Brussels, Mar 2006
"I can't decide this now – I'll sleep on it." This is easy to say but for tens of thousands of us, it is not that easy to do. Sleep and sleep disorders, as a science, are equally elusive because laboratory research conditions are, frankly, not well matched to the calm and cosy environment needed for sleeping. A new EU-funded research project is out to change this with its radical use of information technology and sensitive micro- and nano sleep sensor equipment.
The project, called SENSATION, brings together no less than 46 partners in 20 countries to delve into the sea of sleep research. The aim is to take sleep research to a whole new level by developing a multipurpose sensing platform made up of 17 micro sensors and two nano sensors all connected through a local area network (LAN).
This has a huge advantage over conventional sleep research, which can only manage a certain number of subjects per study and the conditions are not particularly conducive to sleeping. Dr Evangelos Bekiaris, coordinating the project, says the sensors mean subjects can sleep at home. The way it is today, he explains, "you will have to go to a hospital sleep lab for 1-2 nights and have your sleep measured". These tests are costly and often not that reliable.
The sensors will be integrated into a wide range of materials, such as bedding and pillows, wrist straps, seat linings and the frames of glasses. The wireless signals are sent to a computer network which measures brain activity, heart rate, as well as eye and muscle movement during waking and sleeping hours.
The data will be collected in a so-called "body area network" before being wirelessly transferred to a LAN and then beamed to the hospital for analysis. Additional uses are also envisaged for the sensors, such as for monitoring and alerting people while driving or operating equipment, or carrying out critical tasks (i.e. air traffic controllers, nuclear power station supervisors)
The project is now halfway through its funding period and, according to IST Results, is already showing tangible results. The first stage of the project involved data collection, setting up of databases (one with data of normal sleep with 350 participants and another one with sleep data of 400 people monitored while working or driving) and formulating recommendations for the development of sensors.
Sleepy heads at work
"About half of these people were tested in a driving simulator but the other half were tested driving on highways," says Bekiaris. The tests show a significant difference in their reactions. Knowing they were part of a test, was a calming sensation for those in the simulator. But those on the highways really had to fight the desire to sleep, he noted. "It shows that when monitoring people in their daily lives there are significant differences in result."
According to statistics, 25% of traffic accidents in the UK and 40% in the USA are related to driver fatigue. With the project's unique approach to sleep research and its sensor system, serious road accidents could be cut by up to 30% and industrial accidents by over 15%. In fact, there is huge potential for developing sleep alert systems in an array of workplace situations.
"Today, you have techniques that might alert you as you fall asleep, but we want a system that will alert the person already before this happens. In some work environments or while driving, it might already be too late if you have started dozing off," Bekiaris told IST Results. Monitoring for cot death is also an obvious application for the system, as well as in sleep analysis for shift workers to learn more about their natural biorhythm and plan shifts accordingly.
"We currently have nine sensors ready and plan to present the first sensors and prototypes at the international conference on 'Monitoring sleep and sleepiness – from physiology to new sensors'", says Bekiaris. The event will be held in Switzerland, 29-30 May 2006. Real-life tests of the prototypes and the entire sleep platform are planned to start by the beginning of next year, reports IST Results.