Acquisition speed, post-processing automation and ever-increasing applications all combine to make efficient multidetector CT workflow imperative. Training, high-functioning yet easy to use workstations, a plethora of protocols and more help them make the most of the modality.
Efficient multidetector CT workflow requires a good look at who is doing the post-processing and where the data are visualized, says Pedro Diaz, PhD, physicist at MetroHealth Medical Center in Cleveland, Ohio, and assistant professor at Case Western Reserve University. The facility has been using Philips Healthcare’s new 256-slice Brilliance i CT scanner since last fall.
“It’s been amazing to see the speed and image quality we’re getting,” he says. “The radiologists are thrilled with the images.” The scanner’s first task was detecting a renal stone, which took 4.2 seconds. With that kind of speed, some workflow changes are in order.
One of the biggest workflow challenges is the interpretation process, particularly in the trauma setting. Since the scanner generates data faster, the trauma and emergency medicine doctors want the radiologists to provide their interpretations faster. “Sometimes it’s a challenge to get that data out to PACS or a workstation,” Diaz says. “The network takes longer [to send the data] than the scanner itself [to do the study].” However, with increasing use of 3D viewing and sagittal and coronal axes, radiologists can go through the body in a more efficient fashion.
Since the radiologist is the most expensive piece of an interpretation, “our job is to handle this data in the manner that makes the radiologist the most efficient.” To do so, the technologists who work with Diaz do as much of the data reprocessing as possible. If a tech can perform 80 percent of the processing, then leave a bookmark in the portal, everybody with access to the data can benefit from having that work already done. Thin-client advanced visualization applications really help, he says. “We send all this data to the portal server and then multiple people immediately have access to this data and, more importantly, have access to the processing tools.”
Elliot Fishman, MD, director of diagnostic imaging and body CT at The Johns Hopkins University School of Medicine in Baltimore, Md., uses the Somatom Definition dual-source 64-slice CT scanner from Siemens Medical Solutions. Multidetector CT scanning demands better workflow, he says, because “when you go to new scanners, your possible applications change.” For example, coronary CT, virtual colonoscopy and CT angiography all increase.
Protocols and automation
The other new consideration is the speed of studies. With 10- to 15-second scans and reconstruction done almost in real time, “the speed of imaging patients is not determined by the scanner, but by workflow.” That includes, Fishman says, the design of proper protocols and taking advantage of background automation.
About 10 percent of Fishman’s volume is 3D imaging. “For a cost-effective environment, you need to make sure the equipment is used well,” he says. That includes learning the software to analyze those datasets, such as fly-throughs for virtual colonoscopies and tools specific to cardiac imaging studies. Fishman also recommends purchasing a scanner that is easy to use and easy to train new people to use. With the right scanner, “complicated applications really aren’t that complicated to do.”
Another step Fishman took was meeting with referring physicians to teach them what he could do with the multidetector scanner. He makes prints of all his 3D studies and delivers them to referring doctors.
Thinking in 3D
Frederick Barnett, MD, director of radiology services at Metroplex Hospital in Killeen, Texas, and director of cardiac CT angiography services at Doctors Hospital at Renaissance in Edinburgh, Texas, has been using the Aquilion 64-slice CT system from Toshiba America Medical Systems for more than two years.
A move to more invasive cardiology services meant that “it made sense to get a scanner that would let us do noninvasive coronary artery disease screening.”
Barnett went from a single-slice, helical scanner to the Aquilion. That forced him to start thinking in 3D mode—“an entirely different concept.”
Automation is an important part of that different concept. Once the scanner generates the data, the computer takes the raw data, reconstructs it into axial scans and from those axial images, performs reformats. All