Wibree In Medical Links

Wireless Connectivity – the key to enabling Personal Medical Technology Wireless Connectivity out of the box eHealth, Telecare and Assisted Living are becoming key phrases in any discussion of the evolution of national healthcare policies. They are seen as fundamental to moving responsibility for healthcare back to the patient – a move dictated by changing demographics and the growing incidence of long term chronic diseases. To enable this, the availability of mass market wireless technology that supports these devices is necessary. That technology is maturing in the form of Bluetooth, Wi-Fi, and Wibree (Ultra Low Power Bluetooth). As well as improving the usability of existing medical devices, wireless connectivity is set to spawn a massive explosion of devices aimed directly at the consumer. This White Paper looks at the standards serving this growth and discusses the issues that the industry and healthcare providers need to address to deliver new services based on the availability of “Always ON” patient data.

Nick Hunn CTO - EZURiO Ltd

EZURiO Ltd Saturn House Mercury Park Wycombe Lane Wooburn Green Bucks HP10 0HH United Kingdom Tel: +44 1628 858 940 Email : [email protected] Web: www.ezurio.com

An

©2007 Copyright EZURiO Ltd All rights reserved BLUETOOTH is a trademark owned by Bluetooth SIG, Inc., U.S.A. and licensed to EZURiO

white paper - explaining wireless

Page 1 WHP-07008 1v1 Wireless in Medical technology.doc

Wireless in medical technology Wireless is a hot topic in medical device circles. It has always exerted a fascination for the medical community through its ability to remove the cables which hinder the mobility of patient monitoring equipment. The more recent, and greater interest is in its ability to enable a new generation of mobile, personal healthcare devices. There is a growing acceptance that we need to increase the degree of personal health monitoring to cope with the pressures imposed by changing demographics and the increasing incidence of long term chronic conditions. Wireless is seen as a key tool in moving that belief from concept to deployment. With such a requirement apparent, wireless groups are busy trying to convince medical device manufacturers that their wireless standard is best. In an ideal world, there would be one wireless panacea. In our real world, physics intervenes and dictates that the conflicting requirements of different medical devices will need different solutions. That doesn’t mean that there is a not a need for standards. Interoperability between devices will be critical if they are going to be deployed in millions, and eventually in billions. As a result the wireless debate is coalescing on a limited number of wireless specifications, with work on interoperability being conducted by standards groups in conjunction with medical device manufacturers. Key movers in this area are the Continua Health Alliance, the Bluetooth Medical Devices Working Group which covers Bluetooth and Wibree (also known as Ultra Low Power Bluetooth) and the IEEE 11073 Personal Health Devices Working Group.

Choosing the Appropriate Standard Three wireless standards are likely to dominate the majority of medical applications – Wibree, Bluetooth and Wi-Fi. Many more exist, but the cold facts are that if medical devices are to become endemic, they will need to follow the lead of the wireless standards that are embedded within the devices that they connect to, which are typically PCs and mobile phones. Today Bluetooth and Wi-Fi dominate – Bluetooth for connections between portable devices and Wi-Fi for connections to fixed access points and infrastructure. Many medical researchers are currently working with ZigBee, or the 802.15.4 radio standard that underlies it. Unfortunately this pair of standards has failed to gain traction in mainstream consumer devise. Continuing to work with them presents the risk for researchers that they will divorce themselves from the commercial market. Instead the mobile phone industry is embracing Wibree, which is not part of the Bluetooth family of standards. Wibree will serve Ultra Low Power peripherals. Between these three – Wi-Fi, Bluetooth and Wibree, a set of short range wireless standards with critical market mass is in place. The exception to this triumvirate of standards will be the field of implantable devices. Here a wireless link needs to be maintained between an implanted sensor and an external transceiver that the patient wears. This is the one area of medical wireless where standards are largely inappropriate. Wireless links in these devices are highly proprietary with extreme requirements in terms of battery life and reliability. Specialist companies work closely with medical device designers to implement micro-miniature wireless chips at specific frequencies reserved for these applications, which are better suited than the 2.4GHz band for close proximity sensors on the human body. Wireless standards and chips from these applications are unlikely ever to enter the mainstream of medical wireless. Leaving implantable devices aside, from clinical equipment through personal wellness and healthcare products, there is a diverse range of applications. Despite this diversity, four primary parameters dictate the requirements of short range wireless technology for the vast majority of medical use cases:    

the the the the

range over which the device needs to operate, amount of data that needs to be transferred frequency of these transfers – how often data needs to be sent, and power available – typically whether it is battery or mains powered.

