My buddy built Loop, the iOS app for managing this (which, thru another org just got FDA approved). I was living near him in Oakland when he was first building it and I just feel really proud of what he was able to do not only for himself but for others.
For all of you out there who are trying to use tech to solve your own problems, please keep at it, one day your work may help thousands or millions and be featured in Nature.
My spouse is also using Loop. It's a huge improvement over just the pump alone. Even just changing basal settings is easier in the app versus the pump device interface, which is no longer required using Loop on iOS with the OrangeLink device.
https://loopkit.github.io/loopdocs/
They have a new web browser build method using TestFlight, which no longer requires an up to date Mac running the latest version of Xcode. The web build mode also enables someone to update the Loop app using only their smartphone. Something which is handy for travel or long periods of time without access to a Mac. It only lasts 90 days instead of the 1 year of the Xcode build, but is easy to rebuild on TestFlight.
That's impressive as hell. The FDA is, by nature, a very conservative and slow-moving organization. They set a very high standard of evidence for anything that's actually called "medicine". (As opposed to supplements and devices that pretend not to make medical claims, in which they are largely hamstrung.)
It takes a ton of effort to get FDA approval. Navigating the process is expensive and aggravating.
I work in the space, and back in 2018 the FDA was communicating very clearly that they wanted to see someone make a closed loop artificial pancreas. They do go a little easier on companies that are developing the new stuff they want. But I suspect the bigger companies didn't want to touch it because it's inherently risky: any minor bug or glitch can literally kill children any you could lose the whole company.
It's awesome that the DIY / hacker / open source folks were able to step up to the opportunity and build this great think, but it is all very much proceed-at-your-own-risk: there's nobody with a big pile of cash to sue if you or your loved ones get hurt.
I'd love to see more of these sorts of trailblazing initiatives in the future, and it's tremendously encouraging that the FDA is apparently open to this as well.
For all of you out there who are trying to use tech to solve your own problems, please keep at it, one day your work may help thousands or millions and be featured in Nature.