I commonly receive requests for information about connectivity and enabling technologies like indoor positioning systems. Here's an example:
...I am currently undertaking research into RFID technologies and WLAN to use within a hospital. In particular I am interested in implementing the use of patient/infant tracking tag, panic tag or status tag, asset tag and also temperature control tags, all of which are working with a WLAN.
Is there any information that you are able to share with me please.
While I don't normally provide services for free (I have bills to pay like everyone else), I have no problem providing some initial information or value to get to a place where an actual project can be considered. So, in that spirit, here's my reply:
I would be glad to share some info with you. From your email it appears you have a number of different indoor positioning applications which you want to undertake. The key to RFID is that there is no one best system or technology for all applications. You have to match the requirements of your positioning applications to the various capabilities of different systems, and many hospitals end up with more than one RFID system as a result.
You also mentioned Wi-Fi in conjunction with RFID. Many people mistakenly think that Wi-Fi is the only common sense solution for RFID. Sadly, this is not true. To get reasonable RFID performance from Wi-Fi you need a lot of access points. If your WLAN is already designed to support wireless VoIP, then you probably have enough APs to get decent RFID performance. If haven't deployed wireless VoIP plan on at least doubling the APs you have installed for typical data applications like computers on wheels. The actual locations of APs also has a big influence on indoor positioning performance, and differs from where they would be sited for optimal data or wireless VoIP performance. As a result, you will likely have to move some of the APs you have. And even with a gob of APs, all in the most optimal positions, you still won't get reliable room level accuracy - which is critical for some RFID applications.
There are two key variables in RFID performance, spacial accuracy and reliability. Spacial accuracy is the resolution or specificity with which an RFID system can place a tag in space, e.g., plus/minus 10 meters, 3 meters or 1 meter. Reliability is the RFID system's ability to consistently indicate the correct location for a tag.
While RFID system performance can be thought of as a continuum, there are two common performance groupings - RFID that can resolve the general location of tags (e.g., the west wing) and those that provide room level accuracy. Most RFID technologies that are good at determining the general location of tags use some sort of RF triangulation between tags and multiple readers. Examples are Wi-Fi based systems from AeroScout and Ekahau and plug in readers from AwarePoint and Radianse. With RF triangulation systems, the greater the density of readers deployed the higher the spacial resolution. Even with a high density of readers, I'm not aware of any RF triangulation system that provides reliable room level accuracy. RFID systems that utilize ultrasound (Sonitor) or infrared (Versus or Centrak) are often the best at room level accuracy. It is also possible to use RF based RFID readers at choke points (in halls or doors) to provide room level accuracy.
Asset management applications for finding and generally tracking equipment like IV pumps and wheelchairs typically provide general locations. Room level accuracy is required for workflow automation such as ensuring medical devices are cleaned prior to being put back in service, or clearing a nurse call when a caregiver enters the patient's room.
Infant abduction systems can be implemented using readers at the locations where people an enter or leave the unit, or systems that track infants from room to room. The latter systems are more expensive than the former due to the greater number of receivers or APs.
You should also be aware that the hidden cost of RFID systems is the cabling and installation costs of readers, The costs of receivers and tags themselves is pretty transparent. Tag replacement and maintenance (battery replacements) costs also tend to be hidden.
Application software - especially for specialized applications like infant abduction - are often tied to a specialized type of RFID system. An alternative is RFID application software that is RFID hardware agnostic; examples include Intelligent InSites and ConnexAll.
Hear hear Mr. Gee!
Any of us in the healthcare technology business would’ve thought that at this point location technologies would’ve become standardized and unbiquitous. I remember going to HIMSS in 2003 or 2004 thinking that next year would be the year.
Sadly in the real time location/Active RFID arena, we are far from standardization of any kind, and customers keep searching for the one-size fits all solution, often focused on infrastructure, rather than performance results and goals. It’s truly sad how murky the market remains.
Given that standardization exists in the passive RFID world, and the benefit of worldwide economies of scale driven by supply chain/logistics types requirements, I have wondered why hospital haven’t more widely adopted more traditional RFID as a means of solving specific and tangible problems.
Tim, I’ve not paid that much attention to passive RFID as of late, what are you seeing in that realm?
Yes, the RFID hockey stick in sales growth has been expected for several years. I think that “market murkiness” is a major factor, along with cost, that is holding back this market.
I see 3 ways for this to resolve: 1) for the market to slowly educate itself sufficiently to make solid cost effective purchase decisions, 2) for a single vendor to achieve dominance and then impose clarity, or 3) for vendors to get together to implement standards and actively clarify the market.
The first option has slowly been happening since day one. The problem is that it’s really slow. The second option I think is unlikely, but if it occurs, it will likely be a hardware agnostic company like Intelligent InSites. The last option - a strategy pursued in a lot of other technology markets - seems extremely unlikely. It’s been a while since I’ve taken a broad number of vendor’s temperatures, but it seems they’re still rooted in the zero-sum game: “if I win, you have to lose.”
Tim
It is nice to see more people writing about RTLS and RFID solutions these days. Agreed that the adoption is slower than anticipated but hospitals are typically conservative about deploying new technologies. I believe hospitals should first determine what are the use case they believe they look to apply technology to such as tracking medical equipment to improve utilization or track patients as they progress through the operating room to post operative care. Once the use cases are picked then the technology both software and hardware can be determine by what level of precision can the solution can achieve (room level or bed level pinpointing of location data) to match the use case requirements. There are many good vendors out there such as the ones mentioned in your blog. I would recommend to your hospital readers, start with the end in mind.
Another early “fantasy” for RTLS was that manufacturers of medical devices would instill tags at the time of purchase so that their utility could be easily realized once deployed in the hospital. This has not occured for many of the same reasons cited above including lack of standardization, lack of customer demand, and possibly a lack of clarity on why we want to do it and whether it is worth the cost. I once asked the question in a somewhat different context of whether our use of technology was getting harder or easiar. I think this applies here as well. (The response was laughter rather than an actual answer.)
Interesting debate, I too have been expected the “explosion” of RTLS/IPS systems in healthcare for years and years. Speaking to vendors it often seems that RTLS is the “be-all-end-all” for any problem you can throw at it, but still the uptake seems to be on the slow side.
While the real answer probably contains many known factors and quite a few confounders, an interesting article was recently published in IJMI, (http://www.ncbi.nlm.nih.gov/pubmed/22857790) investigating the (long-term) success of 30-ish hospital deployments in the US. It paints an interesting, but bleak, picture of RTLS contributions to clinical flow…