Despite common preconceptions to the contrary, across the three wireless technologies, the range for a given output power and data rate is largely identical. At very high data rates the range will decrease. Obviously if a smaller range is required, such as body or personal devices, the output power can be reduced, which further saves battery life. Range Body < 1metre Personal 1-5 metres Room 5-10 metres House 10-100 metres

Wibree YES YES YES YES

Bluetooth YES YES YES YES

Wi-Fi YES YES YES YES

So although the decision on output power may vary with application, each of these technologies can be tailored to meet the range required. Ultra low power is effectively synonymous with very low amounts of data transferred – that’s because physics demands a certain amount of power in sending data over the air regardless of the standard. Hence a device that streams large quantities of data can never be low power. However, many devices in the healthcare arena send data occasionally, so the overall power consumption can be low. Examples are weighing scales, “lifestyle” devices such as pedometers, “fitness” products, and the range of sensors required for assisted living and dementia care, which are typically devices sending simple state information, such as door position, temperature or bed occupancy. Daily Connection Connections per day Type of data

Wibree ~ 50 Event (ON/OFF)

Additional Features

Bluetooth ~5 File Transfer Voice

Wi-Fi Always ON File Transfer Streaming VoIP

Conversely, if the major requirement is to transfer large amounts of data, the choice is likely to gravitate towards Wi-Fi or Bluetooth. Data Rates Typical Application Rate

Wibree < 50kbps

Bluetooth < 500kbps

Wi-Fi < 10Mbps

A defining reason for the choice of wireless can also be the topology of the connection. If a device is static and connects to the internet, Wi-Fi is a sensible choice. If it’s a mobile device that is more likely to connect to a mobile phone, then Bluetooth and Wibree become obvious contenders. In all cases, it is straightforward to relay any data received via short range wireless onwards to the Internet. Connection Topology (AP = Access Point)

Wibree to Phone to PC to AP

Bluetooth to Phone to PC to AP

Wi-Fi to AP

Mesh topologies remain an area of constant interest, yet for consumer devices they are likely to be of little more than academic interest. They are problematic to configure and do not lend themselves to user installed products. In many cases, mesh is promoted for its ability to provide a robust wireless link. Advanced wireless standards such as Bluetooth and Wibree largely remove this requirement as they incorporate adaptive frequency hopping regimes that provide excellent robustness. The form factor of the device can be equally constraining: how big it is and how it’s used. In an increasing number of consumer oriented devices this may be the primary design requirement that dictates the choice of wireless standard. For example, in a running shoe it will never be acceptable to embed anything bigger or heavier than a small button cell. For the battery to have acceptable life, the only choice of wireless is Wibree. Conversely,

in the clinic, where mains power is available, either for direct power, or to charge a battery, the choice of wireless technology becomes accordingly broader. Power Source best matched to the Wireless Technology Power Source Wibree Bluetooth Wi-Fi Button Cell YES AA Batteries YES YES Rechargeable Li-Ion YES YES Mains YES

Finally, for consumer devices, cost becomes an important part of the equation. The more complex the radio, the more it costs. Cost always depends on the specific implementation, but as a general rule, Wibree enables very low cost devices, Bluetooth medium cost, and Wi-Fi predominantly high-end or professional equipment. Implementation Cost Including associated components and stacks (at 100k volumes)

