Digital image workstations have progressed rapidly since first introduced to radiology more than 25 years ago, but they have yet to achieve the intuitive ease of use that characterized the light boxes they replaced, said presenters at an educational session on Saturday.
Digital image workstations have progressed rapidly since first introduced to radiology more than 25 years ago, but they have yet to achieve the intuitive ease of use that characterized the light boxes they replaced, said presenters at an educational session on Saturday.
Session speakers identified two general problems with workstations today: functions that exist in specialized settings and need to be easily available, and interface problems that stand in the way of effective use. Both problems tend to interrupt workflow and have kept radiologists from achieving additional efficiencies that workstations should provide, they said.
Dr. Steven C. Horii, a pioneer in the development of workstations, said there have been some notable successes with workstations. They have proved useful for interpreting cross-sectional images and projection radiographs and in teleradiology. But there are other common applications, including mammography, PET/CT fusion, and studies that combine 2D, 3D, and 4D images, that require specialized workstations.
When physicians must leave their general PACS workstations to perform these specialized tasks, the workflow is broken and efficiency sacrificed, Horii said.
Workstation problems also lurk outside the reading room. PACS are gaining ground in orthopedics, but most orthopedic surgeons still rely on film for surgical planning, Horii said. Medical images are increasingly being used outside the reading room, but often, referring physicians rely on technologically inferior computers and software that lack the power and sophistication to display images.
Images could be invaluable in the operating room, but calling them up and manipulating them is challenging in a sterile environment. Surgeons don't want to scrub up each time they touch a computer control, Horii said. One solution is to put a mouse in a sterile bag, but surgeons are still resistant because they want to concentrate on surgery and not work a mouse, he added.
Dr. David L. Weiss, head of imaging informatics at the Geisinger Health System in Danville, PA., found fault with workstations that fail to achieve full efficiency for radiologists. He noted that view boxes achieved a fairly high level of user efficiency before PACS led to the development of workstations. View boxes were intuitive with hanging protocols that could be customized, read, and dictated without radiologists taking their eyes off images. Tasks such as annotation, magnification, and measurements could be performed with tools readily at hand.
Try the same thing with a workstation and the task may not be quite as easy. A wax pencil for annotating film can be put down, but a workstation tool that performs the same function must be turned off after it is used. Weiss compared that to having the wax pencil stick to the radiologist's hand. And workstation magnification tools can sometimes be cumbersome enough that radiologists resort to hand magnifying glasses to accomplish the same task, he said.
Other issues involve the user interface. Too often workstations require users to take their eyes off the image to change settings--an enemy of efficiency, Weiss said. Players can use gaming systems and devices without taking their eyes off the screen and radiologists should be able to view images the same way.
There should be no icons to click and no keyword searches. Interfaces should be intuitive, and rely on two hands, 10 fingers, other body parts, and five senses, Weiss said.
Comparative AI Study Shows Merits of RapidAI LVO Software in Stroke Detection
February 6th 2025The Rapid LVO AI software detected 33 percent more cases of large vessel occlusion (LVO) on computed tomography angiography (CTA) than Viz LVO AI software, according to a new comparative study presented at the International Stroke Conference (ISC).
Study: Mammography AI Leads to 29 Percent Increase in Breast Cancer Detection
February 5th 2025Use of the mammography AI software had a nearly equivalent false positive rate as unassisted radiologist interpretation and resulted in a 44 percent reduction in screen reading workload, according to findings from a randomized controlled trial involving over 105,000 women.