Detects osteoarthritis many years before the onset of artificial intelligence

Pittsburgh, September 21, 2020-Researchers at the University of Pittsburgh School of Medicine and Carnegie Mellon University have created a machine learning algorithm that can detect subtle signs of osteoarthritis. Radiologist-MRI scan that takes years before symptoms begin These results will be published on PNAS this week.

This predictive approach allows patients to be treated with prophylactic drugs one day instead of undergoing joint replacement surgery.

“The gold standard for diagnosing arthritis is x-rays. As cartilage deteriorates, the distance between bones narrows,” said co-author Pitt’s orthopedic associate professor Kenneth Urish. Said UPMC Magee-Director of the Bone Joint Center at Womens Hospital. “The problem is that if you see arthritis on x-rays, the damage has already been done. Preventing the cartilage from falling apart is much easier than re-growing the cartilage.

Currently, the main treatment for osteoarthritis is joint replacement. Also, the condition is so prevalent that knee replacement is the most common in the United States for people over the age of 45.

In this study, researchers examined knee MRI of the osteoarthritis initiative. They focused on a subset of patients with little evidence of cartilage damage at the beginning of the study.

Looking back, we found out which of these participants developed arthritis and which did not. Computers can use that information to learn subtle patterns of MRI scans of presymptomatic people who predict future risk of osteoarthritis.

“When a doctor looks at an image of these cartilage, there is no pattern that pops out with the naked eye, but that does not mean that there is no pattern there. It just means that you cannot see it using conventional tools. The lead author, Dr. Shinji Nikundu, said he completed the project as part of the Pit Medical Scientist Training Program and the graduate training in Carnegie Mellon Biomedical Engineering.

To validate this approach, Kundu, now a resident physician and medical researcher at the Johns Hopkins Department of Radiology, trained the model on a subset of knee MRI data and tested it in a patient never seen before. .. Kundu did this dozens of times, with different participants each time, to test the algorithm on all the data.

Overall, the algorithm predicted osteoarthritis with 78% accuracy from MRI performed 3 years before the onset of symptoms.

Currently, there are no drugs that can prevent presymptomatic osteoarthritis from developing into serious joint deterioration, but very effective drugs that can prevent the development of rheumatoid arthritis, a condition associated with patients. There are several.

The goal is to develop the same type of drug for osteoarthritis. Some candidates are already in the preclinical stage.

“Instead of recruiting 10,000 people and following them for 10 years, we only need to enroll 50 people who are known to have osteoarthritis in 2-5 years,” Urish said. “Then we can give them drugs and see if it stops the progression of the disease.”

###

This study was funded by the National Institute of Arthritis and Musculoskeletal and Skin Diseases (Grant K08-AR071494).

Additional authors include Beth Ashinsky of Drexel University. Dr. Mustapha Bouhrara of the National Institute on Aging and Dr. Richard Spencer of Medicine. Dr. Eric Dam, University of Copenhagen Doctor of Medicine, Shadpool Demefri, Johns Hopkins University. And Dr. Mohammad Sifat E. Rabbi and Dr. Gustavo Rohde of the University of Virginia.

To read and share this release online, please visit: https: //www.upmc.When/media/news/092120-Kundu-Urish-OA [when embargo lifts]..

As one of the nation’s leading academic centers for biomedical research, the University of Pittsburgh School of Medicine has harnessed the power of new knowledge to engage in advanced technology and a wide range of sciences in its ongoing search to improve human condition. It integrates basic science. Pitt is primarily driven by the School of Medicine and its affiliates and has been ranked in the top 10 in funding from the National Institutes of Health since 1998. In a ranking recently released by the National Science Foundation, Pitt ranks fifth among all federal universities in the United States. Science and Engineering R & D Support.

Similarly, the School of Medicine aims to improve the quality and strength of medical and graduate education programs recognized as innovative leaders, and to engage highly skilled and caring clinicians and creative scientists. World-class research that is equally committed to training sufficient abilities. The School of Medicine is UPMC’s academic partner who has worked with universities to raise the bar for medical excellence in Pittsburgh and to position healthcare as a driving force in the local economy. For more information on the School of Medicine, see: http: // www.Medical college.pit.edu..

Carnegie Mellon University’s Faculty of Engineering is a prestigious faculty of engineering known for its focus on interdisciplinary collaborative research. Colleges are renowned for addressing scientifically and practically important issues. Our “maker” culture is rooted in everything we do and leads to new approaches and transformative results. Our highly regarded faculty focuses on innovative management and engineering, producing innovative results that drive the intellectual and economic vitality of communities, countries and the world.

http: // www.upmc.When/media

Contact: Erin Hair

Mobile: 412-738-1097

Email: [email protected]

Contact: Amy Charley

Mobile: 412-738-3511

Email: [email protected]