Implementing 3D visualization without experiencing some downtime can be a challengeespecially, as the trend now is “to do everything at once,” according to Osman Ratib, MD, PhD, professor and vice chairman, Information Systems Department of Radiology, University of California Los Angeles, which means the UCLA laboratory has experienced additional pressure to get things done in time.

“We have to be creative about our workflow to do that, and so we are putting pressure on the vendors to provide tools that are convenient and higher performance and that can be used remotely or directly at the workstations,” Ratib says.

In the meantime, however, the laboratory has had to experiment with methods for streamlining workflow. At first, Ratib says, it seemed most efficient to take a centralized approach, and combine resources to maintain the laboratory.

“That approach was effectiveuntil it reached the point where we could not handle all the work because the volume was too high,” Ratib says. “So we have decentralized most of the equipment back to specific reading rooms where radiologists are working, and to imaging suites, where we do the 3D processing. We still do have some centralization for very advanced procedures, however.

“Now that volume exceeds 10-15 cases a day, we need tools to be available more widely,” Ratib says. “We need more workstations, we need to be able to do processing remotely, and we need to have more people trained to use the technology.”

The laboratory at Massachusetts General Hospital, Boston, likewise spent a lot of time working out the integration of 3D with all hospital and departmental procedures, protocols, and systems.

“We had to work out PACS routing and workflow, all of our RIS, billing, and compliance issues, reimbursement issues, budgetary issues, and staffing,” says Gordon J. Harris, PhD, director of the 3D Imaging Service at Massachusetts General Hospital’s Department of Radiology, at Harvard University in Boston, and director of the Radiology Computer-Aided Diagnostic Laboratory. “There was a tremendous amount of front-end work required.”

Once the front-end work is done, a 3D system can flow smoothly. At Stanford University, the workflow involves a system for flagging those examinations that require postprocessing with a 3D modifier in the PACS tables. That information is routed to a server in the 3D laboratory, which receives a page when the scan is completed.

“The source images are retrieved via network to the 3D workstation from PACS, and the technologist determines the correct protocol for each study,” Laura Logan,? laboratory manager, says. “After the postprocessing is finished, the technologist pushes the images back to PACS with the source images for review and dictation by a radiologist. At this point, the radiologist may tell the technician if more images are required.

“We then print the 3D images and send the hard copies to the referring physician,” Logan says. “The images are likewise permanently stored on PACS and the results of quantitation are stored in our databases for future reference.”

E. Finch