Wearable Sensor Based Rehabilitation for Shoulder Pain

  • Research type

    Research Study

  • Full title

    Wearable Sensor Based Rehabilitation for Shoulder Pain

  • IRAS ID

    228905

  • Contact name

    Luckshman Bavan

  • Contact email

    Luckshman.bavan@ndorms.ox.ac.uk

  • Sponsor organisation

    Clinical Trials and Research Governance

  • Duration of Study in the UK

    0 years, 6 months, 1 days

  • Research summary

    We are carrying out a project that aims to develop a motion-sensing platform that can reliably deliver objective information on shoulder function. The sensors we will use are commercially available and safety tested inertial measurement units (IMUs). These devices contain accelerometers and gyroscopes, and are specifically designed for motion analysis. This study forms the preliminary work necessary to validate the output metrics and create activity recognition algorithms. \n\nWe will recruit up to 30 healthy volunteers and 30 patients with subacromial shoulder pain who are prescribed a conventional rehabilitation programme. The patients will be identified and recruited through the shoulder service, musculoskeletal physiotherapy hub or triage clinic at the Nuffield Orthopaedic Hospital. Participants will be requested to wear up to four sensors (upper limb and torso) while performing the routine prescribed exercises. \n\nWe will collect upper limb motion data for sets of exercises performed in a supervised setting and unsupervised in a home environment. Following one episode of routine exercise data collection at home, paticipants will also be asked to provide three hours of free living motion data. Activity logs of exercise routine and activities during motion data collection will be recorded. Patients will be requested to complete a shoulder function and general health questionnaire, and all participants will complete a feedback questionnaire. \n \nRaw data will be processed and analysed to identify features that are unique to rehabilitation tasks and enable development of recognition and function grading algorithms. Machine learning algorithms will be tested using data collected from the same cohort of patients. Once satisfied with the accuracy and reliability of the algorithms we will proceed to the next stage of the project, which will be a clinical trial (a separate application is being submitted for this).

  • REC name

    South West - Central Bristol Research Ethics Committee

  • REC reference

    17/SW/0217

  • Date of REC Opinion

    19 Sep 2017

  • REC opinion

    Further Information Favourable Opinion