Going with the Flow: Integrating Multislice CT Into Your Workflow

High-speed multidetector CT scanners are a marvel of technology, producing unbelievably detailed images of anatomic structures, yet they have created some large challenges in workflow for busy radiology and cardiology practices. Users on the cutting edge have utilized the best solutions the manufacturers have developed and devised methods to insure appropriate image renderings that facilitate clinical care and enable efficient image management and storage. A team approach of highly skilled imaging professionals facilitates the entire process.

Image management



The imaging chain that includes acquisition to post processing and interpretation to storage and archiving raises issues that must be addressed.

Jason Cole, MD, MSc, a cardiologist with Cardiology Associates in Mobile, Ala., has been using the GE Healthcare LightSpeed VCT 64-slice scanner since April to accomplish a total of 500 to 600 scans, 90 percent of which are cardiac studies. They use beta-blockers for those patients who need them, and then run three quick scans, one for calcium scoring, one test bolus and then one with contrast.

Their highly experienced technologist processes the raw data and then creates formatted images of the coronary arteries. They have customized their workstation screen so that Cole can read the raw axial data for an overview, and then perform an integrated reading that incorporates both the volumetric and cross-sectional images.

"The images that provide the ultimate diagnosis are MIP (Maximum Intensity Pixel) images," says Cole. He reviews two-dimensional MIP images with a certain number of slices that the computer formats into the best longitudinal section for each area along the vessel. He correlates the cross-sectional view with the longitudinal view. "The longitudinal view is the one that most cardiologists would be most comfortable with because it's the closest approximation of what we would be looking at in a cath lab."

They store the raw data as well as any formatted views of individual vessels in their archive for future needs. Additionally, they save a "snapshot" image of either the curved reformatted or the MIP data or any portion of the exam that has raised concern.

Because the GE workstation provides the capability for color mapping, Cole finds benefit in setting the color code to soft plaque, calcified plaque and contrast to help identify suspicious lesions. The color diagram of the entire vessel provides the curved reformatted images.

Cole explains that in reading a study there are literally thousands of images he might review for each patient. As he looks at curved reformatted views and rotates through 180 degrees tracing down the longitudinal study of the vessel, there would be no way to save all of those individual images. That's why the "snapshot" version serves to save the specific areas of interest that raised concerns.

Once the imaging exam is completed, a standardized reporting system is used to document findings of either soft plaque or potentially obstructive plaque in each section of every major vessel. In addition, typically they record a paragraph that summarizes all of the findings that is added to the standardized reporting form.

Automation and integration


Pedro J. Diaz, PhD, vice chairman of radiology at MetroHealth Medical Center in Cleveland, describes the use of their 64-channel system to accomplish a variety of imaging studies. They have worked to integrate their Philips Medical Systems Brilliance CT and their disparate information systems, including an electronic medical record (EMR), PACS and hospital information system and radiology information system (which are all from different vendors) by taking advantage of the open architecture Philips offers as native inside their CT application.

In the past, the CT scanner was just a device that you set up for scanning to acquire images, explains Diaz. As they entered the digital environment where the technologist might need to download patient information from the HIS/RIS, review lab results from the EMR and plan the imaging study, integration became a key element in managing all of these systems.
He says that their Brilliance Workspace has incorporated the capabilities of a remote workstation into the scanner itself. Automatic reconstructions of image data sets produce additional views.

"If we know we're doing a spine protocol and we need to generate sagittal slices from the axial data, we can program that up front," Diaz explains. When the technologist employs the MPR auto-launch feature and produces a set of axial slices, those data sets get passed automatically into a reconstruction application right at the scanner, which then creates the additional 30 to 50 sagittal slices.

Besides those capabilities at the scanner, they also can program shipment of the image data set to a remote Brilliance workstation where the 3D technologist develops high-end volume resolution images. This means that they do not tie up their integrated scanner cockpit monitor to accomplish these tasks that may take 5 to 15 minutes.

"The workflow in the Philips software allows us to handle on a protocol-specific or study-specific basis the images we need to generate for the radiologists or referring physicians," says Diaz. "At the core of all of this is the Philips software with automated protocols to let us do our work efficiently."

One of the other features they appreciate is the ability to capture one large study, for example a chest/abdomen/pelvis, but they can segment the specific portions and send them to different clinicians. Because they are a major trauma center, this feature proves valuable when several specialists need to read only one portion of the scan. They are able to perform the scan, develop the correct segmentation and reconstruction, and attach appropriate billing and accession numbers assigned to each segment.

"We can program the scanner up front to take the three orders and assign them to be chest slices, abdominal slices and pelvis slices and send them to PACS as three different acquisitions if you want to," says Diaz, who notes that patient care has benefited from this capability.

Flexibility enabled


Cleveland Clinic Foundation is making good use of 64-slice CT, according to Stacie Kuzmiak, RT(R)(CT), cardiovascular CT technologist. They use their Siemens Medical Solutions Sensation 64 for a variety of imaging studies. She says that with conventional workflow, they would scan the patient and then do the reconstruction and reformation before reading the image. But with this scanner, they are able to eliminate additional reconstruction to produce 3D images on the navigator at the scanner.

"With multidetector CT, there is so much flexibility for the technologist," Kuzmiak says. For example, if he or she scanned the entire abdomen and pelvis in three-millimeter slices, and then determined that higher resolution was needed through the pancreas, he or she can create 1mm slices through that area. "So you could have a routine abdomen/pelvis with 150 images and an additional 100 images that are 1mm thick through the area of interest," she says. They can accomplish those goals either by setting those parameters prior to scanning or after the scan has been completed.

