Here’s our first guest post, from D-dad Lane Desborough. When Lane’s son Hayden was diagnosed with type 1 diabetes four years ago, Lane decided to devote his life to reducing the burden of diabetes. Leveraging his engineering experience in industrial automation, he’s spent the past four years on closed loop insulin delivery, remote monitoring, and tools to support the adjustment of insulin pump settings. — Howard
The ability and motivation to improve the lives of people living with type 1 diabetes has never been greater. Let’s tackle the “ability” first. Through four vignettes, I’ll demonstrate some exciting trends and the questions they raise.
1. Software is eating the world
Software is not bound by the constraints of the physical world. It can do things that are impossible in hardware alone. Software can be anywhere, instantly, thanks to the Internet. Software is a harmless, weightless mental abstraction until it is instantiated in the physical world. It does neither harm nor good until it is compiled, interpreted, or otherwise caused to operate on physical objects: computers, smartphones, firmware in devices, etc.
One example is the LAME MP3 encoder. LAME stands for LAME Ain't an MP3 Encoder. The developers of LAME believed they had a better way to encode (create) MP3 music files. So they published their source code as “educational” material and made it freely available. If one was building an MP3 encoder — which they were not — then this is how one might go about doing it. The simple act of publishing source code on the Internet, available freely and instantaneously to anyone in the world, sparked a collaborative effort resulting in the rapid development of an MP3 encoder with "speed and quality to rival and often surpass all commercial competitors".
Is publishing software source code on the Internet — an act of free speech — something we can or should constrain?
2. When is a CD ROM an HbA1c test reader?
At a hackathon in Singapore this fall, Rahul Rajeev built an HbA1c test strip reader using an old CD ROM, a hobbyist computer, and some other readily available materials for less than $100. In 24 hours. A visit to Quantified Self, Geeks with Diabetes, Hacking Medicine, or Diabetes Mine yields dozens of examples of people on the vanguard, rapidly developing creative, inexpensive solutions to tough medical problems.
How can we increase the rate of diffusion of value from those who are combining readily available, inexpensive components to produce new capabilities?
3. How do you safely automate an oil refinery?
I spent my early career architecting, designing, commissioning, and supporting the ongoing operation of automation systems for oil refineries and chemical plants all over the world. These systems are assembled from plug-and-play components sourced from many suppliers. They are installed to meet the unique needs of a particular facility, often while the facility is in operation, during projects with huge cost and schedule pressure. Valves, transmitters, controllers, distributed control systems, advanced model predictive controllers, operator consoles, process historians, emergency shutdowns systems, servers, operator training simulators, and hundreds of other subcomponents are brought together from multiple vendor. These vendors are often competitors and must collaborate for project success. This seems pretty risky considering the hazards involved, yet it has been common practice for decades.
What can we learn from other industries that quickly assemble plug-and-play components to build life-critical systems that are safe and effective?
4. When does grape juice become wine?
This spring I was visiting diabetes clinics and hospitals in eastern Canada and had a chance to visit my brother, who lives in Halifax, Nova Scotia. We sampled some of his excellent homemade wine. Homemade wine is very popular in Canada due to heavy taxation of the alcohol sold in government monopoly liquor stores. The business of homemade wine has now reached the point where from a single store you can buy the grape concentrate, bottles, and corks, you can rent or borrow the corker, carboys, and other production equipment, you can rent space in the store for your wine as it ferments, and you can purchase services for someone to perform the myriad activities involved in turning grape juice into wine. “Homemade” has advanced to the point that all you have to do is select the kind of wine you want, sprinkle a little packet of yeast into the carboy, and return later to pick up your untaxed wine.
How far can the definition of “homemade” be stretched?
OK, we can do things quickly. So what?
As the vignettes above illustrated, it is now easier than ever for existing technologies to be rapidly recombined to create new value. Now on to the question of motivation.
In November at the D-Data Exchange at Stanford University organized by Tidepool and Diabetes Mine, we were treated to a number of examples of people who are already doing amazing things. Their efforts were summarized with the clarion theme “we are not waiting”, which I will now expand upon with six assertions related to motivation:
1. Diabetes is dangerous
There are 8,766 hours in a year. In just a few hours, blood glucose can go from safe to dangerously low or high. Even assuming it takes eight hours for blood glucose levels to become life-threatening, that’s still over a thousand chances to die per year. Three times a day, every day for the rest of your life.
You can never take your eye off the diabetes tiger or it will bite you. Even with our family’s background and motivation and experience, our son Hayden — in one of those “perfect storm” moments — recently hit a blood glucose of 22 mg/dl. If not for the intervention of his friend, he might not be here today.
