I just bought my first smartphone about a month ago, and trying to work it, I’ve never felt dumber in my life.

That's why I found it so fascinating to attend what some call the "epicenter" of the diabetes tech world, the DiabetesMine D-Data Exchange at Stanford University School of Medicine, on Thursday, Nov. 19.

This was the 5th such event organized by our fearless leader Amy Tenderich -- now held twice a year, in June right before the big American Diabetes Association Scientific Sessions, and in November just before the somewhat larger and broader DiabetesMine Innovation Summit. The ExChange mini-summit is focused squarely on deep-tech development, and gathers about 100 key innovators working on apps, algorithms and platforms that leverage D-data, as well as device makers and tech-savvy patients who both use and are often involved in creating this tech.

Yes, this is the incubator that launched the transformative hashtag #WeAreNotWaiting back in November 2013.

I expected not only to be out of my element, but to be totally overwhelmed. In fact, I fully expected not to understand a single word spoken. I came armed with caffeine and I sat in the back row, in the corner, like a naughty schoolboy. 

It turned out the program was far more approachable than I expected. Here's a look at the agenda:

DiabetesMine D-Data ExChange agenda - November 2015

Toward a Common D-Language

In fact, right off the bat, keynote speaker Dr. Joseph Smith started talking my language. Smith, who is Chief Medical and Science Officer of West Health Institute (with a killer resume that includes being named one of “20 People Who Make Healthcare Better” by HealthLeaders Magazine in 2010), compared the health technology world to a symphony orchestra. Paraphrasing (heavily) he said that you can have a group of the best damn musicians in the world, but if they are all marching to the beat of their own drummer, they sound like hell. If, on the other hand, they play together from the same score, you can have the most beautiful music imaginable. In the actual words of the good doctor: "We have to move from cacophony to symphony."

Joe Smith of West Health at DData15He notes that industry has made tech tools basically in a vacuum, just assuming each component works by itself -- whereas they've haven't thought about the whole ecosystem and how these devices all need to be connected. For example, the reality of an ICU is that most everything is still documented by hand, and whatever data exists is locked into specific devices. Interoperability is the missing “ether” here.

Right, the diabetes tech world is filled with awesome toys created by companies with a device focus, but what we need are toys built with a system focus. Your meter, pump, CGM, Fitbit, apps, and smartphone should all be able speak the same language and to share data with each other seamlessly. This is summed up by the haiku-in-a-single-word: INTEROPERABILITY (the theme of this Fall's event being Making It Happen).

Change must be systematic, Smith says, so we no longer just have a collection of “gifted gadgets." Why is it that other industries have interoperability, but we don’t in diabetes and healthcare?

“We need to demystify and democratize healthcare, removing it from the domain of the high priests," Smith said, followed by his strong conviction that the real solutions will come from us, in the community (see Nightscout as an example).

Throughout the day, speakers urged the device makers to realize that the market will actually reward open-source system-integrated designs more than it will the traditional proprietary siloed machines. The message: Being open won’t cause your downfall, rather it will propel you to success.

 

What Nightscout Did

Dr. Joyce Lee of the University of Michigan, aka @doctorasdesigner presented a whole bunch of new data on the Nightscout/CGM in the Cloud community, and use of this DIY remote glucose monitoring system.

Lee received a grant from PCORI (Patient-Centered Outcomes Research Institute) to support the creation of a "collaborative innovation research network" in type 1 diabetes. She and her team conducted a large-scale survey of Nightscout followers -- the largest type 1 diabetes online community on Facebook that is actively engaged in creating collaborative tools through participatory design. If you haven't checked lately, there are now a dizzying array of options for your Nightscout setup!

Some fascinating results of Lee's study, involving 1,276 members of the CGM in the Cloud Facebook group:

  • almost 20% of people say they started using CGM because of Nightscout itself - that’s huge!
  • 6% said they actually switched to Dexcom CGM from another system just because of Nightscout
  • >80% of current CGM in the Cloud members say they plan to stick with Nightscout, even if commercial alternatives become available
  • >85% found the technology “extremely empowering"
  • self-reported A1c results showed that glycemic control improved meaningfully (0.7%-1.2%)

But our favorite results were those where people were asked how much diabetes kept them from doing normal activities, spending time with friends, etc. before and after their Nightscout use. Wow! There seemed to be a huge reduction in the "invasiveness" of diabetes.

