Computer vision to detect muscle fasciculations from video in MND

  • Research type

    Research Study

  • Full title

    Computer vision to detect muscle fasciculations from video in motor neurone disease

  • IRAS ID

    362572

  • Contact name

    Amr Tageldin

  • Contact email

    amr.tageldin@nhs.net

  • Sponsor organisation

    Leeds Hospitals Charity

  • Duration of Study in the UK

    3 years, 0 months, 1 days

  • Research summary

    Motor neurone disease (MND) is a progressive condition affecting the nerves that control movement. There is currently no simple, widely available test for MND, and diagnosis can take over a year from the first symptoms. Nor is there an easily accessible and objective way to track deterioration in muscle function over time. One important sign of the disease is “muscle fasciculations” – tiny, involuntary twitches – which can be difficult for clinicians to see, particularly in the tongue.

    In routine care, detection of fasciculations often requires an electromyography (EMG) test, which uses a needle to record electrical activity from muscles. EMG is not suitable for all patients, can be uncomfortable, and is not easily available. EMG is typically done once only, and the specialist equipment and doctor required mean that is expensive and has a waiting list.

    This study will explore whether standard smartphone video, analysed using computer vision (a form of artificial intelligence), can detect fasciculations in the tongue and limb muscles. Participants will include people with suspected or confirmed MND, attending neurology outpatient clinics at Leeds Teaching Hospitals NHS Trust and Mid Yorkshire Teaching Hospitals NHS Trust, and healthy control volunteers.

    Videos will be recorded during routine clinic visits, in some cases alongside EMG testing already being done for clinical reasons. Videos will be anonymised and analysed both by clinicians and by computer algorithms. The study will compare the computer’s accuracy against clinician judgement and EMG results.

    If successful, this approach could provide a non-invasive, accessible tool to aid earlier diagnosis of MND and to monitor disease progression.

  • REC name

    West of Scotland REC 3

  • REC reference

    25/WS/0184

  • Date of REC Opinion

    2 Dec 2025

  • REC opinion

    Further Information Favourable Opinion