Rasu B. Shrestha, M.D., who holds the dual title of vice president, medical information technology, and medical director, interoperability and imaging informatics, at the 20-plus-hospital University of Pittsburgh Medical Center (UPMC) health system in Pittsburgh, Pa., has been a thought-leader in the imaging informatics world for years now. Indeed, Dr. Shrestha oversees one of the broadest and most complex imaging informatics constellations in any private health system in the U.S., one that serves the needs of more than 3,200 employed physicians and many thousands of additional affiliated physicians in western Pennsylvania.

Shrestha sat down with HCI Editor-in-Chief Mark Hagland on Wednesday, Nov. 28, to discuss his impressions of RSNA 2012, the annual conference of the Radiological Society of North America, being held this week at Chicago’s McCormick Place Convention Center. Below are excerpts from that interview.

Dr. Shrestha, what is your overall feeling about, and what are your overall impressions of, this year’s conference, compared to past RSNAs?

Well, the theme this year is patients first. And it’s a beautiful theme, one I’ve been promoting for years now. Is it where radiology is yet? No, but we need to get there. We’re still very image-centric. Before PACS, we had an AP lateral [radiograph] of the chest. Now, with PACS, we have intelligent hanging protocols on our flat-screen monitors, and we pick up our dictation devices, and diagnose the AP lateral of the chest. And yes, it’s the patient’s image, but we don’t have the patient’s story. And if as radiologists, our task is to support a surgeon who’s about to operate, and everyone’s relying on the radiologist, we can’t just go with “reason for exam: headache.” We can’t work in a vacuum, with no family history, no previous history of migraines the patient has had, no retinal funduscopic scan, no previous pathology reports, including a biopsy on the potential lesion; yet so often, all you have is, “reason for exam: headache.” It’s ridiculous. And we end up generating a report that’s more of a legal document, and is less than useful for referring physicians. So I love the theme of patients first; I want everyone to march to that theme; but we’re not there yet.

Rasu B. Shrestha, M.D.

Among the themes that have been emerging in the discussions I’ve been having with people at this year’s RSNA have been clinical decision support, in the context of, more CDS, less RBM [radiology benefit management]; the VNA [vendor-neutral archive]; diagnostic imaging analytics; and participating in the meaningful use process.

Yes, those are all important elements.

But how do we get beyond the PACS-centric focus [PACS: picture archiving and communications system], and vendors creating products and solutions for the healthcare of five years ago, rather than for the healthcare of the future?

We’re all blinded by buzzwords; that’s a real problem. What’s the theme of the day? The buzzword of the day? Mobility? VNA? Cloud something-or-other? Shouldn’t we actually instead be focused on the workflow? Everyone’s trying to be the first cloud-enabled-something-VNA-something-mobility-something. And I’m a big proponent of the VNA. But the VNA without workflow is useless; it’s just a bucket.

And aren’t the vendors still very “bucket”-focused?

They absolutely are. Here’s what’s happened: we’ve had non-PACS vendors, including EHR [electronic health record] vendors saying, we have got to create this layer on top of the existing technology; but then the PACS vendors panicked and said, look, we’ve got the VNA, too! And it caused confusion and was not patient-centric. So it’s important for us to focus on the workflow, not the technology. You may have storage needs in a very wide variety of specialties; so you need the VNA, the bucket. But you also need a robust and functional middleware layer that sits on top of the VNA, and then a workflow-centric application ecosystem that sits on top of the VNA and is tightly integrated into the applications that you call home. Your home may be your EMR, your PACS system, but this has to be tightly integrated. And above all, you have to look at what is needed to become patient-centric, and to enable the clinician workflow that can make that a reality.