VR/AR Upper Arm Rehabilation : Personalised Upper Arm Rehabilitation within Virtual and Augmented Reality
Research suggests that intensive rehabilitation immediately post-stroke (up to 6 hours/day), for people fit enough to participate, could have a significant impact on a person’s function and quality of life. However, as outlined in the recent NICE Stroke guidelines (2013), research proposing this level of intensity is not robust enough yet to support additional funding for the required additional services and a case still needs to be made which shows that the additional short-term cost would provide long-term cost saving. In the absence of funding for intensive one-to-one rehabilitation a strong case can be made for connected health systems which make use of the latest technologies. This project builds on established research into the use of games and virtual reality (VR) to support and motivate physical rehabilitation. This PhD project will extend this work by looking at the connected-health aspects of our adaptive active VR architecture and in particular will focus on formal and informal social factors. It is planned a student would enhance existing algorithms, games and VR systems already developed. One aspect of the connected health system to be explored will involve monitoring user motion both inside and outside the house (using smart devices and other sensors) and enabling exercise in a multitude of locations.  Stroke rehabilitation in adults. NICE guidelines [CG162] Published date: June 2013 https://www.nice.org.uk/guidance/cg162/chapter/2-Research-recommendations
Supervisor – Dr Darryl Charles
Behavlets: Profiling Game Players: Behavlets: Profiling Game Players using a Model that Embodies both Personality/Psychology and Domain Knowledge of Gameplay Patterns
Most model of game players do not adequately account for personality and play preference in real-time decision making in games. The core objective within this PhD project will be to enhance our existing Behavlet model for modelling player behaviour to improve on current models of players in the literature. It is expected that the student will develop a model for adapting games on the basis of real-time observation of player behaviour using the Behavlet approach. Experiments will be design to test the model and the model iteratively enhanced using novel ideas and integrating established state to the art ideas from related research areas. It is expected that at least some of the experimental work will be completed with volunteers either in Ulster, Finland or both.
Supervisor – Dr Darryl Charles
Fusing sensor, video and depth data for analysis of ADLs and multiple occupancy tracking in a smart environment
This project will investigate the use of traditional sensors, video and depth sensors for the purpose of tracking and monitoring (multiple) occupants within a smart environment. This will involve 3D modelling of the environment itself, image analysis of the video data to extract possible occupants, the use of depth data to determine 3D position of occupants and dynamic updating of the environment model due to changes occurring during image capture. In addition, dynamic identification, labelling and tracking of ‘static’ objects will also be required. The overall aim of the project is to ‘fuse’ the various data sources in order to analyse activities of daily living within a smart environment.
Supervisor – Professor Philip Morrow