Children with diabetes and their parents try to manage, hour by hour. They try and try and try. The burden is crushing and the improvements are coming too little, too late. We are not waiting.
2. We have a lot in common with jellyfish and viruses
The most powerful motivation in the biological universe is species perpetuation. We want our progeny to survive and thrive. Humans share this motivation with every life form, from viruses and jellyfish on up. So to mess with this motivation is a perilous undertaking that will not succeed.
We want our children to be safe and happy. And we know how to do it. We are not waiting.
3. We can, we should, we will
When a parent takes the instructions for assembling a CGM (continuous glucose monitor) remote monitoring system from a website, downloads educational software from a source code repository, and runs it on a commercial cloud service with the press of a button so that they can monitor their child’s blood glucose at night, who shoulders the greatest responsibility if something untoward happens? We do — those of us who live day in and day out on the glucose roller coaster.
Some in the medical establishment have for decades challenged the notion that people living with diabetes can self-manage. When the first primitive blood glucose monitoring strips were introduced in the 1970s, some doctors said their patients couldn’t, shouldn’t, or wouldn’t prick their own fingers.
- “It’s too hard for them.”
- “They will make dangerous decisions with the results.”
- “They won’t bother”.
Fast forward four decades: Children now administer their own insulin at school from pumps containing a lethal dose of insulin. As families, we already shoulder enormous risk. We are not waiting.
4. We aren’t starting from scratch
Consider three types of software to reduce the burden of diabetes: closed loop feedback control algorithms, remote monitoring apps, and retrospective decision support tools. In each case, the fundamental technologies are freely available. The software and algorithms are published in papers, blogs, theses, textbooks, and open source repositories — available to anyone. The hardware is commercially available.
Just like a pressure cooker, the original intended use of the device or algorithm may have nothing to do with diabetes. The example may come from a completely different industry — like commercial aviation, finance, automotive, supply chain management, power generation, or oil refining. But their applicability to diabetes is undeniable. We are not waiting.
5. We still need great devices
Companies that make physical components of this ecosystem — sensors, pumps, smartphones, insulin, strips, cellular networks, cloud servers — will continue to survive and thrive because these elements are necessary to instantiate the harmless mental abstraction we call software.
The system will continue to comprise three essential components: hardware, software, and wetware (humans). What will change is how — and how fast — these systems will be developed. Closed, proprietary systems will die. Open, standards-based, interoperable devices will thrive. The problem and its solution will get closer in person, space, and time. We are not waiting.
6. The future is already here
The artificial pancreas already exists in the wild. So does the ability to remotely monitor your child’s blood glucose continuously while they’re at school or asleep. So do decision support systems to adjust settings on insulin pumps. Harmless software, combined with readily available hardware, is already doing immeasurable good. But it’s only helping a handful of families living with the burden of diabetes. To quote science fiction author William Gibson, “The future is already here, it just hasn’t been evenly distributed yet”. We are not waiting.
Tidepool is at the vanguard of an exciting confluence of ability and motivation: The abilities brought about through the incredible pace of technology development, and the motivation of people personally touched by diabetes, tempered by experiences gained in other high-risk endeavors. We are closer than ever to the prospect of a safer, more effective, less burdensome life with type 1 diabetes for millions of people.
The meanderings and eddies and backwaters and dams in the value stream are going to be removed. Value is going to flow more freely, more quickly, to more people. Everyone will benefit except those who resist the inevitable flow of progress. I encourage you to support Tidepool and become part of the solution.
— Lane Desborough
References / additional reading
- Software is eating the world: http://online.wsj.com/news/articles/SB10001424053111903480904576512250915629460
- The LAME MP3 Encoder: http://lame.sourceforge.net/about.php
- Hacking A Reader For Medical Test Strips: http://hackaday.com/2013/12/13/hacking-a-reader-for-medical-test-strips/
- Chinese man keeps himself alive for 13 years with HOMEMADE dialysis machine: http://www.dailymail.co.uk/health/article-2266412/Chinese-man-kept-alive-13-years-HOMEMADE-dialysis-machine.html#ixzz2o3dpG6zk
- Oil refinery automation: https://www.honeywellprocess.com/library/marketing/whitepapers/WP-LegacyControlSystem-Marathon-Petroleum.pdf
- Homemade wine in Canada: http://www.winebutler.ca/make-wine/
- Glucoaster: http://www.diabetesmine.com/
- ePatients: http://strata.oreilly.com/2012/01/epatients-empowered-patients.html
- Can we hack our way to better health care: http://vectorblog.org/2013/10/can-we-hack-our-way-to-better-health-care/