New development for DIY Insulin Pump
July 4, 2019 12:15 PM Subscribe
For people who want/need to hack their own Insulin Pumps: From Scott Hanselman's blog:
"You probably didn’t see this coming, [Tidepool] announced an agreement to partner with our friends at Medtronic Diabetes to support a future Bluetooth-enabled MiniMed pump with Tidepool Loop.
Read more here:
Translation? This means that diabetics will be able to choose their own supported equipment and build their own supported FDA Approved Closed Loop Artificial Pancreases.
This is fantastic news for folks with diabetes; it's also really important that there now seems to be a way forward for community-developed open source software in developing SAMD (software as a medical device) in a way that the FDA will recognize. Validated / GxP software development is very challenging- figuring out a way to reach the same quality and testing levels required in this area will definitely take some active FDA participation in the open source world. It would be amazing if we could get this approach replicated for other conditions that would benefit from effective software in the loop.
Great news for diabetics, and great news for medicine. Hopefully great news for software developers in the trenches who want to make a difference, too.
posted by jenkinsEar at 2:59 PM on July 4, 2019 [1 favorite]
Great news for diabetics, and great news for medicine. Hopefully great news for software developers in the trenches who want to make a difference, too.
posted by jenkinsEar at 2:59 PM on July 4, 2019 [1 favorite]
From the article: "Open Source Artificial Pancreases will become the new standard of care for Diabetes in 2019."
Well, kinda. Medtronic is already trying to recall their old pumps which can be used with OpenAPS solutions, and has also sent letters to diabetes educators to discourage DIY looping. And not every endocrinologist is happy about it or willing to support it. So there has been some official pushback. A person can, of course, kill him/herself if the software isn't configured correctly. And doctors' offices that can barely handle prior authorizations for aspirin (which is a lot of them nowadays) will not be able to support DIY hardware and software.
And that is pretty much to be expected. For the end user, getting this stuff to work is not easy or cheap (it's already cost me $250). I'm a very savvy Apple user and it took me a few tries to get the software compiled and running on my Apple hardware. There are excellent support groups for the Loop software in Facebook, Zulip, and Github. But how many non-savvy diabetics even know about those things? It is turning into a pretty amazing grassroots effort for diabetics.
But just last week I began looping with an Omnipod, Dexcom G6, and the open source Loop software. It is the best control I've ever experienced in 42 years of T1D. The improvement in my sleep alone is yielding amazing results. The effort required to get this work (writing software, reverse-engineering the software of the insulin pods, designing hardware and software for the RileyLink and the mobile operating systems) is substantial, but the results are amazing.
Insulet, which makes the Omnipod insulin pump, is expecting to release their FDA-approved looping system (Horizon), for which they are partnering with Tidepool, in late 2020. But I ain't waiting that long.
The Loop software can make decisions every five minutes to regulate my blood sugar, whether I'm asleep or not. It's an incredible relief and an incredible improvement in treatment. But whether the FDA or established medical community sees it that way remains to be seen.
posted by the matching mole at 11:05 AM on July 5, 2019 [2 favorites]
Well, kinda. Medtronic is already trying to recall their old pumps which can be used with OpenAPS solutions, and has also sent letters to diabetes educators to discourage DIY looping. And not every endocrinologist is happy about it or willing to support it. So there has been some official pushback. A person can, of course, kill him/herself if the software isn't configured correctly. And doctors' offices that can barely handle prior authorizations for aspirin (which is a lot of them nowadays) will not be able to support DIY hardware and software.
And that is pretty much to be expected. For the end user, getting this stuff to work is not easy or cheap (it's already cost me $250). I'm a very savvy Apple user and it took me a few tries to get the software compiled and running on my Apple hardware. There are excellent support groups for the Loop software in Facebook, Zulip, and Github. But how many non-savvy diabetics even know about those things? It is turning into a pretty amazing grassroots effort for diabetics.
But just last week I began looping with an Omnipod, Dexcom G6, and the open source Loop software. It is the best control I've ever experienced in 42 years of T1D. The improvement in my sleep alone is yielding amazing results. The effort required to get this work (writing software, reverse-engineering the software of the insulin pods, designing hardware and software for the RileyLink and the mobile operating systems) is substantial, but the results are amazing.
Insulet, which makes the Omnipod insulin pump, is expecting to release their FDA-approved looping system (Horizon), for which they are partnering with Tidepool, in late 2020. But I ain't waiting that long.
The Loop software can make decisions every five minutes to regulate my blood sugar, whether I'm asleep or not. It's an incredible relief and an incredible improvement in treatment. But whether the FDA or established medical community sees it that way remains to be seen.
posted by the matching mole at 11:05 AM on July 5, 2019 [2 favorites]
Main obstacle for me is getting a pump funded. Don't get me wrong - I'm incredibly fortunate to not only get insulin for no charge in the UK but everything else too (including a CGM although that's not standard - and I use that with DIY-style add-ons and software too). But pumps aren't that common at all.
I assume for pumps that a bolus has pretty much the same activity profile as a normal injection, or is it faster?
I'd still be wary of closing the loop even if I had a pump mainly out of a fairly strong distrust of software to always work right.
posted by edd at 12:27 PM on July 5, 2019 [1 favorite]
I assume for pumps that a bolus has pretty much the same activity profile as a normal injection, or is it faster?
I'd still be wary of closing the loop even if I had a pump mainly out of a fairly strong distrust of software to always work right.
posted by edd at 12:27 PM on July 5, 2019 [1 favorite]
« Older What Just Happened Also Occurred Before The Last 7... | Caliph, Calafia, California Newer »
This thread has been archived and is closed to new comments
posted by The Underpants Monster at 2:26 PM on July 4, 2019 [2 favorites]