What is #OpenAPS?

The Open Artificial Pancreas System project (#OpenAPS) is an open and transparent effort to make safe and effective basic Artificial Pancreas System (APS) technology widely available to more quickly improve and save as many lives as possible and reduce the burden of Type 1 diabetes.

OpenAPS means basic overnight closed loop APS technology is more widely available to anyone with compatible medical devices who is willing to build their own system.

We believe that we can make safe and effective APS technology available more quickly, to more people, rather than just waiting for current APS efforts to complete clinical trials and be FDA-approved and commercialized through traditional processes. And in the process, we believe we can engage the untapped potential of dozens or possibly hundreds of patient innovators and independent researchers and also make APS technology available to hundreds or thousands of people willing to participate as subjects in clinical trials.

The body of work by the OpenAPS community includes a safety-focused reference design, a toolset, and an open source reference implementation that can be used by any individual – or any medical device manufacturer. We believe this will in turn allow manufacturers (and the academic research teams they work with) to turn more of their attention to designing and testing more advanced APS systems, and thereby accelerate the pace of innovation toward new and improved Type 1 diabetes treatments, and eventually a cure.

Please note that OpenAPS community efforts will be open source and free for use for other people, open source projects, researchers, and non-profits to use, and available on an open and non-discriminatory basis for all commercial manufacturers to use in proprietary products if desired.


Frequently Asked Questions

What is OpenAPS?

  • OpenAPS is an open and transparent effort to make safe and effective basic Artificial Pancreas System (APS) technology widely available to more quickly improve and save as many lives as possible and reduce the burden of Type 1 diabetes. The community has created a safety-focused reference design and a reference implementation of an overnight closed loop APS system that uses CGM sensors’ estimate of blood glucose (BG) to automatically adjust basal insulin levels, in order to keep BG levels inside a safe range overnight and between meals.

How does OpenAPS work?

  • OpenAPS is a simplified Artificial Pancreas System (APS) designed to automatically adjust an insulin pump’s basal insulin delivery to keep blood glucose (BG) in a safe range overnight and between meals. It does this by communicating with an insulin pump to obtain details of all recent insulin dosing (basal and boluses), by communicating with a Continuous Glucose Monitor (CGM) to obtain current and recent BG estimates, and by issuing commands to the insulin pump to adjust temporary basal rates as needed. It follows the same basic diabetes math that a person would do to calculate a needed adjustment to their BG – but it’s automated and precise in its measurements. For more details, read the OpenAPS reference design.

Where can I go for help when building my own OpenAPS implementation?

  • You’ll have to build your OpenAPS implementation yourself (no one can/will do it for you!), but there is documentation and advice from a community of people who have already closed the loop. Make sure to 1) read the reference design to understand how the system is designed to work; 2) join the openaps-dev Google Group mailing list to read through previous troubleshooting guides and tips; and 3) introduce yourself and start asking questions in the Gitter chat room where most of the OpenAPS users hang out. There are users around the world, so there is usually someone online in Gitter who can help answer questions and point you to resources. (You can also use the “search” feature on the web version of Gitter to search through previous help conversations, such as help with uploading data or building your cron job). And much of what you’ll need is available in the documentation.

Does using an OpenAPS implementation mean I won’t need to bolus for meals or ever push any buttons?

  • No. OpenAPS is what a lot of people call a “hybrid closed loop”. Because the current insulin we have available to us takes so long to reach peak activity, it’s better to do (most of) the manual meal bolus as you would otherwise do. There are features in the OpenAPS algorithm to help if BGs rise faster or drop faster than expected during or after a meal, but they don’t replace a regular meal bolus. Overall, it’s important to understand that OpenAPS is not a “set and forget” type of system. You’ll still be actively managing your diabetes and doing basic self-care as you were before – this includes everything from meal boluses, checking BG and calibrating the CGM, changing out pump sites, etc.

Is OpenAPS safe?

