Engineers are developing electronic and analysis technologies designed to help keep chronically ill people out of hospital. Stuart Nathan reports.
More than 17 million people in the UK have some kind of chronic illness, such as diabetes, asthma, high blood pressure, and angina. These patients take up most of general practitioners’ time and account for many hospital admissions when their conditions become life-threatening.
Keeping these people out of hospital as much as possible is one of the main targets of the National Health Service.
Why is this so difficult? Part of the reason is that monitoring someone with a chronic disease is difficult. Take Type 2 diabetes, for example. A few blood tests a year and a twice-yearly check-up provide snapshots of glucose levels, blood pressure and so on, but what goes on between those visits is a matter of informed guesswork, both for the individual and the doctor.
It is between doctor’s appointments that diabetics tend to go out of control, and often end up hospitalised with hyper (or hypo) glycaemia. Cumulatively, these loss-of-control episodes lead to the severe effects of diabetes — loss of sight, damage to limbs and cardiovascular problems.
Care of chronic illnesses is an important focus for the fast-growing area of digital healthcare, which uses electronic monitoring and analysis technologies, generally taken from other engineering disciplines, to provide a constant picture of an individual’s health. This can alert them, and their doctor, when their condition nears a danger zone.
Frequently, and often to their surprise, UK engineers and technologists are working with medics and finding their expertise can have unexpected benefits. The UK leads the world in this area and with technologies now beginning to come to market, the sector is set for growth even in difficult economic times.
One such technologist is Lionel Tarassenko, a professor of electrical engineering at Oxford University, who is now the director of the university’s Institute of Biomedical Sciences. He has a long-standing interest in the use of electronics in medicine — he believes he was the first electrical engineer in the UK to devote his PhD to paediatrics, some 25 years ago — and is now applying his expertise in industrial data-gathering and analysis to the medical field.
The aim of his work is to help avoid the need for intensive hospital treatment. This saves money and nursing resources but, more importantly, it keeps people healthier. He is also concerned about stopping previously healthy or stable people from sliding into crisis. Good monitoring, said Tarassenko, is key to understanding a patient’s condition, but even in intensive care units, it is more difficult than it seems.
He knows a great deal about condition monitoring. ‘In the 1990s, my lab group designed the engineering health monitoring system for the Rolls-Royce Trent 900 engine,’ he said. ‘In the last five years, we’ve moved that system from the aero engine to the hospital patient.’
The system, called Visensia, is designed to help avoid what Tarassenko calls ‘a law of unintended consequences’.
Patients deemed at risk of deteriorating, and those on organ support, in post-operative care and coming out of intensive care, are monitored electronically all the time. They are festooned with sensors for their heart rate, respiration, blood oxygenation, temperature, blood pressure but, often, their condition deteriorates.
‘And we miss it happening,’ said Tarassenko. ‘Why? It’s because alarms go off all the time. There are literally constant alarms, and the nurses ignore them. And that’s because 86 of every 100 alarms is false. A probe falls off, or the patient moves awkwardly, and the alarm goes. The nurses have to ignore these false alerts — they only trust the information they’re getting when they can see both the patient and the monitors, when they make their rounds every four to six hours.’
This means they will often miss the 14 per cent of alarms that are real. The problem could be even more acute with new hospitals where, to minimise the risk of infections and improve patients’ privacy, more private rooms are being built. The consequence is that nurses will not be able to see their patients from their station, said Tarassenko.
The Visensia system borrows from industrial condition monitoring to provide a more detailed picture from the sensor readings. Using a technique called data fusion, it decides whether an abnormal signal from a sensor is a false alarm, or a true indication that the patient’s condition is deteriorating.
‘Let’s say an ECG sensor pad falls off — normally an alarm would sound. But our system thinks: the heart rate reading is zero, so what’s the blood oxygen doing, and what about the respiration? If they are both fine and haven’t changed, then the patient obviously doesn’t have a heart rate of zero. So it won’t sound an alarm, and the nurses can replace the sensor next time they make their rounds.’
Visensia has been on trial in the US for the past two years and has been used on about 1,000 patients, said Tarassenko. ‘In one high-dependency unit, they used to have two or three cardiac arrests every month; they’d had 50 in the 18 months before we went there. In the 18 months we were there, they had zero cardiac arrests, with exactly the same nursing staff and procedures.’
It will be in use at the Radcliffe Hospital, Oxford and Guys and St Thomas’s, London this year, he added.
The problem of assessing patients’ conditions between GP visits remains pressing, so Tarassenko has devised an application for a familiar piece of technology, the mobile phone.
‘People have looked at using the internet for monitoring but there isn’t universal access, especially for poorer people, who are more likely to suffer from chronic conditions. But more than 90 per cent of people in the UK have a mobile phone and we’re working with Vodafone to supply a monitoring system that will work even on simple phones.’
The system consists of software on the phone that allows patients to keep a regular diary of their condition by following a series of prompts.
For an insulin-dependent diabetic, for example, the software will ask when they last ate, then for a blood sugar level from a pinprick test. The patient can use a Bluetooth-equipped monitor or enter the figure manually. It may then ask for a blood-pressure measurement, and enquire whether the patient is going to do less exercise than normal, the usual amount, or more than usual, over the course of the day.
The system then shows the patient a chart of where their blood sugar levels are and where they have been over the past day or so, marked with green (healthy), yellow (not so good) and red (bad) areas.
The patient then has to enter the number of units of insulin he or she intends to inject to stay within the green zone. ‘We could have put an insulin calculator on the system, but British doctors prefer the patients to make the decision themselves, rather than relying on a machine,’ said Tarassenko.