 

The Ultimate UI (User Interface)

Speaking of beautiful interfaces, past DiabetesMine Patient Voices winner Sara Krugman—who is now with Tidepool—gave a killer presentation describing how she designed the user interface for the iLet bionic pancreas. This uber-creative interaction designer started off by sharing her decidedly mixed feelings about her current tech: She once made 100 plaster casts of her pump and smashed them against a wall as part of a piece of cathartic performance art.

But Krugman’s challenge went to the core of developing tech: “What is it like to wear a (fully integrated) artificial pancreas?” she asked. “No one knows!” because no one has done it yet. For the iLet, she created piles of sticky notes of “need to knows” and “nice to knows” and from them developed a system architecture. In designing the user interface that overlays the architecture, she totally stepped out of the box and even re-defined the way users would record a day, moving from "breakfast, lunch, dinner" as button choices to a "beginning-middle-end-sleep" model, recognizing that many people do not live by clocks.

She even went so far as to think about the subconscious messages that various type fonts and sizes give us, and chose appropriate fonts and patterns of capital letters for the pump’s functions. For example, "DONE" in all caps is screaming and "Done" with an initial cap is a command, whereas "done" in all lower case is a simple user recognition of completion of a task.

In the end, despite a universal lack of user experience to tap into, Krugman created a simple, logical, and beautiful first draft of the user interface for the next generation of insulin delivery systems that will get people talking.

 

FDA on DIY D-Tools

The FDA has come a long way, Baby! Not long ago, the idea that the agency would address do-it-yourself solutions or actually help open-source data platforms to thrive seemed far-fetched at best. Now in the past two years, we’ve seen real action.

Stayce Beck, FDA's Branch Chief for the Diabetes Diagnostic Devices Team, gave a great overview of FDA's approaches, with a generous helping of encouragement for this developer crowd. Her slides were decidedly different too, departing from the dry text-only stuff we usually see from FDA to include movie references to Back to the Future and Star Wars!

She stressed that FDA's main job is to look at all the things that could possibly go wrong with any new tool or treatment, and do its best to mitigate that risk. After all, people do crazy things, and while you can't anticipate everything, you need to test the device thoroughly with users, especially the non-tech-savvy. There was some discussion about FDA's Human Factors requirements being focused on this risk element, rather than attempting to serve as guidelines for pure good design.

"Don't think of FDA regulation as 'red tape' but as meant to raise red flags if needed," Beck implored the group.

In addressing DIY technology, an intriguing question was posed to FDA by Tidepool CEO Howard Look: Is publishing instructions on how to make a homemade medical device different than actually providing that device to the masses? Beck could only say that "there is no hard, clear line" and FDA would have to consider instances on a case-by-case basis.

We're pretty sure Beck broke Twitter records when she openly posted not only her email, but her office telephone number, at the end of her presentation; and invited a two-way conversation between her agency and developers early in the design process, to help speed up the approval process.

 

Deep Tech Panels

The half-day whirlwind event featured two panel discussions—one looking at early successes in interoperability and one on standards—and a series of mini-presentations on new devices and apps being developed.

The Interoperabililty panel, led by Adam Brown of Close Concerns, explored the definition of this term, and then dove into good and bad examples by representatives from Open mHealth, Human API and MedHelp, which btw just announced an expansion of its diabetes management platform with its Sugar Sense app for Apple Watch. Also on this panel was Jorge Valdes, Chief Technical Officer of Dexcom, and it was HUGE to hear him giving kudos to Nightscout and the #WeAreNotWaiting movement for helping to usher in faster R&D and regulatory clearance of mobile device connectivity and data-sharing. He acknowledged that Dexcom itself was trying to get there faster, but simply couldn’t do it as quickly as the community could.

Also in the house was Annika Jimenez, Dexcom's new Senior VP of Data, hired specifically to focus on next-gen data platform, analytics, and software; and Nate Heintzman, who joined Dexcom last year as a Visiting Scientist focusing on User Innovation and Collaborative Research. In other words, Dexcom is all over it!

Worst interoperability example cited by this panel? Fitbit. A lot of the Alpha User types present were harping on how bad the time stamp function is in this tool, which makes integrating data an enormous chore.

"If you want to be a walled garden, you’d better be Apple. If you don’t embrace openness, you are dead. Open up the data!" was the rally cry. And the panelists emphasized "you have to do it right – over and over, we’ve seen companies opening up data and then putting one person on it and forgetting about it." That won't cut it, they said.

