FDA clears Radiobotics' osteoarthritis diagnosis algorithm

The U.S. Food and Drug Administration (FDA) has announced the 510(k) clearance of Radiobotics’ RBknee algorithm, an AI-based analysis tool aimed at improving the efficiency and accuracy of radiographic knee osteoarthritis diagnosis.

Osteoarthritis (OA) is a very common joint disease, also affecting knees and is a major cause of disability worldwide. 10% of the world’s population above 60 years, are estimated to have joint related problems that can be attributed to arthritis, according to the WHO Scientific Group on Rheumatic Disease. 

Radiography (x-ray analysis) is commonly used to support the diagnosis and track progression of the disease, and additionally the demand for medical imaging is increasing, while the healthcare system is experiencing a global shortage of radiologists. 

The osteoarthritis algorithm RBknee from Radiobotics works by analysing digital x-rays of knees and identifying common radiographic findings associated with the diagnosis of OA. RBknee is intended for use in an orthopaedic or a radiological setting, where the x-ray is read to determine the diagnosis and treatment of OA. By supporting clinicians with clinically validated AI technology, RBknee has the potential to save time and money as well as improving clinical outcomes through consistent, objective, and accurate measurements and analysis.

Ayse McCracken, chair of the board at Radiobotics, said: “If you look at the increase in demand for services for radiologists and the rate at which radiologists are burning out, we've got a huge problem. As the global population ages, the demand for health care services is only going to accelerate, and with the first product FDA cleared we’re now ready for the next step in our contribution to the digitisation of the health industry and I’m very proud of all the hard work the team has put into it.”

RBknee can identify presence or absence of osteophytes, subchondral sclerosis, joint space narrowing, and conclude on the presence or absence of OA by using proprietary machine learning algorithms. Furthermore, RBknee can accurately measure the joint space width in both compartments of the knee.

The software is approved based on rigorous testing and clinically validation of the algorithms, showing high sensitivity and specificity on the identification of findings, and a high precision of the measurements of the device when evaluated on an unseen dataset. 

Co-founder and president of Radiobotics, Stine Mølgaard Sørensen, said: “RBknee is our first product with FDA clearance accompanied with a European MDR certification earlier this year. This emphasises the potential of our technology and our AI based solutions continue to demonstrate quality and safety. We are ambitious on behalf of the development of the field of radiology and orthopaedics and the FDA clearance is important to both Radiobotics as well as patients in not only Europe but now also in the US.”