Advancing the speed and accuracy of radiologist workflow

May 19, 2021 -- The world of radiology is changing rapidly, but one constant remains: Providing timely, accurate reads is critical to optimizing patient outcomes. Today, meeting those goals means relying on technology, which is increasingly being driven by artificial intelligence (AI).

It all starts with the worklist. Improving the worklist can improve the workflow and lead to greater reading productivity, clearing time for more consultations and patient interactions, as well as open space for additional imaging. In fact, one study found a nearly 30-minute decrease in total interpretation time across all readers of MRI scans when using an analytics-driven worklist.1 In a 100-radiologist practice, that translates to 50 saved hours a day.

"Given the huge volume of exams we deal with today, knowing which are most important and which require a quick intervention is critical," said Dr. David Mendelson, a radiologist at Mount Sinai in New York City. Today, he and his team use GE Healthcare's Centricity Universal Viewer with Autoserve, designed to reimagine the traditional workflow to ensure the right radiologist reads the right scan at the right time.

"Using artificial intelligence-integrated technology has made a dramatic change in the way a worklist works for us," Mendelson said.

A significant factor inhibiting efficient turnaround time is appropriate study prioritization. Unless a radiologist knows how important a study is or the circumstances around a patient
A significant factor inhibiting efficient turnaround time is appropriate study prioritization. Unless a radiologist knows how important a study is or the circumstances around a patient's condition or care, he or she can't produce a report in the timeliest fashion. This can be especially difficult during peak times when everything appears to be urgent.

Mendelson and his colleagues use algorithms from a couple of vendors, and what they have learned is that when AI notification mechanisms are used, such as those built into the GE Healthcare PACS, those cases are not only looked at sooner but perhaps more importantly, there are times when the AI algorithm identified findings that were not noted by a radiologist. Being notified of a discrepancy has been useful in making patient care decisions.

Another element to consider when designing reading workflows is whether to utilize case assignment automation. In any of the modern enterprises, there are quite a few radiologists with many different roles, many different abilities.

"You don't want those radiologists to be thinking about a work list and what they should read next," explained Dr. Elizabeth Bergey, president and CEO of Quantum Imaging and Therapeutic Associates. "You want them in the zone thinking about the medical question, not the operational question."

Therefore, she considers workflow orchestration critical to Quantum's business to enable the highest level of productivity for her team, to ensure they are exceeding their service level agreements and meeting their quality goals, and to promote greater radiologist work satisfaction.

Hear Bergey's perspective in this 60 second video.

Improving efficiency in the pediatric setting

Referencing normal anatomy and its variants is critical in diagnostic radiology. This is particularly true in pediatric radiology, in which normal growth and development create a broader spectrum of normal appearances. However, reference example studies are not always easily accessible to radiologists in their daily practice.

"The biggest challenge in my field is trying to recognize what normal looks like," said Dr. Sanjay Prabhu, pediatric neuroradiologist at Boston Children's Hospital. "We see a huge gamut of patients, from fetal life to 18 years, when the body is developing at a rapid pace."

Having the ability to load normal images from our institutions
"Having the ability to load normal images from our institutions' own scanners with specific parameters like the ones we scan patients with is a game changer!" -- Dr. Sanjay Prabhu, pediatric neuroradiologist, Boston Children's Hospital.

The challenge is even greater for radiologists who don't read large numbers of pediatric scans, he said.

Which is where GE Healthcare's Integrated Reference Study Library comes in. It loads matched neuroradiology MR reference sets directly into the Centricity Universal Viewer, enabling radiologists to read pediatric scans with greater speed and accuracy.

"This capability can help inform a more accurate study interpretation of childhood brain abnormalities," Prabhu said. "It can help distinguish potential brain pathology from normal developmental changes."

Hear Dr. Prabhu's perspective in this 60 second video.

"Using these integrated data sets has made our reads faster, more accurate, and our radiologists more confident in making the diagnosis," Dr. Prabhu said.

To hear from your peers and other thought leaders on advances that are positively impacting radiologists, sign up to receive Better Radiology in 60 Seconds.

Looking forward

Today, new productivity tools such as intelligent worklists and AI-based triage2 are already demonstrating their promise to improve radiologist efficiency and accuracy. While some are incremental, such as rules-based case routing, others, such as AI-based clinical applications, are poised to revolutionize radiology just as dramatically as digitization did decades ago.

These new layers of artificial intelligence are allowing radiologists to spend less time performing manual tasks and more time interacting with other members of the care team, as well as with patients. In other words, to deliver care that is intelligently efficient.

To learn how GE Healthcare can help bring the power of AI into the radiologist workflow visit our website.

1 Wong TT, Kazam JK, Rasiej MJ. Effect of analytics-driven worklists on musculoskeletal MRI interpretation times in an academic setting. AJR Am J Roentgenol. 2019 Feb 26:1-5.

2 Based on AI algorithms regulatory clearance/approval.

Copyright © 2021

To read this and get access to all of the exclusive content on create a free account or sign-in now.

Member Sign In:
MemberID or Email Address:  
Do you have a password?
No, I want a free membership.
Yes, I have a password:  
Forgot your password?
Sign in using your social networking account:
Sign in using your social networking