The Standards panel is where I actually started feeling like that lost schoolboy again. This was seriously deep-tech talk about FHIR H7 standards and how they facilitate importing and exporting data from EHRs (electronic health records). This is where the whole DiabetesMine team simply nodded in appreciation of the experts pushing to make this stuff happen. The panel was moderated by Peter Levin of Amida Technology, who's a Stanford consulting professor in  Aeronautical and Astronautical Engineering, and former Senior Advisor to the Chief Technology Officer of the U.S. Department of Veterans Affairs, among other things. Wow-smart, in other words.

There was some heated debate about the role of standards, and how necessary they really are. We have to hand it to Melanie Yeung of the Canadian Centre for Global eHealth Innovation, the group working on diabetes-specific data standards, for holding her own on the importance of this work (not to mention as the only woman on the panel).

 

Show and Tell: #OpenAPS and More

My personal favorite in the "Lightening Round" of demos was Sean Saint’s prototype smart pen. It’s that Bluetooth re-fillable inPen that we introduced here at the 'Mine with an extensive article this summer. It automatically beams dosing data to an app that tracks insulin-on-board (IOB) and more. The prototype app has a beautiful interface that also provides a dosing calculator and some great graphs showing history and usage. Think of it as pump-smarts for a pen. The app also features dose reminders, an easy-to-check most-recent dose screen, and—coolest of all—a truly hot feature: A temperature alarm that not only alerts you if your insulin exceeded the maximum recommended temperature, but tells you how long your insulin was in the “red zone.” I hope Saint’s pen is available by the time I run out of Snap supplies. I would totally use it.

The conference closer was an amazing talk by Chris Hannemann, a U.C. Berkeley alumni and mechanical engineer who in August became the 5th person on the planet to go live with what is known as OpenAPS, a fully functional homemade artificial pancreas. I kid you not! Fifteen people and counting are now using this open source closed-loop system, which is made up of a jumble of products cobbled together: a Medtronic insulin pump, Dexcom CGM receiver, raspberry pi device for running Linux OS, a CareLink USB stick to enable communication to the pump, and a battery pack. Whew...

Chris himself was diagnosed with type 1 at age 8, and has a sister who got T1D at age 7, and his father has type 2 DiabetesMine D-Data ExChange brochure - November 2015diabetes. So it's no surprise that he jumped in with both feet as soon as he discovered the work of Dana Lewis, Scott Leibrand, Ben West and others on this #OpenAPS system in March of last year.

He now contributes to the code and documentation, and he shared data from 83 days of his own OpenAPS experience. He saw a 13%-point improvement in daily time in range (up to 83%, equivalent to under a 6% A1c) and a 15 mg/dl decrease in daily mean glucose (to 129 mg/dl) since starting the system. He said he wakes up pretty much every morning at approximately 100 mg/dl, and his Dexcom Clarity output showed an incredibly narrow spread of values in the early morning hours.

He talked about how lovely it is to be able to sleep through the night without BG worries, and shared other insights into the feeling of wearing this system. And (somewhat surprisingly), he emphasized how important it is for patients not to become too reliant on tech tools, at the expense of really understanding what's happening in their own bodies. "Is automated insulin delivery actually a 'tool' or a 'therapy'?" he mused. 

Finally, out in the foyer, folks were hovering around the unusual work of Justus Harris, Visual Artist and Technologist, who has developed a method for creating sculptures out of blood glucose data. We featured his work here recently. He brought color and excitement to the day, 'cause who doesn't love a bright shiny thing you can hold in your hand, as opposed to a bunch of charts of data? (OK, maybe most attendees here would choose charts :)   

 

"Alpha Users" Lead the Way

There was much talk about the Alpha Users: The folks at the cutting edge. That’s not me. I realize that I was one of the few Omega users in the room. In my case, the last adopter. It’s a bold new world, and I’m loving it. But it has its challenges. For this report, I tired to ask Siri when I bought her and she said, “Who, me?” Then I tried to voice-text my wife and apparently said, “The conference is on fire,” instead of saying, “The conference is fine.”

I’m sure if she’d been wearing a Fitbit, her heart rate would have jumped when she got that missive. (And now I know the time data on that might be hard to decipher -- oy.)

So, returning to Dr. Smith's analogy, the violins aren’t in sync with the flutes just yet, but the music is getting better. And in the future it will be a beautiful symphony indeed.


{See also: diaTribe's 'Closer Look' report on this latest DiabetesMine D-Data ExChange}


Disclaimer: Content created by the Diabetes Mine team. For more details click here.

Disclaimer

This content is created for Diabetes Mine, a consumer health blog focused on the diabetes community. The content is not medically reviewed and doesn't adhere to Healthline's editorial guidelines. For more information about Healthline's partnership with Diabetes Mine, please click here.