Awareness is growing about the challenges of developing and maintaining safe and effective wireless medical devices. What with IEC80001 moving forward (due to be finalized next year) and the recent series of wireless medical device workshops, people in hospitals and among vendors are asking more of the hard questions about wireless. Amongst the turmoil, participants are jostling for position. This post looks at common problems with Wi-Fi, a report from U.K. alliance ERBI, and some alternatives to Wi-Fi.

Problems with Wireless

Those of us who are old enough, think back to the golden age of wireless medical devices -- channelized analog telemetry. These systems were so basic and limited in scope (a couple dozen transmitters typically covering just a single 30 bed unit) that they had few problems and required little maintenance.  Today, larger hospitals are pushing the envelope with a few hundred patient monitors and a thousand or more wireless infusion pumps. These wireless devices are using sophisticated client radio/access point (AP) communications protocols to maximize capacity, whether using Wi-Fi or WMTS. We've since left the golden age far in the past.

Radio frequency (RF) spectrum is a shared resource. There's no getting around that fact, even with "dedicated" spectrum. The ether in which wireless signals move is like gases in the atmosphere or chemicals in water. There are no ways to practically segregate RF signals to specific areas, except for a Faraday cage. In a health care facility, some shielded rooms in Radiology qualify as Faraday cages, but little else. Much of the rest of a health care facility consists of objects and structures that seem to perversely confound and obstruct RF communications in  ways like partially blocking and attenuating signals, creating multipath interference, and radiating both intentional and unintentional interference. Intentional interference is where two or more users of a portion of wireless spectrum get in each others way, disrupting or degrading the communications of one or both parties. When there are problems with two or more wireless devices using the same spectrum, this is intentional interference, often referred to as coexistence problems. Unintentional interference comes from electromechanical devices that accidentally spew RF signals as a consequence of some degradation or failure. Common sources of unintentional interference are florescent light balasts, blow dryers, paper shredders, elevator motors, or faulty microwaves. You can see a bunch of examples of RF interference on a spectrum analyzer (which everyone doing wireless medical devices should have, and know how to use) here.

Another common variable to wireless communications is capacity. Wireless capacity is measured by the number of transmitters and receivers that can use a given amount of bandwidth or spectrum, and the amount of information -- be it data in bits, or modulated analog information -- that can be reliably moved through an allocated frequency range. Capacity is determined by the amount of bandwidth (more is generally better) and technical tricks used to increase the efficiency of moving information. Channelized analog is the least efficient means to move information wirelessly. More efficient techniques entail digitizing the information, moving it more quickly (higher data rates), and using schemes to share bandwidth among users more efficiently (things like FHSS, ODFM and MIMO). Another important part is the definition and adoption of a common scheme that ensures coexistance between the various users.

Reliable wireless communications has three basic requirements: 1) the wireless application must detail the proper requirements and technical specifications necessary for proper operation, 2) the wireless system must be properly designed and tested to verify the resulting wireless infrastructure's performance, and then 3) the wireless system must be managed and maintained to ensure ongoing operation within specifications. This is a lot of work. And if you want to use wireless stuff, there's really no way to avoid this work -- although some may suggest otherwise.

Blog Mobihealthnews picked up a press release from ERBI, an industry alliance of biotech firms in the U.K., that suggests that, "A dedicated frequency band for medical devices would boost confidence and stimulate uptake of wireless technology within a healthcare environment." You can download your own copy of the report here (pdf).

The "study" is not really a study but a record of, "a facilitated workshop, which brought together experts from the fields of medical device development, wireless technology and healthcare" (i.e., members of ERBI). The usual wireless challenges were trotted out, like congested frequencies and problems of interoperability and coexistence. No data is presented to quantify or characterize these issues, and one is left to assume that there are no solutions -- besides dedicated spectrum -- to any problems that do exist.

Dedicated Medical Spectrum: Benefits and Costs

Yes, the list of devices using the 2.4GHz portion of the ISM band (where Wi-Fi is located) seems endless. But because applications of the ISM band are relatively short range, it is only the specific devices using 2.4GHz within a location that have to share the spectrum. In a hospital you might have computers on wheels (COWs), wireless VoIP handsets, smart pumps, patient monitors, maybe PDAs. Each of these devices associates with just one out of hundreds of APs in an enterprise. Is this such an overwhelming amount of devices?

