Core – Wearable AI for Rheumatoid Arthritis Management (WARAM)

Rheumatoid Arthritis (RA) is a chronic autoimmune disease caused by the body’s immune system attacking its tissues resulting in inflammation of joints (e.g., fingers, toes) characterized by swellings, pain, warmth, redness and stiffness, and general body fatigue. It takes a toll on the mental health of patients with some patients reporting feeling “helpless” and “hitting bottom” from the pain. RA is pervasive worldwide, affecting about 0.5-1% of the population worldwide and 70,000 people in Switzerland. It is managed through oral or injection medication which controls the inflammations.

Though RA patients take medication, they frequently have disease ‘flares’ which correspond to a worsening in disease activity level characterized by several days of severe RA symptoms. If the disease activity is not managed well, it could lead to worsening of symptoms, deterioration of joints, reduced mobility and bodily functions of the affected joints, and frequent hospital visits resulting in increased health care costs, loss of productivity, and overall negative impact on the quality of life of patients. Annual RA-related medical costs of patients with poorly controlled disease activity (non-remission) were higher than the remission cohort ($10,002 vs.  $8,594). Yet, the current standard of care is a hospital appointment every 3 to 6 months with a physical examination of joints, blood tests, and a change in medication prescription if needed. In between appointments, clinicians are completely in the dark about the experience of RA patients and there is no extra attention when flares occur. 

Meanwhile, it is recognized that behavioral patterns like limited physical activity, poor sleep, and stress are linked with these flares. Healthy behaviors such as medication adherence, sufficient physical activity, adequate sleep, and stress reduction are key to preventing these flares and improving the management of RA. Given the behavioral aspects of the disease’s symptoms, mobile and wearable technology could be leveraged to track and use them to monitor and predict flares even before they happen. Doing so could give the opportunity to deliver interventions such as encouraging medication adherence, physical activity, stress reduction, and adequate sleep, or changing their medication to prevent these flares potentially.

The ubiquity of smartphones and the rapid adoption of smartwatches have enabled the collection of sensor data from patients in their daily life. Furthermore, advances in machine learning provide the opportunity to develop novel multimodal machine learning algorithms that leverage smartwatch and smartphone data for disease activity monitoring. Yet, it is currently unclear how these can be used to improve disease management among RA patients. The usage of mobile and wearable technologies in developing digital biomarkers and interventions for RA is a nascent field. 

We are developing a digital biomarker for monitoring and predicting disease activity in rheumatoid arthritis patients in daily life using smartwatches and other ubiquitous devices such as smartphones, smart rings, thermometers, etc. 

Potential impact

A digital biomarker to monitor and predict disease activity in rheumatoid arthritis would enable an unobtrusive way to track disease severity and progression for patients with RA and other inflammatory arthritis. Early detection of flares could inform early interventions such as the change of medication or mobile and wearable-based nudges for behavioral change that would result in improved quality of life for millions of people across the world who live daily with the burden of RA.

This work is motivated by the personal experience of George, the Core Lead who has been living with RA since 2017 as he understands first-hand the physical and mental toll this disease takes. With this collaborative work, he has the unique privilege to leverage his expertise in applied machine learning, wearable computing, and mobile health to improve the experiences of millions like him who have RA.

Relevant Literature

  1. Bergman, M., Zhou, L., Patel, P., Sawant, R., Clewell, J., & Tundia, N. (2021). Healthcare Costs of Not Achieving Remission in Patients with Rheumatoid Arthritis in the United States: A Retrospective Cohort Study. Advances in therapy, 38(5), 2558-2570.
  2. Chehade, L.; Jaafar, Z.A.; El Masri, D.; Zmerly, H.; Kreidieh, D.; Tannir, H.; Itani, L.; El Ghoch, M. Lifestyle  Modification in Rheumatoid Arthritis: Dietary and Physical Activity Recommendations Based on Evidence. Curr. Rheumatol. Rev. 2019, 15, 209–214. 
  3. Creagh, A. P., Hamy, V., Yuan, H., Mertes, G., Tomlinson, R., Chen, W. H., … & Clifton, D. A. (2022). Digital health technologies and machine learning augment patient reported outcomes to remotely characterise rheumatoid arthritis. medRxiv.
  4. Davergne, Thomas, Antsa Rakotozafiarison, Hervé Servy, and Laure Gossec. “Wearable Activity Trackers in the Management of Rheumatic Diseases: Where Are We in 2020?.” Sensors 20, no. 17 (2020): 4797
  5. Hewlett, S.; Sanderson, T.; May, J.; Alten, R.; Bingham, C.O.; Cross, M.; March, L.; Pohl, C.; Woodworth, T.; Bartlett, S.J. “I’m hurting, I want to kill myself”: Rheumatoid arthritis flare is more than a high joint count-an international patient perspective on flare where medical help is sought. Rheumatology (Oxf. Engl.) 2012, 51, 69–76.
  6. Swiss Polyarthritis Association. Living with Rheumatoid Arthritis. 2019.

Share this post

General Information

Develop a digital biomarker for monitoring and predicting disease severity in rheumatoid arthritis patients in daily life using smartwatches and other ubiquitous devices.

Working Group

George Jojo Boateng, PhD


Prof. Dr. med. Caroline Ospelt – Professor of Rheumatology at the University of Zurich (UZH), Dr. med. Raphael Micheroli  – Rheumatologist at the Universitätsspital Zürich (USZ), PD Dr. Przemyslaw Blyszczuk – Researcher at the Universitätsspital Zürich (USZ) 

UniversitätsSpital Zürich

January 2023 – present

George Boateng
George BoatengPostdoctoral researcher