  • Yes. It’s designed for safety, by people with diabetes. It is designed to monitor and respond to changes in BG levels, ultimately lessening the numbers of high and low BGs a person with diabetes may deal with. There will still be some highs or lows, but most users self-report less of both highs and lows, plus more time in range, AND hbA1c reductions – not to mention the quality of life improvements associated with having a system that can auto-adjust basal rates overnight while they sleep.

Are there any kids using OpenAPS?

  • Yes, dozen(s). Approximately one third of the OpenAPS loopers are kids who have parents who have chosen to build closed loops for their families.

Is it a PID controller?

  • Because OpenAPS is a platform and not a product, each user can implement whatever algorithm he or she feels works best. That being said, most people are using an MPC—model predictive control—algorithm that estimates future glucose based on current glucose, insulin doses, carbohydrate consumption, and some personal configurations. The exact inputs, and thus how the system makes its decisions, differ from user to user. Other algorithm types—PID, fuzzy logic, various machine learning techniques—have been studied by academics for use in closed-loop insulin dosing systems. Each has its own pros and cons, but one of the reasons MPC is frequently used is because of its simplicity. People with diabetes already treat their disease using a simplified MPC in their head, so sticking within that framework helps build trust in the system and understand why it decides to take certain actions. Some of the more complex controllers can become black boxes to the user, obfuscating the rationale behind its decisions. Not only can this be dangerous with an experimental system, but it can also lead to frustration, even when the system is “doing the best thing possible.” Read more about the design of the OpenAPS algorithms in the reference design.

Can I buy an OpenAPS?

  • No.

Where/how do I get an OpenAPS?

  • OpenAPS is not an FDA-approved system or device. It’s not manufactured or sold anywhere in the world. It’s an open-source designed system that you can choose to build yourself.

What do I need to build an OpenAPS implementation?

  • For the most up to date hardware list, check out the hardware page in the OpenAPS documentation. In general, you need a compatible insulin pump; a continuous glucose monitor; a “controller” device (like a Raspberry Pi, a small computing device); and a “translator” device (such as a Carelink USB stick) to communicate bi-directionally with the pump.

Where is the code for OpenAPS? Where is the documentation for OpenAPS?

Do I have to know programming or be technical to build an OpenAPS?

  • No, but you DO have to be willing to learn a few technical things. (Not necessarily a programming language, but basics like Linux command line.) There are resources to help you figure it out. The DIY part of OpenAPS is important, because there’s no customer service if it breaks or stops working – you’ll have to fix it yourself. So the building process is valuable because you’ll be well-equipped to figure out what might have stopped working (and why!) once you have a working closed loop implementation.

Will my pump work with OpenAPS?

  • Right now, the only pumps that work with the OpenAPS safety-focused reference design are older versions of Medtronic pumps. Find more details about which versions of the pumps are compatible in the hardware section of “Phase 0” of the OpenAPS documentation. There is also a Slack channel for people from the broader #WeAreNotWaiting community who are working to understand communication with other insulin pumps.

Will my CGM work with OpenAPS?

  • Most CGMs work with OpenAPS, including: Dexcom G4 (including with Share); Dexcom G5; and Medtronic CGMs.

Does it work with an iPhone or Android based phone?

  • Make sure to read the above questions (or the documentation) about what equipment you need for a standard #OpenAPS implementation. You can *view* your data on a phone or computing device, but most OpenAPS implementations still require additional hardware like a Raspberry Pi or Intel Edison in order to run the closed loop system.

How much does OpenAPS cost to build?

  • It depends on how much equipment you already have. The hardware components (Raspberry Pi, battery, Carelink USB, etc.) usually cost under $100. But many people also need to find a compatible pump, which ranges from free (ask a friend or around your local community) to several hundreds of dollars. Also be warned: most people end up building a second “rig” (hardware setup) or more so they can have a “mobile” rig and an “at home” rig, etc.

Why would someone build an OpenAPS?