Wibree $3

Bluetooth $10

Wi-Fi $20

Accessing the data Before looking at the different application areas, it’s important to understand what happens to the data when it is transferred over a wireless link. A cable is a pretty dumb component – limited to transferring data between two fixed points. Wireless access typically extends that transfer to a database at a remote location. As such it allows data to be aggregated from multiple devices, whereas most cables only display data on a nearby monitoring device. To date many wireless medical products have been designed with a wireless capability that does little more than communicate information to a nearby PC. Such a solution is not going to change the way we use healthcare devices, nor the paradigms of using the resulting medical data. Widespread deployment of wirelessly enabled medical devices requires a consideration of the impact of the remote databases, typically via the Internet, where the patient’s data can be analysed and acted upon. It is the development of these back-end systems that will transform the way that we can deliver personal healthcare. At present the lack of such back-end services is a major obstacle to the greater use of wireless devices. There is a “Catch 22” within this. Without a reasonably sized and representative set of data it’s difficult to develop accurate expert rules that allow medical staff or users to make the data monitoring statistically significant and useful. However, until the analysis associated with it yields a high percentage of positive feedback and alerts, the medical staff loses faith in the system and stops adding new patients to the trials. So the database remains too small to produce statistically significant and useful results leading to too many false positives… It’s unclear what will break this impasse and move from what is essentially a trial environment into mass deployment. Too many trials have been performed whose only apparent deliverable appears to be the funding application for a further trial. The NHS in the UK has just approved some significant trials that acknowledge the problem and will attempt to start collecting a suitable mass of data, but even these only involve around 2,500 patients each. The UK’s recently announced Assisted Living Innovation Platform goes further, envisaging a large scale deployment trial of 10,000 units in the next few years. The importance of scaling up trials cannot be over-emphasised. The solutions that the world needs cover hundreds of millions and ultimately billions of patients and devices. Today the number of remote healthcare devices deployed remains miniscule. For the

simplest device – the fall alarm, global deployment is only a few million. For anything more complex, deployment numbers rarely exceed the low hundreds. Over the coming years, if the services that we envisage are going to exist and become endemic, then we need to extend this knowledge base. Outside of the UK, it is unclear who will take the lead in capturing this quantity of user data and developing the analysis and subsequent services behind it. Institutional inertia means it may not be our National Healthcare Services that create the breakthrough; instead it may be driven by others with more focussed interests, such as Private Medical Companies, Medical Equipment Providers, Health Insurers, the Pharmaceutical Industry and even Expert Community Groups on the web. This last group could well be the most dynamic source of innovation in telecare. An unexpected addition to this list could the mobile phone network operators. They see a real benefit in offering additional services to their customers, not least to increase customer loyalty. Their 3G networks provide the wide area connections to retrieve patient data from phones and they have already spawned an innovative content management and delivery industry around their services. Alongside them we may also see offerings from other companies that depend on our loyalty. We could even end up trusting our health to our supermarket – it’s a sobering thought that they probably know more about our long term diet than does any medical professional we deal with. One good piece of news is that the industry is developing standards for the way in which data from devices is structured, under the auspices of the IEEE 11073 Personal Health Devices Working Group. This should ensure that similar devices from different manufacturers will be able to send medical data in the same format. It will allow a wider range of medical devices to contribute to patient databases and provide the opportunity for innovation from a diverse range of application developers. As an aside, the value of these databases should not be dismissed. Today most medical knowledge is based upon an understanding of human symptoms and reaction to medication only at the time that the patient is unwell. The ability to extend this with massively improved granularity to cover everyday life will generate levels of information that will allow governments, healthcare systems, insurers and drug manufacturers to gain a far better understanding of the value of treatments. This in turn should allow medication regimes to be better tuned to the need of individual patients. The value of this data and improved drug usage may well be used as a business case to help fund some of the deployments.

Device Sectors – traditional healthcare As we saw above, there is a fairly well defined set of criteria that will determine where different wireless standards are deployed.

The Clinical Environment is where the most complex equipment will be found. Typically this will be used to retrieve significant quantities of data that needs to be transferred to a central server, where it will be analysed and appended to the Patient Record within that institution. Such equipment is likely to be mains powered, not least because of the complexity of its monitoring task. This is the simplest task for wireless: it is the base functionality of removing the physical nuisance of cables. It is probably the area where wireless will have least impact on the design or usage of the medical device that employs it. The obvious choice for such applications is 802.11, also known as Wi-Fi or Wireless LAN. Many hospitals and clinics already contain the infrastructure for wireless LAN, in the form of Wi-Fi access points. Some of these products will extend to high-end consumer medical devices, where users want to keep ahead of their neighbours. However, most Wi-Fi enabled products are likely to be seen within the clinical environment. In many ways, this defines where medical wireless is today. Where wireless standards really come into their own is in enabling the next generation of personal medical devices.

