Investigating patient engagement during physiotherapy using EEG

Stroke care is the fourth biggest killer in England and Wales, and the third in Scotland and Northern Ireland. There are over 1.2 million stroke survivors in the UK; 80% suffer from motor impairment, and 74% require long-term support with basic daily activities. Motor function can be improved with physical therapy. Research indicates that in the UK stroke survivors do not receive the recommended amount of therapy. The reasons are complex but include costs and availability of physiotherapists.
Emerging technologies, such as Virtual Reality (VR), offer a potential long-term cost-effective method of providing support to healthcare professionals. One approach is to use Time-of-Flight (ToF) cameras (e.g. Microsoft Kinect) for human pose estimation and motion tracking to enable accurate assessment of patients. These systems offer a way for patients to complete rehabilitation exercises at home. It also gives real-time feedback to encourage patients to stick with their rehabilitation programmes.
Other emerging technologies, such as Virtual Reality (VR), have also been researched as a method of assisting rehabilitation efforts. In a VR environment, the interactions taking place can much more closely represent real life as opposed to traditional computing systems. The idea is that encouraging more lifelike interactions, with exercises mimicking Activities of Daily Living (ADL), can not only provide substantial support for standard rehabilitation regimens but also increase patient engagement.
Nevertheless, assessing the difference made by physiotherapy and other rehabilitation methods remains unreliable. One study looked at the differences between patient self-assessment, physical therapist assessment and posturographic measures when assessing patient performance. Findings indicate that current methods of assessment suffer from a lack of objectivity and are likely affected by biases.
Electroencephalograms (EEG) have been utilised to measure brain activity throughout rehabilitation from brain injury. Studies suggest that it is possible to use EEG biomarkers to predict motor recovery in patients following a stroke. This information might facilitate the tailoring of patient rehabilitation programmes based on the brain activity of patients.
While the existing research into technologically aided rehabilitation has shown promising results, most use one method of assessment/tracking. Recent studies have suggested that a multi-domain approach is beneficial providing objective, comprehensive and personalised feedback on patients undergoing rehabilitation for brain injury. Existing approaches are high cost and the number of sensors attached to patients can lead to discomfort. Many of these drawbacks are surmountable using emerging technologies such as ToF cameras. Pose-estimation/motion tracking, EEG and VR offer a means of providing patients with more engaging exercises. In addition, a greater amount of quantifiable data about the condition of patients will be available to aid clinicians and improve patient compliance.
The overall aim of this research is to analyse patient recovery from stroke, and other brain injuries, to assess their current state and to assist in making recommendations for recovery. The objectives are
1. To bridge the gap between standard neuro-motor rehabilitation methodologies and new technologically aided methodologies.
2. Improve patient engagement and adherence to home rehabilitation exercise programmes.
3. Provide an accurate and quantitative assessment of the patient's neuro-motor capabilities.