  • Individuals who build OpenAPS setups have decided that #WeAreNotWaiting to be able to automatically adjust basal rates to safely keep BGs in range and reduce the burden of living with Type 1 diabetes. For context, the first DIY closed loop that was the inspiration for OpenAPS was completed in December 2014 – back when the forecast for commercial closed loop systems was still 3-4 years away. And today (around May 2016), the forecast for a closed loop is still another year or more away.

What does the FDA think about OpenAPS?

  • You can Google it, or ask them yourselves! Our take on what they’ve told us directly as well as what they’ve said publicly: their responsibility is to regulate products on the commercial market and help safeguard the public. OpenAPS is NOT a commercial product and is not sold or distributed in anyway. Individuals who build an OpenAPS are essentially doing an (n=1) experiment, which they have a right to do to/by themselves. That is not a regulated activity by the FDA.

What do you think of (Bigfoot, Beta Bionics/iLet/Bionic Pancreas, Medtronic, TypeZero, Tandem, Cellnovo, Animas, Insulet, Roche, Cambridge)’s efforts to close the loop and create an artificial pancreas?

  • We support all of the companies working to commercialize a closed loop system. We just wish they would each get their solution(s) to market more quickly! This is part of what we’re working to do as a community. We are working to demonstrate the need of the community to have closed loop devices; demonstrate the safety of basic closed loop systems and efficacy compared to “standard of care” without closed loops; and help break down communication barriers between the FDA, these companies, and the community that needs the devices widely available as quickly as possible. And in the meantime, many of us are not waiting (#WeAreNotWaiting) by building and running our own DIY, OpenAPS implementations.

Who is behind OpenAPS?

  • There is an amazing community of people behind OpenAPS! Dana Lewis and Scott Leibrand built the first DIY closed loop system (called #DIYPS) that was the inspiration for the open source version of a closed loop system now known as OpenAPS. They spend a lot of time supporting the community in various ways, including continuing development on many of their algorithms that are now used by the community. One of the other founders behind the OpenAPS movement is Ben West, who identified the way to communicate with Medtronic pumps (in addition to other open source work he does with diabetes devices). Dana and Scott worked with Ben as he built the openaps toolkit, which became the building blocks of the modern OpenAPS implementation. Many other people have also contributed to the community in various ways: from writing documentation, building unit tests, testing new features and algorithms, incorporating OpenAPS data with visualization tools like Nightscout and making the data easily viewable on Pebble watches, building new translator devices to communicate with pumps, building new software for new controller devices to host algorithms, etc. You’ll find a lot of people from the community sharing their experiences using their OpenAPS implementations and documenting new tools in progress via the #OpenAPS hashtag on Twitter.

I have more questions. Who can I talk to?

In the News

There has recently been some great coverage & stories by/about some of the users and builders from the OpenAPS community. You may be interested in checking out some of their stories:

You’ll also see many stories shared on Twitter – make sure to check out the hashtag #OpenAPS to see more insights about what life with an #OpenAPS closed loop, DIY artificial pancreas is like!

From the #OpenAPS blog:

Real-World Use of Open Source Artificial Pancreas Systems – Poster Presented at American Diabetes Association Scientific Sessions

The OpenAPS community submitted an abstract that was accepted for a poster presentation at the American Diabetes Association’s Scientific Sessions meeting. The embargo lifted this morning, so you can read the abstract here or read below for the content from

#OpenAPS Winter 2015 update

The #OpenAPS community has made some amazing progress toward the end of this year. There are now, as of 12/31/15, (n=1)*22 people with DIY closed loops! (As of 4/28/16, it’s (n=1)*53!) And as always, several others in the testing and

Early Fall 2015 #OpenAPS Update

It’s been a busy few months for the #OpenAPS community, going from n=1 to (n=1)*7! (And as of 11/19/15, (n=1)*16.) And there are several people working on their implementation, so it might be possible to see a dozen OpenAPS implementations