The growth of personal healthcare devices Today most medical devices are purchased by health professionals. Over the next few years that will change as lower prices and simpler operation mean that they are increasingly offered as consumer products. Initially these applications are likely to be centred around wellness, fitness, sports and lifestyle devices. However, the experience gleaned from these will see the sector expand to everyday medical monitoring, with the resulting devices having more in common with consumer electronics goods, manufactured by everyday brands. Within the next decade expect to see the market progressing as far as a designer branded glucose meter. For these personal devices, two linked standards will dominate almost all of the short range wireless connections and the way that they are used – Bluetooth and Wibree. Bluetooth is the most successful of any short range wireless standard. At the start of 2007 it reached a level of 2 million silicon chip shipments every day of the year and continues to grow. That success has led to a high level of development effort, resulting in a wireless connection that is low cost, interoperable and highly robust. It has also attracted an active working group that is cooperating with medical device manufacturers to define a Bluetooth Medical Device Profile that enables products from different manufacturers to interoperate. That level of interoperability covers connecting to a monitoring device that can either display information, record the patient data, or act as a gateway to forward it to a remote database or an Internet destination. It is anticipated that the work on the Bluetooth Medical Device Profile will be completed early in 2008. This will coincide with the first standards from the IEEE 11073 Personal Health Devices Working Group to provide a common data exchange protocol and definition of device data formats. Together they will enable the first generation of low cost, interoperable medical devices. It’s likely that the first devices we see appear will be blood pressure meters, glucose meters and weighing scales. Where Bluetooth has struggled to compete is in the realm of very low power products, where there may be a requirement for the medical device to operate for many months on a button cell battery. This is poised to be an area with explosive growth, covering everything from active clothing (which isn’t just science fiction, but covers many down to earth applications like pedometers in trainers and incontinence sensors) to assisted living devices such as fall sensors and bed monitors. Here Wibree is set to transform the market. The principle reason for this is that Wibree has a unique design feature in that it can share the key components of a Bluetooth radio chip. What this means is that every mobile phone that includes Bluetooth, and most do, will soon also be able to communicate with a low power Wibree device, at no extra cost for the mobile phone. These phones will then be able to act as gateways to transfer the data from a plethora of Wibree enabled sensors to a remote monitoring service via the mobile network.

Wibree is optimised for sending limited amount of data, as this ensures a long battery life. It has a range of several hundred metres that will allow it to be used for sensors around a typical house, making it equally applicable for personal devices and fixed, assisted living sensors. The first chips are likely to become available in the second half of 2008, when they will be available at low cost to incorporate into a wide range of consumer medical, healthcare and fitness devices. The Bluetooth Medical Devices Working Group is already incorporating Wibree into its work program to ensure a structured approach to the standards. There will be applications where more than one wireless standard is combined. Smart clothing with multiple sensors may well use Wibree links to a central data-logger that’s worn somewhere else, and which itself uses a higher bandwidth Bluetooth link to upload the data. There are plenty of other wireless standards, but the market for medical devices demands interoperability. 802.11 provides that for infrastructure links of complex products within the clinical environment that need to be network enabled. The combination of Bluetooth and Wibree, with the support of the medical device industry provides a complete set of wireless capabilities to serve the growing personal healthcare market. The building blocks for wirelessly enabling medical devices are coming into place. Designers of medical devices and the services behind them should be seriously considering these options to provide wireless connectivity for their next range of products. If they don’t, they may find they lose their market to the consumer electronics industry.

Nick Hunn CTO – EZURiO November 2007 http://www.ezurio.com Further reading: The promise of eHealth – an analysis of barriers to implementing telecare – www.ezurio.com/wp