Employing Multi-Modal Sensors for Personalised Smart Home Health Monitoring

Applications are sought for a fully funded Research Studentship (PhD) in Artificial Intelligence at Robert Gordon University.

Duration and Funding
The project will be up to 42 months duration, commencing in October 2018. The studentship is funded by Albyn Housing Society (AHS) and The Data Lab, and includes Home/EU tuition fees as well as a tax-free stipend of £14,777 per annum for three and a half years. Non-EU students may also apply but will be required to pay the difference between Home/EU fees (£4,260 per year) and International fees (£15,640 per year).

Proposed Research
This PhD project will develop software solutions that analyse data collected by sensors in smart homes being developed by AHS. The key elements of the solution are to create representations from low-level sensor data that capture the residents’ activities and behaviours of daily living, (e.g. sleeping, dressing, cooking etc.); and then to assemble these activities into personalised profiles on which we can reason about their health.

The project will build on work being done in the FITsense, project developing a solution using in-home sensors to predict increased risk of falls for residents of 16 smart homes being built near Inverness. New solutions are required that consider how additional health conditions and different home layouts can be integrated into a generic transferable solution. This project will develop a framework-based solution that extends work already done by:
• developing solutions for a broader range of conditions e.g. dementia;
• investigating a wider range of sensors to capture richer data; and
• exploring the transfer of solutions to smart homes with different layouts.

The work will advance research in a number of areas. The application and advancement of state-of-the-art approaches in Artificial Intelligence and Machine Learning will support a shifting in focus for health care from a hospital-based model to a real time, data-driven model.

The project is based at RGU, Aberdeen, UK. The initial 6 months will focus on research training as well as learning the key concepts of Human Activity Recognition and health event prediction to create a literature survey that documents state-of-the-art and related research. The following 18 months will focus initially on developing solutions to benchmark problems, and then on building solutions to the real-world datasets available from the smart homes. The final 18 months will work on evaluation of the algorithms, developing refinements that consider personal factors and context, publication of results, and development and submission of the final thesis.

Key Skills
Applicants should have a good Honours degree or a Masters at Distinction level in Computing Science or related discipline. Strong programming skills, in particular in Python, are highly desirable. Some knowledge of AI, in particular Case-Based Reasoning and / or Machine Learning techniques, is also desirable, although not essential. Applicants should have good personal and communication skills, strong professionalism and integrity and be confident working on their own initiative.

Applications
Applications should be emailed to Virginia Dawod at [Email Address Removed] by 12 noon on Monday 27th August 2018. The applications should consist of a covering letter or personal statement of interest, and CV. Further information such as passport details or transcripts may be requested during the short-listing stage. Interviews are expected to take place on week commencing 3rd September 2018.

All enquiries should be addressed to Dr Stewart Massie, [Email Address Removed]