As we saw above, there are inherent risks in the use of any wireless technology. If wireless medical devices received their own dedicated portion of the radio frequency (RF) spectrum it might make wireless communications better in some ways.

With dedicated spectrum it is possible that there would be less interference and fewer coexistence problems. But there is more to this than must having dedicated spectrum. Certainly there will be fewer devices trying to use the same spectrum. However, there still must be sufficient bandwidth available to medical devices so they all have sufficient "elbow room". Current WMTS is a case in point. Users of WMTS get just 13MHz of bandwidth on 3 separate frequency bands (608-614 MHz, 1,395-1,400 MHz, and 1,429-1,432 MHz). Vendors only support two of the three bands, the 608 and 1,400 bands.

The creation of a dedicated frequency band would certainly increase the cost of wireless medical devices -- which is no way to "stimulate [the] uptake of wireless technology" in health care. Most current devices leverage tremendous economies of scale from commercial and consumer technologies like Bluetooth and WiFi. This would be lost with a specialized band where units sold will be measured in tens of thousands, rather than hundreds of millions.

Some proponents of dedicated spectrum suggest using spectrum that's close to the ISM band. This would potentially allow for the re purposing of existing ISM technologies to the new dedicated medical spectrum. This approach has some validity, if adjacent spectrum is available and existing technology can be re purposed for a reasonable cost. An example of this is Philips' wireless patient monitors that use DECT as the underlying technology for their WMTS infrastructure.

The communications capabilities on a new dedicated frequency would be feature poor. Right now, health care leverages lots of important advanced features from horizontal market wireless technologies. Want to keep your advanced authentication and encryption? Out the window! Want quality of service or fast roaming with that? Forget it! How about special power save features for longer battery life? Sorry! These are features supported by the 802.11 standards and increasingly built in to today's wireless medical devices. Again, some of these features may be available if re purposed Wi-Fi or other technology can be used.

The devil's always in the details, and in any re purposed wireless technology, the question is how many capabilities can be simply ported over and how many must be re engineered in part or in whole.

What Dedicated Spectrum Won't Fix

Dedicated spectrum for medical devices would not greatly reduce interference. Most interference comes from unintentional sources like faulty florescent light ballasts, blow dryers, paper shredders, microwaves and elevator motors. Only a small portion of interference comes from other intentional users in the same band. Yes, coexistence problems among devices in the ISM band do exist, but it is manageable. Oh, and because it's a big market, we have lots of fancy tools built into the infrastructure for monitoring and fixing interference and coexistance -- more features we'd likely loose with dedicated spectrum.

Getting wireless regulatory bodies and governments worldwide to select a frequency band for medical devices would take forever -- if it is even possible. Note that portions of ISM used by WiFi aren't even consistent world wide. Precious RF spectrum goes to national security (public safety, military) and serves major engines of the economy (broadcast, commercial data, etc.) While we can all agree that we spend a lot of money on health care, wireless spectrum devoted to health care won't generate much economic activity or lower health care costs. It seems to this observer that it is unrealistic to think governments around the world will cough up spectrum for something that's already working in established frequency allocations.

Health care will loose bandwidth with a new dedicated frequency allocation. Current bandwidth for 2.4GHz and 5GHz 802.11a/b/g is 383MHz (not including the 255MHz recently added to the 5GHz band). This compares to WMTS, which gets a measly 13MHz -- barely enough for a few hundred wireless patient monitors in the same hospital. Current 802.11 bandwidth is 30 to 400 times bigger than WMTS. Another dedicated band is MICS, the Medical Implant Communications Service, that get's just 3Mz of bandwidth (more on this below). Bandwidth is directly related to capacity; the more bandwidth, the greater the number of users that can be supported. Bandwidth also provides the elbow room to manage coexistence problems. To be safe and effective, wireless medical devices need bandwidth.

Once a new frequency allocation is dedicated to health care it will be time to develop standards so solutions from different vendors can coexist with the same spectrum. According to Wikipedia, the first 802.11 standard was released in 1997 after several years of work. What followed is over 10 years of further enhancements and refinements.  The alternative to time consuming standards development is a hodge-podge of proprietary products that don't coexist, or do so poorly.

