close
close

Christianluscher

Breaking: Beyond Headlines!

AI robot scanner performs as well as rheumatologists in assessing RA
aecifo

AI robot scanner performs as well as rheumatologists in assessing RA

WASHINGTON — A fully automated ultrasound system combined with artificial intelligence-based disease activity scoring, along with expert rheumatologists to jointly evaluate patients with rheumatoid arthritis (RA), new search found.

The system, made by a Danish company called ROPCAincludes an ultrasound system called ARTHUR (RA Ultrasound Robot) which interacts directly with the patient and scans 11 joints per hand and a neural network-based software system, DIANA (Diagnosis Aid Network for PR), which evaluates images and monitors activity of PR. .

photo of a woman whose wrist is scanned by an ultrasound machine called ARTHUR
A woman has her wrist scanned by an ultrasound machine called ARTHUR.

The combined system classifies the degree of RA according to the joint standards of the European Alliance of Associations of Rheumatology (EULAR) and Outcome Measures in Rheumatology (OMERACT) for the diagnosis of RA. It received CE marking in Europe in 2022 and is currently used in six rheumatology clinics in Denmark, Germany, Switzerland and Austria, with more to come, said ROPCA co-founder and chief medical officer, Søren A. Just, MD. Medscape Medical News.

“Automated systems could help rheumatologists detect and monitor arthritic diseases early. Systems can be placed or moved to areas where there is insufficient rheumatologic expertise,” Just said during a special end-of-session presentation at the American College of Rheumatology (ACR) 2024 Annual Meeting.

He said Medscape Medical News“Currently, there are so many people referred and fewer and fewer rheumatologists. So we have to think differently. We need good automated assistants. As a screening tool, the system can determine whether a person with hand pain has RA or simply osteoarthritis “and can also give the patient an immediate answer, instead of sometimes waiting up to 6 months to get the information.”

Just, who is also chief physician of the department of internal medicine at Odense University Hospital, in Odense, Denmark, said his department also uses the system to assess flares in patients with established RA. “They can have a blood sample taken. They are scanned by the robot and you can see if there is any pathological activity. But I think screening patients for joint pain is just the beginning.

Asked to comment, session moderator Gregory C. Gardner, MD, professor emeritus in the division of rheumatology at the University of Washington in Seattle and member of the ACR conference program committee, said Medscape Medical News “One of the reasons we chose to present this summary is that we are interested in the science at convergence. We really thought this was a potential way to move the field forward for rheumatologists.

Gardner said it’s a benefit that the patient can potentially take an ARTHUR exam with a DIANA report and have blood work done before a visit to the rheumatologist. “Doing these studies is really time-consuming for a human being, so if you automate them, it’s a step forward in terms of having the data available for the rheumatologist to view and use sequentially to track the condition of the patients.”

Introducing Just’s presentation, Gardner called it “the coolest summary of the meeting.”

DIANA and ARTHUR did at least as well as human rheumatologists

In the study, 30 RA patients underwent two examinations by ARTHUR, followed by an examination by a rheumatologist specializing in musculoskeletal ultrasound. The scans were sent to DIANA, who classified the images according to the overall OMERACT-EULAR synovitis score, just like the human rheumatologist.

A “ground truth” was established by another human expert who evaluated the images of ARTHUR and the other rheumatologist, blind to the scanning method. The image with the highest disease activity was considered “ground truth” and agreement with this was assessed for both individual methods.

photo of 2: A woman has the metacarpophalangeal joint of the index finger examined with the ARTHUR ultrasound machine
A woman has the metacarpophalangeal joint of the index finger examined with the ARTHUR ultrasound machine.

I just showed a video of a patient scanned by ARTHUR. The machine guided her verbally to remove her jewelry, apply the gel and place her hand on the screen under the scanner. ARTHUR’s arm moved over the patient’s hand, locating the best angles to take grayscale images, Doppler images, and Doppler videos. The analysis takes 15 to 20 minutes and the images are stored, just said.

The study patients had a mean age of 65 years, and 23 of the 30 were men. Their average disease duration was 11 years, and the average disease activity score in 28 joints using C-reactive protein was 3.86, indicating moderate disease. The majority (73%) of patients were taking disease-modifying antirheumatic medications and approximately one third were taking biologics. ARTHUR scanned a total of 660 joints and 564 scans were successful.

For repeatability between the two ARTHUR exams, the exact percentage agreement was 63% for synovial hypertrophy, 75% for Doppler activity, and 60% for the combination. The percentage of agreements concluded (within one point) was 93%, 94% and 92% respectively. Binary agreements as to whether the joint was healthy or diseased were 88%, 91%, and 85%, respectively.

At the common level, the exact percentage agreement of ARTHUR and DIANA with the ground truth was 49% for synovial hypertrophy, 63% for Doppler activity, and 48% together. Binary agreements between illness and health were 80%, 88%, and 78%, respectively.

Human rheumatologists have obtained very similar results. The percentage of exact agreement with the ground truth was 51% for synovial hypertrophy, 64% for Doppler activity and 50% for all cases. The percentages of agreements concluded were 94%, 94% and 92% respectively. And the binary agreements between sick and healthy were 83%, 91% and 80% respectively.

At the patient level (all joints combined), the binary disease assessment of ARTHUR and DIANA (health vs. disease) showed agreement with the ground truth of 87% for synovial hypertrophy, 83% for activity. Doppler and 87% combined. Here, rheumatologists had lower scores, at 53%, 67% and 60%, respectively.

“In this study, we believe that the accuracy of ARTHUR and DIANA was comparable to that of an experienced rheumatologist, both at the joint and patient level,” Just said.

Gardner pointed out another advantage of the system. “DIANA doesn’t get tired… With human reading, accuracy can change depending on the time of day or stress level… But with DIANA, you’ll get consistent information.”

I just mentioned that the German Arthritis Foundation recently put ARTHUR and DIANA on a bus and took them to cities that lacked a rheumatologist. Patients stood in line, answered a questionnaire, had blood drawn and received their scans. A rheumatologist on the bus then interpreted the data and consulted individuals about their RA risk. “During the last trip, we screened 800 patients in 6 days. So there are definitely possibilities here.

Just is co-owner of ROPCA. Gardner had no disclosures.

Miriam E. Tucker is a freelance journalist based in the Washington, DC area. She is a regular contributor to Medscape Medical News, and other work appears in the Washington Post, NPR’s Shots blog, and Diatribe. She’s on X: @MiriamETucker.