Session decries lack of integration for workstation features

Reprising the old joke about how many people it takes to change a lightbulb, radiologists were asked Friday how many workstations it takes to solve a given clinical question. Their answer: way too many.

In a special presentation at SIIM organized by Drs. Khan M. Siddiqui and Eliot L. Siegel, both of the VA Maryland Health Care System, clinical challenges from the ER, cardiology, and nuclear medicine were placed into a PACS and workstation environment while two radiologists and one cardiologist worked through the interpretation process. They moved hither and yon through a bank of workstation monitors as they resolved clinical questions. In the end they got their answers, but they suggested that the process involved far more steps than it should have. Efficiency, and potentially quality of care, may have suffered.

"Every single action that happened required these clinicians and radiologists to go through multiple workstations," Siddiqui said after the presentation. "It's not an easy workload to handle."

SIIM and its predecessor (SCAR) have focused on steps to ease the growing workload for radiologists as the number of scans, and the amount of data produced by those scans, continues to increase.

The first case involved a man who was shot in the buttocks and ended up with a bullet lodged in his upper lung. A head-to-pelvis CT scan produced cross-sectional images showing injuries that traced the path of the bullet through the liver and into the lung. Still, the radiologist, Dr. Kathirkamanthan Shanmuganathan of the University of Maryland School of Medicine, was having no luck explaining to the ER surgeon, played by Siegel, what had happed until he moved to a workstation with volumetric imaging capabilities. The nature of the injuries then became clear to the ER surgeon and he realized that he had better head back to the surgical suite to attend to the patient.

Shanmuganathan told Siegel that profuse bleeding in the liver and the lung may require the attention of two surgeons simultaneously.

"It was a good day in shock trauma," he said.

Another demonstration showed a complex set of images in cardiology requiring multiple modalities and workstations and demonstrated by Dr. Anwer Quershi, a cardiologists at the Geisinger Medical Center in Wilkes-Barre, PA. A third demonstration showed a pair of nuclear medicine studies (one myocardial perfusion, one PET/CT) also requiring multiple workstations for an accurate diagnosis, this time demonstrated by Siegel.

"Traditional thought is that clinicians don't need anything advanced. But as you can see, clinicians are advanced and sophisticated and want to use the tools radiologists have and maybe more," Siddiqui said. As for nuclear medicine, the RSNA's Integrating the Healthcare Enterprise project released compatability standards for nuclear medicine three years ago, but no one is adapting them.

"The purpose of the session was to highlight for the users as well as the vendors the need to take steps to better integrate data so that life is easier for all radiologists," Siddiqui said. "The volume of imaging is going up, the number of images is going up. It's becoming a real problem for us."

The data sets used in the demonstration were made available to all vendors who exhibited PACS and workstations at the meeting. Workstations used in the demonstration were supplied by Agfa, GE, Philips, and TeraRecon (thin-client and dedicated).