With reconstructions of the images, they will generally accomplish those tasks at the scanner. If they need uninterrupted workflow, they can reconstruct the image in the background and continue scanning, using WorkStream 4D on their Leonardo workstation. Kuzmiak says they will soon be implementing a Cosmos PACS from Siemens to integrate the RIS, PACS and post-processing functionality.

Burd Schoener, RT(R)(CT), manager of the CT scanning department at the University of Michigan Medical Center in Ann Arbor, says that they have been managing imaging studies from their GE VCT 64-slice scanner since April. Although they usually acquire images in the axial plane, when they reconstruct images he says it is almost impossible to detect whether they acquired the data in axial, sagittal or coronal views.

As their technologists reformat images, some through automatic function and some manual, they are responsible for presenting the radiologist with images that will prove useful in diagnosis.

The technologists who work in the 3D lab have become very specialized, and they have to know their anatomy "stone cold," he says. Because data manipulation could require a fair amount of time, the technologists are charged with setting up the images in appropriate configurations so that the radiologist does not need to do much reformatting.

"The scan times are very quick now. Four or five years ago, a typical chest CT would take about 5 minutes to acquire all of the images with multiple breath holds," Schoener says. "We can do a chest CT in one breath hold now, so the timing of delivering IV contrast is unbelievably critical." If contrast is administered at the wrong time, the entire study could prove worthless.
Typically they acquire images in thin slices and then reconstruct them into thicker slices for reading. He says that they used to acquire thick and reconstruct thin, but that process is reversed with these new machines.

Describing their deep archive as "huge," Schoener says that they usually send only the reconstructed images to the archive, hold onto the raw data for a short period of time, and then discard it. This helps them manage the amount of traffic on their network, and the amount of space they require in their archive. They employ a layered archive configuration with deep archive for long-term storage and another layer for short-term archive and more rapid accessibility. Their RIS pre-fetches prior studies when they are needed for outpatient visits.

Christi Newton-Foster, RT(R)(CT), lead technologist at the Mayo Clinic Jacksonville in Florida, describes their outpatient practice where they average 150 studies per day using a Siemens Volume Zoom 4 slice, Siemens Cardiac 16 slice and GE LightSpeed 4 slice scanners.

For some of their routine, straightforward imaging they use a slice thickness of 5 mm, which means the volume of those studies is not great. They are using a Siemens PACS with images sent from the scanners to the MagicView where they are prepared for the radiologists to read electronically. All of the CT scanners' images are sent to a stand-alone MagicView.

"On these multislices [units] you can use a collimator and scan at a thicker image, but if you need to go back and retro smaller, you have the ability to do that," explains Newton-Foster.

Robert Burns, BS, RT(R), technical coordinator for the South Nassau Communities Hospital in Oceanside, N.Y., explains their use of the Siemens Somatom Volume Zoom 4 slice scanner in their main radiology department. Because when they scan they only see every fourth image, they must reconstruct the remaining images.

They do not store the raw data that are produced, and they are saving their reconstructed images on optical disk. Their radiologists read studies from film. They anticipate installing a PACS within a year or two.

In addition to their Volume Zoom, they have a Siemens Sensation 10-slice scanner in their level 2 trauma center emergency department.

Teamwork is vital


Given the complexity of these studies, and the need for efficient and effective functioning, the team that runs these multislice scanners must be extremely capable.

Cole of Cardiology Associates, who is using the LightSpeed VCT scanner for cardiac studies, emphasizes that while they currently have one extremely experienced technologist who is both scanning and processing data, they are in the process of hiring a second technologist who will be responsible for completing the scans while their primary technologist continues processing the images. In addition, they have two nurses who work in the preparation area and a radiologist who "over-reads" all of the non-cardiac aspects of studies.

"In our studies, heart rates are extremely important, so all of our patients are pre-treated with oral beta blockers," he says. "If their heart rate remains elevated, anything over 60, our nurses give IV beta blockers and sometimes sublingual nitroglycerine." When all proceeds smoothly, they generally scan three patients an hour, but that schedule can be disrupted when something goes awry.

In addition, he notes that their current technologist is extremely capable and produces excellent images for his review. In one of the local hospitals with technologists who were not so experienced, they found that fact made a tremendous difference from a physician workflow perspective.

Brett Milner, CT interventional supervisor at MetroHealth Medical Center in Cleveland relates that he has two technologists running their Philips Brilliance CT 64-slice scanner.

"One technologist is doing everything on the scanning side, and because of all of the integration of programs on the scanner, he or she doesn't have to leave their chair to accomplish scanning, billing and looking up medical records," he says. "The second technologist runs the patient side of the room to get the patient on the table, start the IV and position the patient."

Conclusion


High-speed multislice CT scanners have proven extremely valuable for numerous aspects of patient care. But like a thoroughbred horse that requires taming and careful management, the personnel who run CT scanning must give careful consideration to how to manage the image data sets from acquisition through storage and archival and all of the steps in between. Highly skilled and experienced professionals and equipment that is designed to address the challenges these machines create are absolutely critical to success.

Around the web

The new technology shows early potential to make a significant impact on imaging workflows and patient care. 

Richard Heller III, MD, RSNA board member and senior VP of policy at Radiology Partners, offers an overview of policies in Congress that are directly impacting imaging.
 

The two companies aim to improve patient access to high-quality MRI scans by combining their artificial intelligence capabilities.