I'll be giving an overview of the project, as well as my role within it; it won't be a tech-heavy talk. The talk is part of a larger session to the SE group in Infolab, hence the short timeframe, but no reason why anyone can't tag along for just one talk. Personally, I think this is a great example of multi-disciplinary cross-institution research.
For those that don't know, SAMS is attempting to provide novel techniques to passively detect early signs of dementia (particularly Alzheimer's disease). I have been developing the monitoring software that is deployed on participants computers, it runs with no-user interaction, and is deployed for about nine months. The software collects lots and various data/text from participant's everyday activities. This data is then mined and analysed, looking for patterns/metrics/heuristics changing over time. It is a £1m project across three universities, with myself and Ann Gledson (Mancs Uni) being the only two technical RAs. I have been working on this project for two years now, and we're entering the final 6 months (the analysis phase). The project shows promise (of course I'd say that), but it's had its fair share of significant ethical and technical challenges; which makes the research all the more interesting.