Chip technology enables us to improve existing measurement and diagnostic methods for conditions such as cardiac and neuro disorders. It makes the equipment more compact, more economical and more comfortable for the patient, too. In 2015, we carried out a number of projects in this area. Of course, you need to have the right expertise, but in itself, this is not the greatest challenge in terms of medical sensor systems.
So, what is the greatest challenge? Developing new methods, that’s what! For example, our research group is looking at how sensor systems can make a contribution in the diagnosis and/or monitoring of heart failure, stress, sleep apnea and head trauma. Working with medical specialists, we’re examining which parameters are relevant and how we can measure them accurately. The difficult thing in all this is that the method has to be demonstrated and approved in trials with a sufficient number of patients. Which of course means that you need a robust and mature demonstrator – and that is by no means straightforward in the research phase. But it’s not impossible: this year we succeeded in setting up trials for heart failure (30 patients) and stress detection (1,500 people).
In the area of sensor systems for lifestyle applications, there are all sorts of other challenges. These include genuine ease of use, the personalization of algorithms and the creation of convincing applications that help persuade us to change our behavior. Most of the gadgets you find on the market today tend to be disappointing when it comes to accuracy. They are very good for checking whether fit people manage to do their 10,000 steps or cycle enough kilometers, but they are of no use at all for the other 90% of the population. They are simply not accurate enough for measuring whether your elderly aunt is getting up and moving about the house enough, or whether your overweight uncle is increasing his level of fitness by doing the extra exercises given to him by the doctor. Overall, current devices are not at all inspiring and don’t actually anticipate your individual needs and habits.
How can you use sensors to encourage older and obese people to exercise more? How can you get someone to stop smoking? How can you help a person to keep their stress levels under control? We at imec and Holst Centre are confident that sensors can help to recognize habits and make adjustments to behavior. But it is certainly no easy task: not technologically, but also not because psychologists and behavioral scientists tend not to be very familiar with modern technology. As a result, there is still some skepticism about whether or not sensors are of any value in changing people’s patterns of behavior. We are currently working with some enthusiastic behaviorists from UZLeuven and KULeuven to investigate the usefulness of sensors for stress management.
One of the main problems with using sensors to change behavior is the personalization required for the sensors themselves. Take stress, for example, which expresses itself differently in each individual. One person may start sweating, while other gets heart palpitations – and so on. This is in stark contrast with heart rhythm measurements, for instance, where all of the signals are more or less the same. They are also well known and any discrepancies are clearly identifiable. Personalized sensors and algorithms are needed to identify behavior correctly with any accuracy and then make adjustments. In practical terms, imec and Holst Centre took the first steps in 2015 to validate the measuring technique used for stress and to recognize people’s habits and trigger moments using sensors and artificial intelligence technologies. In 2016, the emphasis will be on providing feedback, for example to reduce stress. A project will also be started to help smokers to quit their bad habit with a ‘virtual coach’, as we also call our sensor approach. Because one thing is certain: if we were all to have a personal coach who kept an eye on us 24/7, we wouldn’t have to make a list of New Year’s resolutions any more. Or maybe we would – even if it was simply to pass them on to our virtual coach.
