Skip to main navigationSkip to main content
The University of Southampton

Research project: COntinuous, Mobile Patient Assessment of Rheumatoid Arthritis (COMPARA)

Currently Active: 

We will make it possible for patients to give 'activity samples' of their daily life, based on data collected using smartphones.

Rheumatoid Arthritis (RA) affects over 400,000 people in the UK. RA is a chronic autoimmune disease causing pain and loss of function. It also causes an acceleration of a number of features associated with ageing such as sarcopenia, reduced mobility and dependence on others. Multiple factors affect a RA sufferer’s perception of their condition: joint inflammation, mobility and pain, levels of physical activity, and mental wellbeing. Monitoring these factors is challenging for both patient and clinician. With RA outpatient appointments often up to six months apart within the NHS, only snapshot measurements inform treatment and patient feedback.

We will make activity sampling throughout daily life possible, based on data from micro-electro-mechanical system (MEMS) sensors, which lie ready to use inside the ubiquitous iPhone, Android and Windows smartphones.

Activities will be classified using methods developed for elite sport
Methods for activity classification

Patient activities will be classified using probabilistic support vector machine (SVM) classifiers, based on features extracted from segmented tri-axis sensor data and GPS, followed by dimension reduction. Such methods have been used to distinguish between moderate levels and sporting levels of activity using accelerometer based data. The touch screen interface of these platforms is also an excellent vehicle through which to combine the more common, and mostly subjective, disease activity scores (e.g. composite disease activity scores such as the 28 joint count disease activity score DAS28 or the patient reported RAPID3) with an objective activity score derived from the MEMS sensors to yield potentially more reliable information for patients and clinicians.

Within this three-year project our objective physical activity scores will be compared in a cross-sectional and longitudinal way (based on a one year follow-up) with disease activity scores. We will also compare the objective physical activity scores with self-reported physical activity measures. This will provide pilot data for our longer-term aim of creating improved composite disease activity scores through continuous mobile assessment with smartphone technology.

Supported by the Southampton Rheumatological Trust and Southampton Musculoskeletal Research Unit.

Related research groups

Computational Engineering and Design
Share this research project Share this on Facebook Share this on Twitter Share this on Weibo
Privacy Settings