Using Movement Intention EEG in Rehabilitation

• Main Author: Rashid, Usman (Author)
• Other Authors: Taylor, Denise (Contributor), Signal, Nada (Contributor), Niazi, Imran Khan (Contributor)
• Format: Others
• Published: Auckland University of Technology, 2019-11-12T01:44:45Z.
• Subjects: Stroke; Rehabilitation; Brain Computer Interface; EEG; Movement-related Cortical Potentials; Thesis
• Online Access: Get fulltext

 Stroke causes brain injury and is a major cause of long term disability in New Zealand and across the world. Scientific evidence shows that our nervous system is plastic and adapts itself to this injury in order to recover from disability. One promising research area which aims to enhance brain's plasticity for faster and greater recovery follow- ing stroke is an intervention which involves temporally associating repeated electrical stimulation to affected limb to coincide with the intention to move the limb. This intervention achieves temporal association by using a Brain Computer Interface (BCI) to interpret Movement-related Cortical Potentials (MRCPs) recorded with Electroen- cephalography (EEG) to identify the intention to move. There is an increasing body of laboratory based research evidence which supports this rehabilitation intervention. However for successful translation of this intervention into clinical practice, research to support the development of a mobile and usable medical device is required. In this thesis I have comprehensively examined the gaps in the scientific knowledge which hinder development of this intervention to support its uptake in clinical practice. To address the key gaps I have proposed a blueprint of a BCI which has potential to be developed into an affordable, mobile and usable rehabilitation medical device. I have proposed and evaluated the different components required for this BCI within three experimental studies. In the first study I have evaluated the performance of a low power small footprint cost effective analogue to digital converter in recording MRCPs against a laboratory based gold standard system. In the second study I have proposed a method for automated labelling of MRCPs. In the third study I have proposed an optimisation method along with extending an existing algorithm for enhancing the detection of movement onset and offset from surface electromyography. Moreover, the conceptual blueprint and the evidence generated by this thesis contributed to the development of a prototype rehabilitation medical device at a national technology challenge. Future research should focus on how such a device can be embedded into clinical practice.