December 11, 2019
Podcast: Business Intelligence
In this episode of “Top Five” we catch up with Jeff Vachon President of Bialogics and Tim O’Connell, CEO of Emtelligent. Top Five is a podcast series recorded at RSNA.
Q: Jeff and Tim, welcome to the Life Image podcast and welcome to RSNA, where we have about 50,000 radiologists and vendors from around the world. While the doctors here spend a lot of time examining and looking at the equipment and the technology around the equipment, interoperability is often an afterthought. Jeff, as a North American company that helps practitioners glean business intelligence, why do you think that could be a mistake? How should they be thinking about interoperability and business intelligence?
A: First off, the importance of collecting data is paramount because the data they collect today will have value for the life of the data. To collect the data, you have to have a strong framework of interoperability. Your systems have to be vendor-agnostic so that they don’t have to be reliant on one vendor. Next, they have to be able to collect data in a live format and be able to do so without complex interfaces and engines. The data is really being collected under the hood or behind the scenes and the practitioners only see the results of that collection of data in a user front end.
Q: Talk a little bit about what Biologics does in helping organizations, hospitals and provider groups in terms of extracting that really important data.
A: Once we collect the data from across the radiology departments, from the moment a patient is registered in the RIS, to the moment the finalized report is sent to the referring physician, all of that data is collected. As result, we can do very detailed pathways of the patients in the entire operational, performance and the workflow analytics of the department. The purpose for increased efficiencies, decreased wait times and turnaround times, but more than anything is to increase the operational efficiency of the department.
Q: Recently you’ve taken on a partner in order to integrate clinical intelligence into that. Talk about that.
A: Exactly, so bringing on operational data into the department is only one half. The real value of data is when we can combine that with the clinical intelligence that comes from the physicians’ reports. We’ve partnered with Emtelligent to really help us paint that full picture of the entire DI and patient experience. Dr. Tim O’Connell is the founder of Emtelligent and can answer to those very detailed questions.
Q: Tim, what were you solving for when you created your company? What did you see as a challenge?
A: Working as a busy radiologist, in a single shift you might read 50 CT scans and you might just be given, “right lower quadrant pain.” You’d open the scan there’d be a hundred priors. The problem I wanted to solve was to get to know my patients better. I wanted to produce the best report I could in that patient’s healthcare context. I really wanted to use natural language processing to be able to read a patient’s chart and then present me as a radiologist with a clinical summary of that patient so that I wouldn’t be recommending inappropriate follow-ups, to chase down stuff that had already been worked up and that I could really give the patients something that would be best for them.
Q: Give a few really practical use cases that can be applied with this combined converged platform.
A: Absolutely. Number one is the solving the needle in the haystack problem. Right now, if you go to a hospital and you say, “how many patients had appendicitis last year,” they might be able to get it from billing data, but we can do better than that. We can actually go into the reports, read and understand all those reports, and find the patients who just have that one condition. We can also look for patients with more than one. So much of medical research is looking for patients with condition ‘a’ and condition ‘b.’ We can easily do that. Show me all of the patients you got this drug, at this side effect, or whatever the case may be. Other use cases, we specifically find follow-up recommendations in radiology and other diagnostics reports. Nothing makes your heart sink more as a diagnostic radiologist than seeing a case where there’s now advanced disease and a follow-up had been recommended a year or two years ago that never happened. That’s one problem I really wanted to solve.
Q: What would you say to your fellow radiologists who are here, and you can see them across the way looking at really nice, fancy, sophisticated equipment, what would you say to them?
A: I’d say that I’m so glad that they’re now taking a much broader view of radiology in the context of the patient’s health. We’ve really moved from a focus on what can this modality do with photons, to what does a digital workflow look like, to why do we get up in the morning? We get up in the morning to help patients. We want to understand everything in the whole spectrum of quality and safety of care and that’s what I want all of my colleagues to focus on. Step back and think, does how I’m providing care pass the mother test? Would I want my mother to be treated like this?
Q: Final question for you both. Dr. O’Connell, what kind of advice would you give fellow radiologists if they want to implement a business intelligence and clinical intelligence platform? What are the steps they have to take in order to get to that place?
A: It’s easier than you think it is. Through the amazing work that Jeff and the Bialogics team have done they can capture all the information flowing through a radiology department in a really straightforward and easy manner, and boil down this huge amount of information into really easy to understand views. The best way to think about business intelligence is not something for the bean counters to use in the back end, but it’s really an easy button to help provide the best quality care. That’s the way to think about it.
Q: And Jeff, what kind of practical advice, or counsel, would you give to the doctors here who are interested?
A: To start right away because collecting data today will facilitate the adoption of AI in the future because AI will only be as good as the data that we can feed into it. By simply starting collecting the data today, even though it’s not fully used today, it becomes a database of real value for the future. There are simple ways to do it. There are ways of looking at the data in very simplistic, intuitive ways. Fundamentally, that’s the future and the value of radiology.
Q: Doctors often think about IT projects and they kind of put their head into their hands, but it doesn’t have to be that way, right?
A: No it’s a much simpler process than it has been in the past, but there are unique ways of collecting data and manipulating data that are different from company to company. You really do need someone who takes that holistic, big-picture view, regardless of the type of equipment that’s sold, or the type of information systems that they’re using, to sit on top of all of those and collect the data from a large perspective.