The patient’s phone transmits the information to a secure web page accessed by a specialist nurse at the local primary care trust (PCT), which is responsible for organising and paying for GP and specialist services in the region. The patient will receive a call from the nurse at least every six weeks but if they are losing control of their condition and their health is becoming a concern, the nurse will call the patient immediately.
The system has been used in 12 PCTs around the UK for a variety of conditions, including diabetes, asthma, chronic obstructive pulmonary disease (COPD) and hypertension. The PCTs pay a fee per patient for the service — about £200 a year, although it varies depending on the number of patients receiving the service. However, the average hospital visit for a patient with COPD costs in excess of £2,000, so the net savings can be huge, said Tarassenko. ‘When we tested this with the Bristol PCT, we halved the admission rate.’
‘I feel like a bit of an imposter, having spent most of my career developing mobile phones,’ he said. ‘But now we’re using that technology not as a fashion statement, but for the good of mankind. If you apply just a bit of this technology to healthcare, you can make major innovations.’
Toumazou’s major technological innovation was to develop ultra-low power electronics for mobile phones, which consume only nanowatts of power. He is now developing this technology into ‘smart plasters’, disposable sensors that can monitor vital signs and transmit the readings to specialists.
‘The digital plaster contains a chip that monitors the full ECG; the latest version will monitor your blood pressure and temperature as well,’ he said.
The chip, which is being marketed by Toumazou’s spin-out company, Toumaz Technology, under the name Sensium, sits beneath the gauze layer of the plaster, is powered by a printable zinc-air battery that uses oxygen from the air in its cathode, and contains an antenna that transmits data to a PDA.
‘The system extracts the data, does some intelligent work with this data — for example, it could detect cardiac arrhythmia — and if there’s an anomaly, it will transmit a signal and alert a GP or medic,’ said Toumazou. ‘You could imagine a situation where the heart surgeon Magdi Yacoub, who we’ve been working with, could be in Egypt and logging on to monitor a patient’s blood pressure in the UK. This is technology we understand — it’s cheap and cost-effective.’
He added: ‘In the next few years, you might be able to go to a chemist and buy a pack of digital plasters for the equivalent of $10 (£6.70) each.’
Top: modelling the pancreas on a chip. Below left: the Sensium chip. Below right: Bluetooth peak-flow meters for asthmatics and blood glucose sensors for diabetic working with mobile phones
The next development for the Sensium technology is combining diagnostics with treatment. ‘The idea is to make a chip a few microns wide that sit on a disposable insulin pump,’ said Toumazou. ‘This chip will model the physiology of the pancreas and provide intelligence for the pump.’
The chip, which incorporates a glucose sensor, would be implanted into the body, and transmit information wirelessly to the pump. ‘In the same way that the pancreas monitors blood sugar and decides how much insulin your system needs, the intelligence on the pump will decide how much insulin you need and provide 24-hour coverage.
‘If you change your diet or do some exercise, it will detect the change in glucose level and secrete the insulin you need to stabilise it. That would remove the need to make spot measurements of glucose and inject insulin, and it would help avoid hypo- and hyperglycaemia.’
Communication technologies are not the only ones being pressed into play. The techniques of microfluidics — using chips etched with nanoscale channels that propel and shepherd liquid samples into minute electronic analysis devices — are also coming into play, as Jon Cooper, head of bioengineering at Glasgow University, explained.
Cooper is particularly concerned with cancers and other diseases of the gastrointestinal tract. These are hard to diagnose and are major killers — one person in the UK dies of bowel cancer every three minutes, said Cooper. It is a bigger killer than breast cancer, because of the extensive screening programmes for that condition.
Bowel cancer, however, is much more difficult to screen. First, the procedure makes inhibited Brits uncomfortable — it involves smearing a stool sample onto a card then posting that off to a central lab for analysis. Second, the analysis involves finding blood in the sample, which is difficult to analyse because it is opaque. The normal tricks, which involve seeing a colour change or shining a light through a sample to detect the presence of coloured compounds, are no use.
Cooper’s team has developed a microfluidic device that uses the electrical properties of blood to detect it in any sample, no matter how opaque and turbid. ‘This has clear advantages in terms of speed and reliability,’ he said.
Taking this technology further, Cooper has now designed a ‘lab-on-a-pill’ to detect bowel cancer inside the body. ‘There have been a number of companies that have launched video cameras on pills, which you swallow and provide a view of your GI tract,’ he said.
‘But there’s a problem when you get the lower bowel, because of its many folds and twists. It’s difficult to get pictures. So we’ve taken the same blood sensor technology and integrated that into video pill technologies.’
The device, like Tarassenko’s and Toumazou’s innovations, is being developed commercially, in this case by a Glasgow University spin-out called Wireless bioDevices.
Known as SensaPill, it has so far attracted more than £500,000 in funding; the company is seeking more funds to take it though clinical trials.
The UK is becoming a leading centre for bioengineering research. ‘We’re ahead of the rest of Europe and well ahead of the US,’ claimed Toumazou.
But this has not always been the case. Tarassenko recalls that when he was doing his PhD, his official designation on the research team, as the sole engineer among medical specialists, was ‘button pusher.’
‘The medics viewed engineers as glorified technicians 25 years ago, but this has changed completely in one generation,’ he said.
‘And the main reason for that is imaging. CT scanning was invented in the UK and so was MRI. These have become such powerful tools for medics that their whole view of engineers has changed; they now see us as equal partners.’
Toumazou sees the establishment of state-of-the-art bioengineering institutes, such as the ones at Imperial, Oxford and Glasgow, as the fruit of this collaboration. ‘They break the silos of traditional university research — the idea that here’s mechanical engineering, here’s electrical, here’s physics and these are all separate. We’ve seen that it’s putting these things together that gets results.
‘Working with medics in this shared environment: that’s where this stuff gels.’