Pervasive sensors to tackle obesity and problems of ageing

A five-year multi-disciplinary collaboration is to develop continually available, permanently connected pervasive sensors for healthcare in the home.

The collaboration will do this through data-fusion and pattern-recognition from a common platform of non-medical/environmental sensors.

Dubbed SPHERE (Sensor Platform for HEalthcare in a Residential Environment), the project has received £12m in EPSRC funding and includes researchers from the Universities of Bristol, Reading and Southampton, plus partners from Bristol City Council, IBM, Toshiba and Knowle West Media Centre (KWMC).

Project leader Prof Ian Craddock from Bristol University’s Faculty of Engineering told The Engineer that the UK faces a series of health challenges, two of which include obesity and an ageing population.

The UK is Western Europe’s most obese nation and Craddock explained that one factor making an impact on this problem is the combination of sedentary lifestyles and the availability of highly calorific food that can be prepared and eaten very quickly.

The ageing issue also has numerous health implications including managing dementia, Parkinsons disease and stokes.

With the obesity scenario, technology developed through SPHERE would help individuals modify their behaviour via prompts from devices including tablet, mobile ‘phone or flatscreen TV.  In more urgent situations, such as dealing with someone who might have suffered an overnight stroke, SPHERE technology would directly alert healthcare professionals.

According to Bristol University, the collaboration does not intend to develop fundamentally-new sensor technologies for individual health conditions.

Prof Craddock said, ‘The aim of the project is to leverage developments in video game controllers, communications technology, energy harvesting technology and video analytics.

‘The idea isn’t to develop new sensor modalities but [instead] look at …one system in everyone’s home that would make a difference to as many healthcare conditions as possible using a platform of common sensors rather than sensors that would have to be tailored and deployed specifically for individual disease conditions.

‘We’re looking at scaleable solutions because you might be looking at lots of different health issues. For example, in a young person you might be concerned that they were eating too much and that they were at risk from obesity and diabetes. In an elderly person you’d also be interested in eating behaviour but because you might be worried that they weren’t eating and drinking enough.’

He added that the placing of digital healthcare monitoring in the home provides an opportunity to study subtle anomalies that would normally go unnoticed and that stroke provides good examples of this.

‘Many people, especially elderly people living alone, have strokes without realising it,’ he said. ‘If you’d observed their behaviour the day before [a stroke] and the morning afterwards, would you have noticed a change? The answer is that nobody really knows. Probably, but nobody has been instrumented to a level where that sort of data is available.’

Mini strokes can lead to subtle changes in gait and SPHERE will be designed to recognise these and other changes – arising from situations including falls, depression and taking medication – in order to make an assessment.

‘The algorithms would have to triage appropriately,’ said Craddock. ‘The video platform would monitor some things that require a gentle nudge in behaviour or might require no action at all. In some cases there may be a need to alert a healthcare professional or in the case of major changes – for example someone collapsing – ringing for an ambulance.

‘Potentially, you can do all those things with one suite of sensors and hopefully in a completely non-invasive way; this can be part of a person’s everyday life. It won’t require them to perform any action, for example to carry an alarm cord.

Craddock does also envisage, sensors embedded in clothing or jewellery and he said the Southampton team will work on powering the embedded devices with piezoelectric materials.

Once practical, user-friendly technologies have been developed further, they will be piloted in a large number of homes over extended periods of time. 

This research project, plus plastic electronics, micro air vehicles (MAVs) and self-assembly in manufacturing, are discussed here in The Engineer Podcast.