A AI-based system has been able to successfully predict the onset of osteoarthritis solely from observations of a patient’s MRI scans. The kicker hear, is that the machine learning system was looking at images that were taken years before symptoms had begun. The study was published in PNAS (Proceedings of the National Academy of Sciences of the United States of America).
The system was created by researchers at Carnegie Mellon and The University of Pittsburgh School of Medicine and in Pittsburgh. The combined team hypothesized that machine learning could be used to visualize sensitive cartilage phenotypes that are signs of of a future osteoporosis progression.
The study authors mentioned, ‘Our approach combines optimal mass transport theory with statistical pattern recognition’. For this study, the research team examined knee MRIs from those who had been enrolled in the National Institutes of Health Osteoarthritis Initiative. The NIHOI initiative is a multicenter, prospective observational study of osteoarthritis of the knee They focused on a subset around of 86 patients who had had little or no evidence of damage to the cartilage at the beginning of the study.
The research team reviewed many knee MRIs in the study. The researchers looked at the knees of 86 individuals that had little to no evidence of damage to their knee-cartilage. The patients were those who had been enrolled in the National Institutes of Health Osteoarthritis Initiative.
The AI-based algorithm was able to positively identify osteoarthritis in over 78% of those that were examined in the study. This level of accuracy was achieved exclusively using MRI scans that were taken up to three years before the first signs of symptoms. The machine learning system was able to detect signs of the disease that are far too subtle for radiologists to notice/.
At the moment, there isn’t any medicine on the market that can prevent getting presymptomatic osteoarthritis from developing into its more serious counterpart, joint deterioration. The good news is that there are, however, drugs that prevent people from contracting a related condition, rheumatoid arthritis.
In press release, the team of researchers said that their goal is to develop the same types of drugs for osteoarthritis, which would drastically reduce the need for more invasive osteoarthritis treatments.
Instead of recruiting 10,000 people and following them for 10 years, we can just enroll 50 people whom we know are going to be getting osteoarthritis in two or five years. Then we can give them the experimental drug and see whether it stops the disease from developing.
Kenneth Urish, MD, Ph.D., associate medical director of the bone and joint center at the University of Pittsburgh Medical Center (UPMC) Magee-Women’s Hospital.
