Sensing Movement using Action Recognition Technology in Babies
Research type
Research Study
Full title
Pilot study into Human Activity Recognition and Classification Techniques for the Early Detection of Movement Difficulties in Infants
IRAS ID
252317
Contact name
Edmond Ho
Contact email
Sponsor organisation
University of Northumbria at Newcastle
Clinicaltrials.gov Identifier
n/a, n/a
Duration of Study in the UK
2 years, 1 months, 20 days
Research summary
The automated recognition of human activity has wide ranging applications and is widely used in healthcare. This project proposes that this technology could be used in paediatric practice to aid with early diagnosis of movement disorder such as cerebral palsy in infancy.
Making an early diagnosis of cerebral palsy or other movement disorders is important but current methods lack sensitivity and specificity and are time consuming. Applying current tools and assessments is dependent upon the availability of fully trained practitioners - these skills take many years to develop. However, all tools are susceptible to observer fatigue, contain a degree of personal subjectivity and are reliant upon a suitable behavioural state of the infant. If infants are upset or too tired when they attend clinic then the assessment may have to be stopped or repeated.
There is scope to improve the accuracy and accessibility of existing tests through computer-based evaluations; these evaluations could also conceivably provide quantifiable evidence to clinicians. The development of automated systems could also help to significantly reduce the time and subsequent cost associated with these current diagnostic practices. We want to try and develop these techniques that will use non-invasive film/video assessments of normal movement in babies at rest. Our pilot work will use existing film clips that were collected during our routine clinic visits.
REC name
London - Bromley Research Ethics Committee
REC reference
19/LO/0606
Date of REC Opinion
21 May 2019
REC opinion
Further Information Favourable Opinion