Skip to content

Parkinson's Disease

Contact: Dr Jane Alty

Applications of computer vision to develop biomarkers of Parkinson’s disease

People with Parkinson’s are at increased risk of developing dementia. The worldwide prevalence of Parkinson’s is increasing year on year, primarily due to ageing populations, and an estimated 10 million people are already affected. If we can detect Parkinson’s at an earlier stage, we have the opportunity to implement risk reduction strategies for dementia. There is an urgent need to find objective tests to aid early diagnosis of Parkinson’s, enable accurate clinical monitoring, and to measure research study outcomes. The central movement feature of Parkinson’s is bradykinesia (slowed movements) but this can be difficult to distinguish from changes seen in normal ageing. 70% of people with Parkinson’s have tremor but this can also be difficult to differentiate from other types of tremor. Our inter-disciplinary research involves bringing together neurologists, computer scientists and mathematicians to develop new ‘deep learning’ methods to detect the clinical signs of Parkinson’s more accurately using non-touch computer vision technologies.

Image taken from Williams, Alty et al J Neurol Sci. 2020. Example videos frames taken from smartphone video showing 6 labelled key points that enable non-touch measures of hand movements

Objectives:

  • Current clinical assessments for early Parkinson’s disease have 25% inaccuracy and our research aims to address this gap by developing computer-assisted technologies that accurately measure clinical features

Research Team:

Wicking Dementia Centre, College of Health and Medicine:

University of Leeds, UK

  • Dr Stefan Williams
  • Dr Sam Relton
  • Prof David Hogg

University of Manchester, UK

  • Dr David Wong

University of Loughborough, UK

  • Dr Hui Fang

Outputs:

Accuracy of Smartphone Video for Contactless Measurement of Hand Tremor Frequency.

Williams S, Fag H, Relton S, Wong D, Alam T, Alty J. November 2020; Mov Disord Clin Pract doi.org/10.1002/mdc3.13119

Supervised classification of bradykinesia in Parkinson's disease from smartphone videos.

Williams S, Relton SD, Fang H, Alty J, Qahwaji R, Graham CD, Wong DC. Artif Intell Med. 2020 Nov;110:101966. doi: 10.1016/j.artmed.2020.101966

Seeing the unseen: Could Eulerian video magnification aid clinician detection of subclinical Parkinson's tremor?

Williams S, Fang H, Relton SD, Graham CD, Alty JE. J Clin Neurosci. 2020 Nov;81:101-104. doi: 10.1016/j.jocn.2020.09.046

Time series clustering to examine presence of decrement in Parkinson's finger-tapping bradykinesia.

Zhao Z, Fang H, Williams S, Relton SD, Alty J, Casson AJ, Wong DC. Annu Int Conf IEEE Eng Med Biol Soc. 2020 Jul;2020:780-783. doi: 10.1109/EMBC44109.2020.9175638

The discerning eye of computer vision: Can it measure Parkinson's finger tap bradykinesia?

Williams S, Zhao Z, Hafeez A, Wong DC, Relton SD, Fang H, Alty JE.
J Neurol Sci. 2020 Sep 15;416:117003. doi: 10.1016/j.jns.2020.117003

Computer vision of smartphone video has potential to detect functional tremor.

Williams S, Shepherd S, Fang H, Alty J, O'Gorman P, Graham CD.
J Neurol Sci. 2019 Jun 15;401:27-28. doi: 10.1016/j.jns.2019.04.016

A smartphone camera reveals an 'invisible' Parkinsonian tremor: a potential pre-motor biomarker?

Williams S, Fang H, Alty J, Qahwaji R, Patel P, Graham CD.
J Neurol. 2018 Dec;265(12):3017-3018. doi: 10.1007/s00415-018-9060-z.