Medical Delta Program ‘From man to machine – early identification of rheumatoid arthritis’

Rheumatoid arthritis (RA) is one of the most common autoimmune diseases and causes chronic inflammation of the joints. People with RA experience pain; the disease burden and societal costs due to work incapacity are high. Early diagnosis can help people with RA enter the treatment process sooner, thereby reducing their symptoms.

Early detection of joint inflammation is currently only possible in hospitals (the "second line") through a physical joint examination by rheumatologists. This is inefficient: expensive manpower is required, and waiting lists delay the diagnosis. The Medical Delta Program 'From man to machine – early identification of rheumatoid arthritis' seeks to find a way to use technology to achieve a more efficient and faster RA diagnosis. Although this screening approach has proven to be efficient, it still requires medical specialist manpower and is not future-proof.

The so-called 'Early Arthritis Recognition Clinic' has been established at the LUMC. When a general practitioner suspects that someone has arthritis, that person is examined by a rheumatologist in five minutes through a joint examination to determine the presence or absence of joint inflammation. This is followed by either a referral back to the general practitioner or further rheumatological analysis. Although this screening approach has proven to be efficient, it still requires medical specialist manpower and is not future-proof.

Smart technology

The Medical Delta Program 'From man to machine – early identification of rheumatoid arthritis' is researching how smart technology can help people with inflammatory arthritis receive a faster diagnosis with less involvement of doctors and at lower costs.

With a short MRI scan, it is possible to identify specific joint inflammations. By developing machine learning and AI algorithms, a quick diagnosis can be made with less involvement of healthcare personnel.

Pilot studies have already shown that these AI techniques can accurately detect joint inflammations in patients with confirmed diagnoses. However, as a screening tool, the accuracy of MRI still needs to be investigated. In preparation for this, the program aims to address the practical and technical limitations in implementation. To achieve this, the program involves all stakeholders—from general practitioners to radiology technicians.

Goals

The overarching goal is to replace expensive and increasingly scarce clinicians with the use of MRI and AI technology. A significant obstacle to this is image interpretation: if more MRI scans are taken for rheumatology diagnostics, radiologists would be further burdened. Since the trained eye of a radiologist is not yet directly replaceable, the program is exploring ways to:

• develop a machine learning methodology using MRI data so that computers can interpret MRI scans, and integrate an initial version into commonly used radiological software;
• adapt the triage process from primary (e.g., general practitioners) and secondary care (e.g., rheumatologists, radiologists, radiology technicians) to this technology, so that technology and clinical practice can complement each other.

Contact

For more information or if you're interested in participating, please contact one of our innovation managers.