CGM in the Cloud: A Community Unites to Tackle Diabetes

Email this to someoneTweet about this on TwitterShare on Facebook0Share on Google+4Pin on Pinterest18

I never thought I would say this, but it is an exciting time to have diabetes.

This decade has brought us continuous glucose monitoring systems (CGMs) that allow us to quickly measure glucose levels and transmit that data to handheld or pump-integrated receivers.

When I got my first CGM system in 2008, my husband Kevin, Principal Applications Architect for a major airline, saw potential in it that I never dreamed of. “What if I could listen in on the conversation between the transmitter and the receiver?” he asked me. “What if I could find out if you hadn’t responded to a low alarm? Set it all up on a dedicated computer here at the house…” This talk continued for months as he dug into the design of the device – the now discontinued Freestyle Navigator. I watched with interest as packets of data flew across his screen every minute. What if?

As a community of people with diabetes, we have fought hard for access to these devices. Then we waited for FDA approval in pediatric patients. We wait for industry, and the promise of each insulin pump’s answer to an integrated cgms/pumping system. We wait for researchers as they tirelessly work toward functional (artificial/bionic) and biological cures.

But more and more people have grown tired of waiting. And out of necessity, comes invention. Brilliant invention.

CGM in the Cloud is a concept first put forward by engineer (and father of a young son with type 1 diabetes) John Costik early last year that, in a nutshell, entails taking a patient’s CGMs device real-time data, sending it to the cloud, and then allowing access to that data – again, in real-time – in remote locations, such as a webpage, a cell phone, or a wearable, like the popular Pebble smartwatch.

CGM Pebble Watch - CGM in the Cloud

“When we first began uploading our son’s BGs to the cloud (in the form of a G4 connected laptop),” says Costik, “it was life changing. As our tools evolved to support uploading from an Android phone and viewing the data on a Pebble watch, we knew it was too good to keep to ourselves. The only way bring it to more people, without waiting years, was to make it open source.”

And bring it, he did.

With the addition of fellow D-dad and innovator Lane Desborough’s “Nightscout” software that allowed the data to be viewed on what Desborough calls “glanceable displays” throughout the home, the two men found that there was a community of people hungry for homemade solutions.

A few dozen families came together in a Facebook community called CGM in the Cloud, founded by D-dad Jason Adams, to support one another through the setup and implementation of the design.

“The group is an amazing assortment of people, motivated by love and a desire to help others,” says Costik. “The testimonials and stories of hope and gratitude are more than I ever expected.“

Now boasting nearly 7,000 families (and doubling from month to month), the CGM in the Cloud Facebook community is a social media phenomenon. Members are quick to help one another with technical issues from setup to webpage outages, appropriate use in schools to use among adult type 1s.

 

James Wedding, father to a daughter with type 1, found a role for himself in the group. “To me it was really the first time I had a place to make a real difference in how my daughter and others deal with type 1,” he explains. Wedding laid out the initial nightscout.info webpage, where new members to the group are first directed to explain setup, system requirements, and for troubleshooting, as well as the @nightscoutproj Twitter feed. He felt that as membership grew, the group needed a more efficient way “to keep people up to date with good information and support people attempting to get the system up and running.“

The community has become an international effort. Well known among the 7,000 members is Manipal (India) Institute of Technology computer science engineering student Rajat Gupta. Gupta has stepped in to help more than 120 less technologically oriented families set up, maintain, and upgrade their systems. He does it all as a volunteer.

“Contributing to Nightscout is one of the best ways for me to use my knowledge for a good cause and learn what difficulties do people face, and hence what solutions are appreciated,” explains Gupta. “I get a chance to speak to new families everyday from around the globe, mainly being the US, which is a great exposure to me. I hope that with all that I learn here, I will be able to make some cheap solution to these problems that can be easily made available to the developing countries.”

This spirit of sharing and support extends to the developers and designs within the community as well.

