A CMIO’s Take on 5G
Editor’s note: John S. Lee, MD, CMIO at Edward-Elmhurst Healthcare, participated in the first-ever CHIME Innovation 5G Executive Forum in December. Co-Developed by CHIME Innovation, AT&T Business and the HCI Group, a Tech Mahindra Company, the small cohort-based forum included 5G healthcare use cases, an immersive 5G experience, a potential 5G roadmap and more. He shares what he learned and the implications of 5G in this article.
The educational 5G series will continue in 2020 with a similar CHIME Innovation 5G Executive Forum on March 26-27 in Plano, Texas, and a new CHIME Innovation event, Impact of 5G in Health and Care, on May 5-6 in Atlanta.
John S. Lee, MD, CMIO at Edward-Elmhurst Healthcare
I recently had the pleasure of attending a conference on 5G wireless technology co-developed by CHIME Innovation, AT&T Business and The HCI Group, a Tech Mahindra Company. The purpose of the gathering was to educate and facilitate innovation. Clearly, this CMIO needed some education. I admit that I really didn’t know what exactly is 5G.
I found out that 5G is not a single technology but rather a generational constellation of technologies. For instance, 5GE (E stands for evolution) sits on top of the current LTE platform and is, in fact, souped up 4G. It is faster than 4G but not really a generational leap in throughput. At the other extreme, millimeter wave signals can deliver gigabits per second data throughput. However, these short wavelengths prevent it from penetrating even minimal physical objects. In the middle is Sub-6, which refers to frequencies under 6GHz. These wavelengths are longer than millimeter waves, which allows the signal to penetrate solid objects and carry longer distances. The signal throughput does not compare to gigabit millimeter wave speeds but still is far faster than 4G. Broadly, in addition to speed orders of magnitude higher than 4G, 5G will reduce latency and improve signal reliability and security.
The question was posed why would we not use WiFi and, in particular, anticipate wide adoption of the new 802.11x WiFi 6 standard? The speeds that 802.11x will provide will be on par with 5G. However, WiFi has inherent instability and latency. The signal spikes that occur because of traffic that moves on and off the network create performance uncertainty.
On the other hand, 5G creates specific lanes upon which the signal is carried. Administrators can then dictate which lanes move faster or are prioritized. For instance, one could force the network to reserve 10 times the capacity for a robotic surgery application versus a channel that would be carrying ventilator data. In a sense, WiFi is like a 12-lane superhighway with varying types of traffic. 5G is a dedicated hyperloop carrying only a single type of vehicle.
Therefore, in many ways, I think 5G will not succeed 4G cellular, at least not in the way we currently conceptualize ubiquitous use of cellular phones. Instead, 5G’s high throughput, high security and high reliability can be considered more of a successor to wired ethernet with the added mobility and flexibility of a wireless infrastructure. As a CMIO, this helps me understand where I may be able to deploy WiFi 6 and 5G within my system. Both will be extremely useful for care delivery and infrastructure purposes.
Of course, outside the hospital, there will still be a demand for classic wide area cellular coverage with next generation speeds. Thus, we will probably need two strategies. The first will be on premise solution where high speed, volume and reliability are needed as a replacement for WiFi and wired connections. The second will be a wide area/off premise strategy where penetration and distance are as important as speed. 5G definitely has a place in this two-strategy approach.
The new wireless technology is impressive, but the ultimate goal is to get the signal on to fiber as soon as possible. If we keep this in mind and think about the characteristics and limitations of the various forms of 5G, we can start considering how we can use the technology. We can create wide areas of mobile coverage, but the highest speed millimeter wave signals need to be deployed strategically.
The gigabit speeds will allow for intense computing on edge. The combination of the cloud with edge computing then could disintermediate and further blur the lines between local computation and cloud-based computation. This concept then opens the possibility for high bandwidth, high demand computing without requiring providers on the clinical front lines to tether themselves to physical hardware.
For instance, one of the demonstrated applications was delivering care with augmented reality Magic Leap retinal displays. As advanced and light as these devices were, I still felt that they were clunky and not usable for long periods of time. However, it is possible that the computational burden and physical bulk could be off loaded to an edge architecture that could be connected to a lighter physical device. This edge device then could be connected to the larger network for less time-sensitive work.
Other potential use cases for 5G were explored:
- We will be able to download massive imaging files wirelessly to various care settings, no longer tethered by wired network cables.
- Training and simulations can be facilitated by augmented reality enhanced by massive multimedia applications.
- Remote robotic surgery could potentially be performed from across the continent or even across the world. (However, I personally feel that there will still be a need for a local human surgeon since we need to deal with the limitation of the speed of light. As anyone who has been on a WebEx knows, there is often a bit of a lag. As most surgeons know, having lag of milliseconds is unacceptable if a vessel is accidentally cut or the surgeon realizes that a strand of connective tissue that he or she was going to dissect was actually a nerve.)
These networks will facilitate placement of all sorts of sensors throughout the medical continuum to constantly feed a data beast and create insights to make medical care safer and more efficient, much as manufacturing and aerospace has used similar sensors to improve their efficiency and safety. The “sensor” that was most intriguing to me was the video as a sensor. As a physician and CMIO, I know that one of the biggest hurdles to improving our care systems is that so much of our processes require human abstraction and reporting. Cameras and microphones could detect hand washing compliance, interpersonal interactions, fall risk or a patient in pain, using AI algorithms. Yes, there is a creep factor, but perhaps we can take a cue from airport backscatter security devices that now only show blocky silhouettes to direct manual searches. 5G can be part of what makes this feasible.
Pamela Arora from Children’s Health in Dallas commented that it is important not only to have the right tool, right person and right time but also to deliver it to the right place. This is especially an issue for specialty pediatric care where there are very few specialists who cover an enormous geographic area. 5G technology creates the potential for distributing care efficiently across a large physical area. I recently read a piece where a small rural hospital closed to initial protestations of the community, but realized that they didn’t really need a hospital. Perhaps this wireless technology can help distribute care that would otherwise not be available and small communities won’t need to support the enormous fixed costs of a free-standing hospital. Could this model be the Netflix to the 20th Century Blockbuster hospital?
Ultimately, I don’t think we can predict the full impact of these new technologies. As an analogy, who could predict that, when smartphone technology appeared in 2007, it would result in Uber, mobile check-in for airlines and ubiquitous GPS with presentation of minute-to-minute graphical traffic? Then again, remember that Apple tried to introduce a similar concept in the Newton over 10 years before with miserable results. Even if 5G starts out as a Newton, I have no doubt that it will eventually completely revolutionize how we deliver medical care. I encourage you to prepare your organizations for this revolution!