Elderly woman suffering from pain in knee. Tendon problems and Joint inflammation on dark background.
Credit: ipopba/Getty Images

Finnish researchers at the University of Jyväskylä and the Central Finland Health Care District have created an artificial intelligence (AI) tool that can detect early knee osteoarthritis with an accuracy that matched 87% of doctors’ diagnoses. The researchers noted that being able to diagnose osteoarthritis early can help prevent unnecessary additional examinations, treatments, and potentially knee joint replacement surgery.

The new AI tool is not currently included in diagnostic criteria, but could be considered by orthopedic specialists to consider as a potential tool for early diagnosis. The technology was developed as a part of the AI Hub Central Finland project, in the Digital Health Intelligence Lab at the University of Jyväskylä and uses an AI algorithm, using a neural network, that has been trained to recognize whether a specific feature, spiking in the tibial tubercles, is present in radiologic images.

“The aim of the project was to train the AI to recognize an early feature of osteoarthritis from an x-ray. Something that experienced doctors can visually distinguish from the image, but cannot be done automatically,” noted Anri Patron, the researcher who developed of the  new method. “Around 700 x-ray images were used in developing the AI model, after which the model was validated with around 200 x-ray images. The model managed to make an estimate of the spiking that was congruent with a doctors’ estimate in 87% of the cases, which is a promising result.”

According to Sami Äyrämö, head of the Digital Health Intelligence Laboratory at the University of Jyväskylä, there are multiple efforts underway worldwide to leverage AI as a diagnostic tool for early osteoarthritis. The goal is to allow AI to serve as a clinical aid that would make it possible for general practitioners capable of making this early diagnosis as opposed to them needing to make referrals to orthopedic specialists.

“Several AI models have previously been developed to detect knee osteoarthritis,” Äyrämö said. “These models can detect severe cases that would be easily detected by any specialists. However, the previously developed methods are not accurate enough to detect the early-stage manifestations. The method now being developed aims for, in particular, early detection from x-rays, for which there is a great need.”

Juha Paloneva CEO of the Central Finland Health Care district, which collaborated on the study, notes that early diagnosis has both patient and health economic benefits.

“If we can make the diagnosis in the early stages, we can avoid uncertainty and expensive examinations such as MRI scanning,” Paloneva, who is also a professor of surgery, noted. “In addition, the patient can be motivated to take the measures to slow down or even stop the progression of the symptomatic osteoarthritis. In the best possible scenario, the patient might even avoid joint replacement surgery.”

Also of Interest