Kate Farnsworth, of Canada, is the mother of a daughter with type 1, and has played a crucial role in the Facebook community, “holding hands” and helping members troubleshoot.

“I am contributing to CGM in the Cloud because it changed our lives significantly,” says Farnsworth, “and I want to improve the lives of the families that have to deal with type 1 diabetes. I want to make life better for the kids who have to live with it, for the parents who have to worry all day/night about their kids and for the people with type 1 diabetes who have lived their whole life without this technology. I want to share this with everyone I can and shout it from the rooftops.”

After initially helping make sure the code could be translated into mmol for those, like her, outside the United States, Farnsworth turned her attention to the popular Pebble watchface for the Nightscout suite of tools.

“I didn’t like that it would report an error code (1-14) and would scare parents who were watching, thinking their child was dangerously low,” she explains. “I wanted there to be icons instead. I designed a new watch face and Christine Deltrap and Jonathan Moore rewrote the existing code to make my design work and improve the functionality of the Pebble watch so it is now amazing”

The many developers themselves, my husband among their ranks, work behind the scenes, contributing thousands upon thousands of lines of code to an open source code base and sharing emails about the development of features.

Jason Calabrese, involved since last summer through exchanges with Costik on Twitter, helped Gupta and Adams get people on Facebook started up on Calabrese’s early fork – or development branch – of Costik and Desborough’s original code.

“Since then, I’ve been working to add features, make the system more stable, and help other people get it running,” Calabrese says. “Due to lack of free time, it took me a while to move over to the community version.”

Lack of free time. Did I mention that this is no one’s day job?

In the beginning, Costik worked on the project while on FMLA time to transition his son to pump therapy and prepare for kindergarten. I’ve watched my husband Kevin spend countless hours cleaning up code and adding features on his own time. This is a community of people who want to make life better and management easier for people with type 1 diabetes.

My endocrinologist asked me this week, “But does it make diabetes too big a part of your life?” as she stared at the Pebble watch on my wrist displaying my blood glucose level.

“No, far from it,” I answered. “I have peace of mind that somebody knows I’m okay when I’m home alone with my toddler, when I’m away from home traveling, when I’m out for a walk. Parents can send children to sleepovers and school and not have to worry so much.”

I imagine what this kind of technology would have done for my mother, who kept me from driving until I was 18 out of fear of low blood sugars.

With each new device seems to come the fear of the patient (or their caregivers) having “too much data,” but as Costik explains, “Beyond the immediate knowledge of a BG, the understanding and intuition that the data brings has transformed how we manage type 1 diabetes. We’ve taken back so much of our lives from type 1, and we can count our son’s immediate and long term health as part of that.”

There’s a feeling among users – even those who may be frustrated by the technical aspects of implementing the system – of relief.

Kate Farnsworth says, “It has changed our lives. Our daughter is 11. She just started middle school this year. While we had the cloud set up last year, we also had someone at the school helping her out. This year, she is on her own with my back-up. It has meant that she can be “a normal kid” which as you know, is all a tween girl wants. She has way more freedom to go places now. She also loves that I have her back at all times. It lets her be a kid, and leave the tough stuff to me. Can’t put a price on that.”

Calabrese has had great success with his 7-year-old’s school, as well. “The way it’s been embraced at Andrew’s school has been amazing.  He has two different nurses and a health tech that monitor Nightscout during the day and enter treatments into the care portal. He’s now much safer at school and we all have more peace of mind.”

James Wedding adds, “It’s little things. Probably the best was letting our daughter spend a solid week dancing this summer. In the past, we’d have been depending on her feeling the lows to catch any that came along. With Nightscout, we were able to monitor and observe the trend line well in advance of a low that would require sitting out, and treat accordingly. She spent her time at dance engaging in intensive dancing, not watching, and I think Nightscout for that.“

Ben West, active in the development community and a type 1 himself, adds that some of his favorite stories are “the first walks alone with grandpa and first sleepovers.” West calls the whole Nightscout endeavor “an expression of freedom” for patients like himself and says that these stories “represent tiny bits of liberty and humanity restored.”

