Reductions in fertility combined with increases in life expectancy are resulting in a globally ageing population. The result is an increasing strain on healthcare resources as members of the older population are more prone to suffer from chronic illness hence requiring long-term care and management. One current focus of research is to investigate methods of alleviating the increased strain on healthcare resources through the use of intelligent environments (IEs). IEs incorporate the use of sensor technology to facilitate long-term, home-based assessment of health condition with the goal of minimising the amount of direct clinical supervision required whilst maximising the frequency and objectivity of patient data collection. This thesis presents the design, development, testing and evaluation of novel methods for the assessment, visualisation and simulation of sensor data generated within IEs. Details of two novel methods for the objective assessment of the severity of the motor symptoms associated with Parkinson's disease are presented. The first method utilises the Nintendo Wii Remote for interaction with motor tasks and the second method uses a computer vision-based approach to monitor activity of daily living performance. Both methods were capable of quantifying the presence of tremor during activity performance. A novel method for IE data visualisation is also presented in the thesis. This method was capable of visualising spatiotemporal data trends using a novel density ring format within 2-dimensional (2D) virtual environments (VEs). Testing on data collected from an active smart lab illustrated the ability of the approach to highlight typical and atypical activity trends. Additionally, a novel method for the simulation of IE data is presented. This method was capable of facilitating the generation of simulated IE datasets through the navigation of an avatar through user-created 2D VEs, facilitating rapid prototyping without access to a physical IE implementation.
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:601204 |
| Date | January 2013 |
| Creators | Synnott, Jonathan |
| Publisher | Ulster University |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
Page generated in 0.0067 seconds