Many disorders and injuries result in abnormal movement that impairs function and affects quality of life. A method that would allow clinicians to easily quantify a patient’s movement pattern could be an invaluable diagnostic tool, as movement patterns could be compared with databases of patterns from a variety of disorders. It could also help monitor disease or rehabilitation progression by comparing movement patterns with earlier measurements from the same patient.
Currently - in most cases - clinicians use observational scales to assess the level of impairment and monitor progression of disease. For example, the Action Research Arm Test (ARAT) is used to assess arm performance in individuals with stroke, brain injury and multiple sclerosis; the Scale for the Assessment and Rating of Ataxia (SARA) includes a number of tests involving hand movements, walking, and standing; and the Unified Parkinson Disease Rating Scale is a rating tool that evaluates typical Parkinson’s symptoms including tremor and rigidity.
Objective measurement of movement is currently only done in specialised labs. They use markers attached to the legs and pelvis of a patient to track their movement pattern while they walk along a lab walkway lined with infrared cameras. The process is time consuming, expensive, and not suitable for all patients. Moreover, this analysis is limited to walking patterns, as it is not clinically feasible to measure arm movements with marker-based systems.
Capturing any human motion without specialist equipment would enable its wider use in the clinic and impact how movement disorders are diagnosed and treated. The aim of this PhD is to investigate how novel technologies such as machine learning and biomechanical modelling can be used to quantify human movement based on video recorded with regular RGB cameras. The methodology will potentially involve semantic segmentation and other machine and deep learning techniques to identify and separate the person from the background and further segment them into relevant body parts, and to then perform skeleton fitting using detailed multi-body biomechanical models.
The student will work in the Aberdeen Centre for Health Data Science (ACHDS). We are a dynamic and multi-disciplinary group of engineers, scientists and clinicians that aim to improve health and care with data. The supervisory team includes Dr Dimitra Blana, lecturer in ACHDS with extensive experience in movement analysis and biomechanical modelling, and Professor Nir Oren, Head of Computing Science and Associate Director of ACHDS. We have access to video data for pathologies such as cerebral palsy and Parkinson’s, that can be used to develop and test the markerless motion capture methodology. We collaborate with clinicians in NHS Grampian and clinical movement analysis laboratories across the UK, which provides a direct route to clinical implementation.
This project is advertised in relation to the research areas of APPLIED HEALTH SCIENCE. Formal applications can be completed online: https://www.abdn.ac.uk/pgap/login.php
. You should apply for Degree of Doctor of Philosophy in Applied Health Science, to ensure that your application is passed to the correct person for processing.
NOTE CLEARLY THE NAME OF THE SUPERVISOR AND EXACT PROJECT TITLE ON THE APPLICATION FORM. Applicants are limited to applying for a maximum of 3 applications for funded projects. Any further applications received will be automatically withdrawn.
This project is funded by a University of Aberdeen Elphinstone Scholarship. An Elphinstone Scholarship covers the cost of tuition fees only, whether home, EU or overseas.
For details of fees: View Website
Candidates should have (or expect to achieve) a minimum of a First Class Honours degree in a relevant subject. Applicants with a minimum of a 2:1 Honours degree may be considered provided they have a Distinction at Masters level.
Yozbatiran N et al. A standardized approach to performing the action research arm test. Neurorehabilitation and Neural Repair. 2008; 22:78-90
Weyer A et al. Reliability And validity of the scale for the assessment and rating of ataxia: a study in 64 Ataxia patients. Movement Disorders 2007; 22:1633–7
Fahn S et al. Recent Developments in Parkinson’s Disease, Vol 2. Florham Park, NJ. Macmillan Health Care Information 1987, pp 15 3-163, 293-304
Baker R, Measuring Walking: A Handbook of Clinical Gait Analysis, Mac Keith Press 2013
Blana, D et al. Real-time simulation of hand motion for prosthesis control, Computer Methods in Biomechanics and Biomedical Engineering, 2017; 20,5: 540-549.
Blana, D, et al. Feasibility of using combined EMG and kinematic signals for prosthesis control: A simulation study using a virtual reality environment, Journal of electromyography and kinesiology. 2016; 29:21-27.
Zhao, Bo, et al. A survey on deep learning-based fine-grained object classification and semantic segmentation. International Journal of Automation and Computing. 2017: 14.2: 119-135.