My husband Kevin has worked on coding an app that sends push notifications directly to his Android phone. While I’m wearing a standard Pebble watch connected to the cloud setup, he has a homemade app that bypasses the setup and has the CGM push data to the cloud as soon as it’s received rather than poll the cloud constantly. He’s got the system up and running on his phone and on the new Moto 360 piece of Android Wear and looks forward to extending the capabilities to other remote monitors.

Most importantly, he now knows when I’m not responding to a low alarm and can text or call me with a swipe of his app.

So what’s next for CGM in the Cloud? The future is limitless.

Calabrese rattles off features he hopes to see, “Pulling in data from pumps, adding #DIYPS like features, more reliable alarms, more customization, group communication using the care portal, there’s no end.”

Wedding agrees. “I would like to see the Nightscout database schemed to accommodate pump and other CGM data, making it possible for other systems to record against the same timeline, allowing overlay of data of different types in a single reporting system.”

With the Facebook group at well over 6,000 members, Ben West notes that “the community has resources to rival big vendors.” Looking forward, he says, “the Nightscout community has just barely scratched the surface of what’s possible when we work together and I’m curious to find out what is possible.” Both West and Costik are in early talks with the FDA.

Costik has high hopes for the future as well. “I am excited to see the addition of Medtronic CGM upload ability (awesome work by @JBerian), and the continued refinement of the existing tools (uploader, Pebble, Nightscout),” he says. “I would love to see the functionality we provide to be commonplace – under a unified set of (device agnostic) tools (Tidepool comes to mind).”

“Most of all,” says Costik, “I want to close up shop because a cure (functional or true) has put us ‘out of business’.”

Of all the things we are waiting for, I think we can all agree that we can’t wait to stop waiting for that one. 

For more information on CGM in the Cloud

visit https://www.facebook.com/groups/cgminthecloud/ and http://www.nightscout.info/

 

Melissa Lee writes the blog Sweetly Voiced.

 

 

 

Email this to someoneTweet about this on TwitterShare on Facebook0Share on Google+4Pin on Pinterest18

Comments (6)

  1. Kathy at

    I’ve been following this group too.  Its a tribute to pooling intellectual knowledge for the common good and finding success.  What they have already accomplished is phenomenal.  I can’t wait to see what comes next.

  2. Kathy at

    I’ve been following this group too.  Its a tribute to pooling intellectual knowledge for the common good and finding success.  What they have already accomplished is phenomenal.  I can’t wait to see what comes next.

  3. The author makes it seem as though the industry and the FDA are slow to respond because they don’t care.  I suspect that the opposite is true.

    Looking at Nightscouts web site I see this disclaimer toward the bottom.

    “All information, thought, and code described here is intended for informational and educational purposes only. Use of code from github.com is without warranty or support of any kind.
     
    Each element of the system can fail at any time rendering the system unusable. There is no password protected privacy or security provided by these tools; all data you upload can be available for anyone on the internet to read if they have your specific URL to view your data.”

    Legal liability is precisely why industry and the FDA have been slow to respond.  No medical product will come with such a disclaimer and the FDA won’t allow it.  No medical device company will release a product that hasn’t been fully tested to reduce their legal risk.

    Sign a waiver promising never to sue for any reason whatsoever and anybody will sell you anything. 

  4. The author makes it seem as though the industry and the FDA are slow to respond because they don’t care.  I suspect that the opposite is true.

    Looking at Nightscouts web site I see this disclaimer toward the bottom.

    “All information, thought, and code described here is intended for informational and educational purposes only. Use of code from github.com is without warranty or support of any kind.
     
    Each element of the system can fail at any time rendering the system unusable. There is no password protected privacy or security provided by these tools; all data you upload can be available for anyone on the internet to read if they have your specific URL to view your data.”

