Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

As a t1d who's been using a cgm for like 8 years at this point, having it there constantly is objectively better, and you normally don't notice the needle (needs a bit of fat though). Fairly routinely, I'll go a couple of days without because the installation is a bit of a pain, and my blood sugar is invariably worse on those days. A big part of it is the reduction in cognitive load devoted to managing your blood glucose. You don't have to estimate what it's doing, just decide on the right action to take. It can wake you up when you're going low, which is much better then waking up in a cold sweat and devouring your kitchen refrigerator from the hunger.

Also, the alternative is pricking your finger so it bleeds, which is much less pleasant.

One thing op doesn't talk about is close loop systems, which automate some of the insulin delivery using the sensors. They're still in the very early days, but they're generally good at dealing with random variation or overnight highs.



The article does mention closed loop systems.

> That's what Beta Bionics, amongst others, are focusing on. Their iLet® system literally bills itself as "a fully automated bionic pancreas".


It's crazy that we've kind of been capable of this for almost a decade now. You can buy all the parts to make an artificial pancreas, and use open source software from GitHub to get them to talk to each other, but none of the biotech giants want to put the parts in a box and sell it as an artificial pancreas because the liability would be insane. One little firmware glitch could literally kill a ton of kids in their sleep.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: