Healthcare's hi-tech lifelines
Wearable body monitors are reaching into sports and healthcare as preventative medicine gets active.
Doctors may have to work out less worrying ways to deliver that kind of news over the phone but it is the kind of call that represents the future of healthcare. Instead of phoning for an ambulance after a heart attack takes hold, the crash team drives to you before the natural rhythm of the heart collapses. An ambulance might not even be necessary. The push to put defibrillators into public places will make it easier for health services to deal with a problem without involving any of its own people. Often, the medical equipment will not even be needed.
Emma MacPherson of the Central University of Hong Kong describes what might happen: "If someone gets arrythmia, the emergency services can track them and contact them to talk to them: calm them down. Get them to sit down. That might be enough to prevent a heart attack."
Jörg Habetha, director of medical signal processing at Philips Research, explains why healthcare needs to change from its current practice: "We have the demographic trend of an ageing society. We have to be more efficient in how we deal with chronic diseases. Healthcare will become more cost-aware."
Professor Guang Zhong Yang of Imperial College, London, says: "There will be an increase in age and in chronic, age-related diseases. The problem with many of these diseases is that the early symptoms are hard to capture. Wearable and even implantable sensors may be the answer."
Habetha describes the process that will come as the "consumerisation of healthcare", in which people wear body monitors that update a database somewhere on the Internet with their condition. "Consumers will take more responsibility for maintaining their health," he says.
In many cases that already happens, with people turning to the Web to perform at least a degree of self-diagnosis through sites such as Medhelp. "We want to extend the space of treatment into the home, with rehabilitation-type treatments after hospitalisation or more preventative care. You can have more frequent treatment than is currently the case."
Philips, like others, is concentrating on heart disease. "The reason why we chose cardio-vascular monitoring is that it is the major case of death in most countries around the world. It is responsible for 41 per cent of all deaths in the US. With the ageing of society, we expect that the problem will become aggravated," says Habetha.
The aim is to produce a garment that will record electrocardiogram (ECG) signals. "It will provide direct feedback to the user or forward the information to a service centre," Habetha explains.
Researchers at the Holst Centre in the Netherlands, a branch of the Belgian electronics research institute IMEC, have concentrated on ECG measurements with the smart plaster. They are gradually adding intelligence to the patch, which is worn over the heart, so that it can perform the kind of signal processing that is, today, reserved for hospital-bedside monitors.
Local analysis reduces the amount of data that needs to be transferred by radio to a data recorder, which improves the autonomy of the patch. With local processing, the patch can run for ten days on its internal power source.
The Philips team is trying to develop a vest that carries the ECG and other sensors needed to monitor the heart, in the hope that this will be unobtrusive enough to be worn all day, every day by the user. The advantage of something like a vest is that it is possible to embed piezoelectric sensors to track breathing. Such continuous monitoring may provide better information to the doctor than what can be picked up in a 20-minute consultation.
"If people become stressed when they visit the doctor, you don't get a representative view of their condition," says MacPherson. "Equally, you can't have devices that put pressure on the patient."
Yang agrees: "With wearable devices, you want them to disappear into the fabric of life." The cardiovascular system is not the only target. Philips has worked on a 'smart bed' fitted with sensors that is intended to help with the monitoring of insomnia and other sleep-related problems. The sensors take ECGs and monitor movement and respiration rate so that doctors can make better diagnoses than is possible today.
"People with problems are sent to a sleep centre where they are surrounded by obtrusive equipment. Our goal is to use simpler technology but with very sophisticated processing," said Habetha.
Habetha claims it is possible to detect the different phases of sleep and disturbance using ECG and respiration data. "The ECG from the bed sensors is more noisy than a standard ECG but you can process it to get reliable results," he says.
Professor Paolo Bonato of Harvard Medical School has been working on the rehabilitation of people suffering from debilitating conditions such as chronic obstructive pulmonary disease (COPD). "There are large patient populations and these are long term conditions where clinical management is a factor in success," he says.
Today, to analyse the condition, patients have to go to hospital and go through exercises while wearing reflective markers. "We use camera-based systems. The procedure is a bit time-consuming and it is very difficult to stop people picking up and repositioning the markers," Bonato notes. Perhaps the biggest problem is that the patients have to go in at all. "The results are not necessarily indicative of happens when the patient goes back into the community setting.
"Walking seems to be less predictive than going up and down stairs because the task is less provocative." Again, the aim is to use wearable sensors to detect the motion over longer periods of time, with one option being the use of a electrotextile, where the sensor is effectively part of the fabric itself. "But the fabric is not 100 per cent comfortable because it is not entirely flexible," says Bonato.
Some of the systems are now going into trials. A 12-month clinical evaluation of the effectiveness of cardiovascular monitoring under the European MyHeart got underway last November at six hospitals in Germany and Spain.
"We are just observing. We will correlate our data with the events and see if we could identify trends in the data before the events actually occurred," says Habetha.
It may be possible to take treatment further and use the data from wearable sensors to adjust medication. Conditions such as Parkinson's disease and bradykinesia can be alleviated with drugs but there are problems. "The therapies are effective but most patients develop motor complications," explains Bonato. "The severity goes up and down over time."
Altering the dosage level can bring the unwanted motions back under control after a few hours. The trick is being able to detect a rise in actions such as dyskinesia.
Blood pressure challenges
Bonato reckons it is probably not feasible in the short term to have wearable sensors tell a patient to adjust medication on a daily basis. But long-term measurement will help the clinician make better decisions about each prescription than what is possible in a 20-minute consultation.
One of the most difficult problems facing technologists developing body monitoring devices is how to measure blood pressure unobtrusively. Jens Mühlsteff of Philips Research at Aachen, says: "It is becoming more important to track blood pressure changes during the day. Physicians would like to see how your body reacts under stress so they can improve blood pressure management."
Professor YT Zhang of Central University of Hong Kong agrees: "Blood pressure variability is a very good indicator of mortality." Unfortunately, traditional measurement techniques involve putting external pressure on a limb, which is a long way from being unobtrusive. One option is to look at how pulses propagate through the arterial tree inside the body - after each beat of the heart, a pulse wave travels through the body ultimately dissipating in its countless capillaries. This measurement can work because the speed of the pulse wave increases as the blood vessels stiffen, which you would expect to happen under high internal pressure.
So far, so good. But there is a catch. In fact, there are several. "It is an indirect method. So you need a calibration step," says Mühlsteff. "The pulse-wave velocity increases with age, the arteries have quite a complicated structure and the readings are influenced by stress and medication.
"Typically, you have to place two sensors above an artery. It is very difficult to place the sensors accurately." One option is to use the time it takes for the pulse to reach a sensor away from the heart and rely on an ECG measurement to provide the baseline time at which the pulse got underway. The sensor might take the form of a ring or of a device worn behind the ear.
The question is how often you need to calibrate the sensor. Calibration would be needed every day. "The arterial system changes day by day," Zhang notes.
Push the button
Doing that with a cuff-based blood pressure monitor is probably not going to work. However, it might be possible to use a pressure sensitive button to perform the calibration. The idea would be that, at the start of the day, the user would press the button, release it and then repeat the cycle. The sensor would then be set up for the day. It implies that users will actually remember to do that. But the framework of healthcare may change so much that consumers will just find themselves having to adjust. "Personal healthcare will happen: there is no alternative," claims Habetha.