Ambient assistive living environments require sophisticated information fusion and reasoning techniques to accurately identify activities of a person under care. In this paper, we explain, compare and discuss the application of two powerful fusion methods, namely dynamic Bayesian networks (DBN) and Dempster-Shafer theory (DST), for human activity recognition. Both methods are described, the implementation of activity recognition based on these methods is explained, and model acquisition and composition are suggested. We also provide functional comparison of both methods as well as performance comparison based on the publicly available activity dataset. Our findings show that in performance and applicability, both DST and DBN are very similar; however, significant differences exist in the ways the models are obtained. DST being top-down and knowledge-based, differs significantly in qualitative terms, when compared with DBN, which is data-driven. These qualitative differences between DST and DBN should therefore dictate the selection of the appropriate model to use, given a particular activity recognition application.

This research aims to evaluate a mobile phone-based video reminder system (MPVS) for people with dementia, with respect to its design and utility, in addition to its ability to satisfy user needs. Carers for those using the system use a bespoke desktop-based system to record and schedule reminders for delivery through the MPVS system. Nine participants were set eight activities of daily living (ADL) tasks and asked to repeat these tasks over a number of days within an ABAevaluation protocol. In the A phase, ADLs were undertaken using standard reminding techniques; in the B phase, the MPVS system was used; following this, a second A phase was evaluated. ADL completion / compliance was rated and recorded by the carer. Carers and participants were interviewed prior to and following the evaluation to gauge their perceived needs and how these are met, in addition to the potential utility of the technology. The generalizability of the outcome of this evaluation is limited due to the low number of participants; however, the participants reported that the MPVS system assisted them to organize their routine, and the phone used to deliver the video messages was of a good size with adequate screen and audio clarity. The carers saw the potential utility of the technology, and although some had to learn how to use the desktop recording system, the wizard-led interface made it much easier to use for people with minimal computer experience.

In the development of technology for people with mild dementia it is essential to achieve a combination of the features which provide both support and monitoring along with the ability to offer a level of personalization. Reminding support by means of personalized video reminders portraying a relative or friend combined with sensors to assess whether the requested task was performed lends itself as an ideal combination to achieve this aim. This study assesses the potential of using low cost, off the shelf sensors combined with a mobile phone-based video reminding system to assess compliance with task completion. A validation study has been conducted in a lab-based environment with 10 healthy young participants. The work presented discusses the implementation of the approach adopted, data analysis of the results attained along with outlining future developments of this approach.

No abstract

No abstract