    Legal liability is precisely why industry and the FDA have been slow to respond.  No medical product will come with such a disclaimer and the FDA won’t allow it.  No medical device company will release a product that hasn’t been fully tested to reduce their legal risk.

    Sign a waiver promising never to sue for any reason whatsoever and anybody will sell you anything. 

  5. nl at

    I don’t feel as if “the author makes it seems as though the industry and the FDA are slow to respond because they don’t care » but I feel as if the author emphasizes on the need to open up research to a bigger world of programmers (open source), no matter where they come from. While I am happy products are fully tested and I understand companies have to reduce their legal risk, I also know that my priority towards my teenager is to reduce his risk of severe hypos while giving him a chance to live his life to the full. If I can help him not being alone in trying to reach safety, then I will do whatever it takes, with him and for him.
    Using CGM is by default a risky business that require a serious pilot on board to keep the plane flying not too high but then again, not too low either, while readjusting the horizon line with sugar or insulin, according to the CGM results on the dashboard (might not be the right word, I am French).
    For having used different CGM products in the past years, I can say that some products launched early on, that were not all that accurate, have been allowed on the market because the only way to evolve is not just to test but also to try on a big scale the product to correct its limitations and work together towards a better solution. We knew then the risks, yet we gave it a try and help the system get more efficient along the years.
    Type 1 diabetes remains a risky business, no matter what device you will use for a simple reason : the “pilot” has left the plane (the pilot’s name : ?cells). We, not a cell but a person, are now on the commands and we have no choice but to take control while making difficult decisions at time, based on too many intangible factors of life. 

    Companies decide to reduce their legal risk while our business managing type 1 diabetes is to take measured risks by having to command a body on our own.
     
    That is why I totally support this project and accept its limitations, because I believe my son should not have to bear the weight of such decisions on his own. Two heads are better than one and that system allows to lighten up the weight for some and share responsibilities that cannot be avoided. Elements of his actual CGM device (FDA approved) has already failed before and rendered the system unusable. But we know that we are in command so we took action to reduce risks knowing that it could happen. FDA approved or not, technology remains technology and can at any time : fail.

  6. nl at

    I don’t feel as if “the author makes it seems as though the industry and the FDA are slow to respond because they don’t care » but I feel as if the author emphasizes on the need to open up research to a bigger world of programmers (open source), no matter where they come from. While I am happy products are fully tested and I understand companies have to reduce their legal risk, I also know that my priority towards my teenager is to reduce his risk of severe hypos while giving him a chance to live his life to the full. If I can help him not being alone in trying to reach safety, then I will do whatever it takes, with him and for him.
    Using CGM is by default a risky business that require a serious pilot on board to keep the plane flying not too high but then again, not too low either, while readjusting the horizon line with sugar or insulin, according to the CGM results on the dashboard (might not be the right word, I am French).
    For having used different CGM products in the past years, I can say that some products launched early on, that were not all that accurate, have been allowed on the market because the only way to evolve is not just to test but also to try on a big scale the product to correct its limitations and work together towards a better solution. We knew then the risks, yet we gave it a try and help the system get more efficient along the years.
    Type 1 diabetes remains a risky business, no matter what device you will use for a simple reason : the “pilot” has left the plane (the pilot’s name : ?cells). We, not a cell but a person, are now on the commands and we have no choice but to take control while making difficult decisions at time, based on too many intangible factors of life. 

    Companies decide to reduce their legal risk while our business managing type 1 diabetes is to take measured risks by having to command a body on our own.
     
    That is why I totally support this project and accept its limitations, because I believe my son should not have to bear the weight of such decisions on his own. Two heads are better than one and that system allows to lighten up the weight for some and share responsibilities that cannot be avoided. Elements of his actual CGM device (FDA approved) has already failed before and rendered the system unusable. But we know that we are in command so we took action to reduce risks knowing that it could happen. FDA approved or not, technology remains technology and can at any time : fail.

Add a comment

Your email address will not be published. Required fields are marked *