Doctor of Philosophy
School of Electrical, Computer and Telecommunications Engineering
Post-stroke rehabilitation is essential for stroke survivors to help them regain independence and to improve their quality of life. Among various rehabilitation strategies, robot-assisted rehabilitation is an efficient method that is utilized more and more in clinical practice for motor recovery of post-stroke patients. However, excessive assistance from robotic devices during rehabilitation sessions can make patients perform motor training passively with minimal outcome. Towards the development of an efficient rehabilitation strategy, it is necessary to ensure the active participation of subjects during training sessions. This thesis uses the Electroencephalography (EEG) signal to extract the Movement-Related Cortical Potential (MRCP) pattern to be used as an indicator of the active engagement of stroke patients during rehabilitation training sessions. The MRCP pattern is also utilized in designing an adaptive rehabilitation training strategy that maximizes patients’ engagement.
This project focuses on the hand motor recovery of post-stroke patients using the AMADEO rehabilitation device (Tyromotion GmbH, Austria). AMADEO is specifically developed for patients with fingers and hand motor deficits.
The variations in brain activity are analyzed by extracting the MRCP pattern from the acquired EEG data during training sessions. Whereas, physical improvement in hand motor abilities is determined by two methods. One is clinical tests namely Fugl-Meyer Assessment (FMA) and Motor Assessment Scale (MAS) which include FMA-wrist, FMA-hand, MAS-hand movements, and MAS-advanced hand movements’ tests. The other method is the measurement of hand-kinematic parameters using the AMADEO assessment tool which contains hand strength measurements during flexion (force-flexion), and extension (force-extension), and Hand Range of Movement (HROM).
Butt, Maryam, Enhancement of Robot-Assisted Rehabilitation Outcomes of Post-Stroke Patients Using Movement-Related Cortical Potential, Doctor of Philosophy thesis, School of Electrical, Computer and Telecommunications Engineering, University of Wollongong, 2020. https://ro.uow.edu.au/theses1/997
Unless otherwise indicated, the views expressed in this thesis are those of the author and do not necessarily represent the views of the University of Wollongong.