A framework to model human behaviour to detect and predict cognitive decline within home environment

Assisted living and diagnosis of early dementia

We have developed a novel cost effective system that uses non-intrusive technology to monitor the normal daily behaviour of elderly people to establish normal behavioural pattern. This allows the detection for any sudden change in behaviour that might be a symptom of dementia or other aging related diseases.

Key achievements
  • Research on sensor applications and NFC led to the development of MOCAD (Mobile Carer Assistant System) which in turn led to the formation of a new business.
  • In the same context, the team developed an innovative methodology and a tool (MOCAD) for monitoring carers’ attendance including scheduling, recording visits and providing assistance to carers to provide more efficient service to patients using the combination of GPS (Global Positioning System) and NFC (Near Field Communications) technologies in an innovative way.
  • Care System Integration Ltd is using these services in systems for councils in England and Wales with a pilot of 45 units in the Rhonda Cynon Taff in Wales to be followed another with seven units in Torfaen, also in Wales.
  • Larger scale deployment is planned in Liverpool as part of Caring Cloud, a multi-million pound TSB-funded project that includes also the mobile operator O2 and others.


Background and technology

The aim of the project is to design and implement a framework that can be used to detect and predict the onset of dementia in home settings for elderly people who are living alone, although similar but more complex sensing technology that includes individual identification can be used in multi-inhabited elderly homes as well.

The framework will rely on the minimal raw context data in order to infer a behavioural model. This model will be used later on in order or detect and predict the onset of dementia.

The framework will be able to alert a third party to instigate actions. The framework will totally rely on the detection of the behavioural change as an indicator for the onset of dementia. In the implementation, wireless sensor networks and utilities’ meters will be used to collect raw context from the home environment.


The Framework’s Hardware Implementation

The Ambient Intelligence Lab

The Sensors