Informatics: The Cause of - and Solution to - Radiology’s Problems

November 27, 2012

CHICAGO - Radiology is woefully behind the times in its use of informatics, said Paul Chang, MD, who detailed how systems can improve efficiency and accuracy.

CHICAGO - Paul Chang, MD, has a lot of titles. He’s a professor of radiology at the University of Chicago. He’s the vice chair of Rad Informatics. He’s one of the presenters of the Eugene P. Pendergrass New Horizons Lecture at RSNA 2012.

But his father, a retired radiologist, has a different title for Chang: The Man who Ruined Radiology.

That’s because Chang is an informatics guy, a PACs person. Chang said his dad told him, “Before PACS, we were the doctor’s doctor. Medicine and surgery rounds started in radiology. ... Every morning the clinicians and the radiologists collaborated.”

Now? Radiology is fast becoming an ancillary service. And Chang admits that his people - the IT guys, the informatics fanatics - helped cause the problem. “The constraints of the immaturity of these systems … facilitates pigeonholing us into commodity level service,” he said.

But Chang said new technology can set the specialty free and put radiologists back at the center of the medical universe. “We should be embracing IT to virtually collaborate and be everywhere,” he said. “In every OR, in every ICU, in every clinic and [in collaborating] with our patients.”

At the University of Chicago, Chang has helped implement this is a couple of ways. One pilot program applies workflow improvements to scanning equipment, treating a modality as an IT device. “The CT scanner programs itself,” he said. “The injector programs itself. After the scan is done, it knows about all the post-acquisition workflow requirements - sending stuff to PACs and 3D work stations, notifying transport.”

That pilot paid off across the enterprise. “These efficiencies are realized as improvements not only in radiology, but also length of stay, hospitalization, throughput in the clinics. We were able to realize about a 66- to 70-percent improvement in cycle time in our CT scanner by automating procedures that originally were done by humans,” he said.

Another program automates the process of lesion tracking for oncology. Deep integration automatically identifies prior lesions and associates them with the new ones. That system reduced the time it took radiologists to do these measurements by 50 percent. “More importantly,” Chang said, “it improved efficiency and accuracy downstream for the oncologist as well.”

And that just scratches the surface of what new technology can do, because radiology itself is woefully behind the times. “My kids have 10 ways to talk to each other, from Skype, Twitter, Facebook, Foursquare… and they can articulate intelligently why they need all of these,” Chang said. “And yet, as a radiologist, I basically have the report and a phone call. We have to do better.”

How much of a difference can it really make? Well, think about how you interact with the Web at home. “We expect a certain level of deep, sophisticated, elegant integration when we’re buying music or watching the web, and yet when we go to the hospital we live with static reports,” Chang said.

Instead, consider the potential of multimedia reports that use hyperlinks to give clinicians - and patients - more information. “Most of our physicians are going to consume reports electronically anyway,” Chang said. “Instead of saying I used this much of contrast, have it as a hyperlink.”

If you’re worried about increasing your workload, Chang wants you to relax. “You can actually create these multimedia reports faster than you can create a static report. Value is optimized, quality as well as efficiency.”

“Technically, this is easy,” he continued. “The real challenge is changing ourselves.”