Screen shot. Looks even better live! :)
The hormonal system of the human body is complex, and many outside factors affect things like blood glucose levels. People with diabetes need to self control their blood glucose levels. To do that, they need to learn what makes the blood glucose rise and go down.
Wellness trackers provide easy, perhaps fully automated, or even fun ways to track many signals. Learning from that data, and converting the information into behavior change is still a challenge.
We believe Personal Health Records like Omakanta make it easier for innovative companies to develop services that help make sense of all the data. PHRs foster collaboration between companies, and put people in the center of it all, allowing citizens to control where their data is used and how. We hope to see ecosystems emerging on top of these platforms.
For instance, Mikael from our team has a startup Sensotrend that concentrates on making life easier and providing insights for one small user group (insulin-dependent people with diabetes) really well. He still sees that people with diabetes are all individuals with their own needs and challenges, and there's no way one app or service can solve all of those. People need to be able to choose from a variety of apps, that can all work with the same set of collected data.
The real health benefits and business case for services like this come from the way we enable healthcare organizations take a major step towards digitalization.
Healthcare organizations can already separate their customers (the patients) into segments. Most people with chronic conditions are already empowered and independent, and capable of managing their own condition independently. Only a small minority requires all the possible attention.
To be able to categorize people with chronic conditions, and to keep tracking the categorization (things change in life, as we know), the organizations need the self-monitored and self-reported data from these people. People will submit their information when provided with easy-to-use tools that bring them real value (like ours), even more so when rewarded with benefits like possibility to manage appointments and ordering supplies autonomously. Healthcare organizations can enable these features for people who do submit their self-measurement data often enough, go take their lab results regularly, and don't have hospitalizations or other events warranting a meeting with a healthcare professional.
With the help of this categorization, healthcare organizations can reduce cost from where it is not required (over-serving people that do well) and target their resources in a meaningful way. This is the easiest place to save. It also has the most potential for saving. And it's logical to start the digitalization efforts with this user group, that contains the natural early adopters.
What it does
We take information from wellness apps and devices (in this case Oura ring and a continous glucose monitor and an insulin pump system) and visualizes that information together. So people with diabetes can see and learn, for instance, how lack of recovery causes stress, which causes blood glucose to rise, and how that can (once you understand the cause) controlled with medication, but also with changes in behavior.
How we built it
We're using open source components Nightscout (nightscout.info), xDrip (http://stephenblackwasalreadytaken.github.io/xDrip/), and other components (https://github.com/pazaan/640gAndroidUploader) built by the #WeAreNotWaiting community (see http://www.healthline.com/health/diabetesmine/innovation/we-are-not-waiting) to read data from medical devices.
We're integrating directly to the cloud API of the Oura ring, as well as some other wellness trackers, and use aggregator services like Wellmo to extend our support to dozens of different apps and devices, as well as manual tracking. Wide coverage of supported apps and devices gives our users the freedom to choose the tools that best suit their individual lifestyles. We believe this is absolutely essential!
We use Sensotrend's exising technology to fetch the data and mash it up as insightful visualizations. We don't store any data in our (or Sensotrend's) service, rather fetch it from wellness apps directly, and store medical data to KanTa (or Taltioni, another Personal Health Record system). This way the data is always in full control of the citizen.
Both KanTa and Oura connections were built from scratch during the Junction hackathon, demonstrating how easy it is to integrate systems. Almost all modern API's are based on OAuth2 authentication and authorization, JSON data format and REST calls.
Challenges we ran into
KanTa system had a problem with the firewall, not all interfaces were fully functional for developers outside the organization (with a VPN access, everything works...). Oura system worked well for us initially, but some other developers required more relaxed security features, and when Oura enabled those it caused some trouble for us.
A very positive challenge was that we weren't able to fully concentrate on hacking, as there was a semi-constant stream of partners, people interested in diabetes treatment, and especially government officials, allowing us to discuss and share our vision of the emerging ecosystems on Personal Health Record platforms.
Accomplishments that we're proud of
We had very good discussion and feedback both ways with both KanTa and Oura representatives. We hope we have given them valuable feedback, and also encouragement that their APIs and services are really required and being used. The same for us, we've received a lot of feedback stating that there is a true need for services like ours.
What we learned
We had some interoperability issues with both Oura and KanTa. Even when an API is well specified and tested internally in an organization, you can really tell whether it works after you bring it to a hackathon or interoperability event.
What's next for diabetesdata
We're actively participating in the development of KanTa system, building integrations to sandbox environments and helping define the data structures. We are also eagerly waiting for the production environment to become available. We're also continuing our discussions with the public healthcare organizations, and plan to bring our service (or services) online in production as soon as KanTa is ready for it.