Remember the early days of WMTS? With the advent of digital television, medical device vendors went to the FCC and said they had to have RF spectrum to replace that lost to digital TV (for some reason, they didn't want to use the ISM band). The FCC came up with 3 non-contiguous slices of spectrum and named medical as secondary users. Vendors informally agreed to 1) implement support for all 3 slices of spectrum and 2) work together to develop interoperable coexistance across all vendors using the frequency. Sadly, neither happened. Existing vendors just recrystaled their existing telemetry systems for 6o4MHz (they added 1.4GHz support later). No effort was made to develop any standard to ensure coexistance. In fact, for the first few years, WMTS based systems from two different vendors couldn't be installed in the same hospital. Over time, GE and Philips worked out reasonable WMTS coexistence. Besides Philips and GE, only Spacelabs and Datascope released WMTS based systems. All other medical device vendors have looked to 802.11 for their wireless devices. Which brings us to the final issue with dedicated spectrum, currently installed products.

There is a large base of installed wireless medical devices. Most of these devices use Wi-Fi, and some WMTS. The typical life cycle for a medical device is 7 to 12+ years. Providers and their vendors will be managing these existing wireless devices for many years to come. Products based on any new dedicated spectrum won't hit the market for at least 2 or 3 years, leaving the hard work of managing today's wireless technology a requirement for a decade or more. That's not to say new wireless technologies won't be developed and come to market, they will. And when they provide tangible benefits, they should be adopted.

Alternatives to Wi-Fi

Given the increasingly well known issues with Wi-Fi networks, I frequently get questions about alternatives. In fact, there are alternatives, and for some applications, attractive alternatives. Wireless applications in health care are divided into short range cable replacement and  enterprise network "last 100 feet" cable replacement. Choices for connecting to the enterprise network are limited, and include WMTS and the ISM band. Short range cable replacement applies to wireless sensors in body area networks (BANs) and are limited to ISM and (due to a recent FCC announcement) MICS -- or the Medical Implant Communications Service.

For connecting to the enterprise network using the ISM band your options are 802.15.4/ZigBee, the high powered Class I version of Bluetooth,  and components based on the now defunct standard HomeRF. Both ZigBee and Bluetooth have specific strengths and weaknesses that must be carefully matched to effective requirements. (Hint: eliciting good wireless requirements for those just getting into wireles is difficult.) While commercial products based on the HomeRF standard haven't been manufactured for years, it is likely possible to license the technology or  even components, for the right price. Finally, you can roll your own custom or semi-custom protocol (a perennial favorite among some engineers).

Regardless of what wireless technology you choose, you must have good requirements, and do a good job matching technology with requirements to pick the best tech. Most wireless medical devices are deployed enterprise wide, so your Wi-Fi alternative must be cost competitive with the added burden of a duplicate enterprise-wide wireless network.

Presently, every dollar invested in improving the performance of the enterprise Wi-Fi network (whether driven by wireless medical device adoption, wireless VoIP phones, or COWs) is justified across multiple uses. A dedicated wireless medical device infrastructure -- that's not dedicated to all of a hospital's medical devices, just one device type from one vendor -- has to be cost justified based on that one medical device application.

Oh, and you still have to meet the three basic requirements for reliable wireless communications (there's no getting out of that). This means having the human resources to provide the service and support necessary to develop, install and support wireless devices. These costs may be less when deploying a dedicated wireless infrastructure, but probably not a lot less.

Wireless sensors are really a short range cable replacement application. Wi-Fi is not an option here. Many of the issues discussed above are applicable to wireless sensors, and there are also many specialized issues and requirements. Unlike wireless enterprise networking applications, wireless sensor implementations vary considerably. Options range from commercially available proprietary radios like the ANT, to "science project" technology like NASA's Radio Frequency Health Node wireless sensor system, and academic projects like Matt Welsh's CodeBlue.

Can We Fix Wireless in Health Care?

Hmm, I don't think it's broken. Do we know everything? No. Are we figuring some things out as we go? Sure, especially as technology evolves. Is there anything that indicates that existing technology is incapabile of supporting wireless medical devices, or is inherently unsafe? No. Are there some who have this mostly or completely figured out? Yes, on both the vendor and provider sides.

This is just like designing a medical device. There is no "one way" or even a "best way" to implement wireless. Like other areas of connectivity, the phrase, "you don't know what you don't know," applies to wireless -- requirements, enablement, deployment, management and support.

If you're looking for something to chew on, be sure to check out the FDA's Draft Guidance for Industry and FDA Staff - Radio-Frequency Wireless Technology in Medical Devices. If you're a hospital biomed or IT network person, you should read this document too.

If you need help, call a connectologist.