prior studies. The American College of Radiology listed the tools that radiologists found most valuable and access to a prior study was number one on the list. Having access to prior studies, even as patients move from location to location—whether we are talking about a partnering healthcare organization or even an entirely new city and hospital system—it’s critical to have access to patient information and studies, to re-standardize it so it can be consumed in multiple locations so we can best help our patients and reduce our costs.
What steps should healthcare organizations be taking to get to a point where they are best supporting vendor-neutral images and informatics?
Mark Watts: The industry has spent 15 to 20 years accumulating high resolution digital images, and we’ve just tossed them into these picture archiving and communication systems (PACS) and we rarely access them again. It is an untapped image data bank that has tremendous value if we store them in a logical manner, make them accessible, and have the right metadata to associate them with pertinent information.
The first step is to realize you have all these really useful data sets being stored in siloes that have no governance to them whatsoever. No awareness, no back-up, no continuity, and no audit trail for who might have accessed that information. You need to establish an enterprise architecture that allows you to ingest images in a native format so you can use them and apply the analytics and business intelligence you need moving forward. You won’t be able to avail yourself of these opportunities until you do.
In addition to this, you need the right partners. You need partners who will provide you with the ability to take advantage of new tools and to transfer this data in a vendor-neutral manner. If you have a partner who is privatizing this data, they are preventing that important data liquidity that you need to move, act, share, and collaborate on these data sets. It makes secondary and specialty readings problematic. Once you’ve digitized vendor-neutral data, you can have multiple people accessing one data set from remote locations. You then can facilitate collaboration, where two doctors can look at an image simultaneously and discuss the diagnosis and the treatment plan. And that’s good for your organization—and it’s good for the patient.