Comment by ijustlovemath | Hacker News Reader

Comment by ijustlovemath | Hacker News Reader

Parent Full thread

we're building an artificial pancreas for hospitals, so I know a good bit about CGMs. Noninvasive blood sugar sensing is horrifically difficult. Every few years, people come along and say "oh this is just some simple DSP on spectroscopic information, piece of cake" before inevitably running up against:

- skin conductivity changes over time

- the ways in which skin tone changes signal absorption (which itself changes over time)

- the ways in which different levels of fitness affect blood flow, material density etc.

You also can't use it in a hospital setting, due to how your skin and bloodflow changes during serious conditions like sepsis (though I'm guessing they're not thinking about that market).

Really smart people have been trying to use Raman spectroscopy to solve this problem for decades at this point (early patents go to early 2000s). Apple is an extremely strong hardware vendor, and I wish them luck, but I would not hold my breath for this. Plus, I'm guessing they will not open the signal up for looping, which would really leave the T*DM community out to dry.

Honestly, none of those sound like blockers for the use case I and many other diabetics would like - monitoring for general blood sugar responses (rough curve) after eating. Sure, you wouldn't be able to use the measurements to dose insulin or even measure your actual (numeric) glucose level, but measuring my A1C every three months is good enough to do that in mine and many other cases. I've had my blood sugar controlled through diet and metformin with it being in the range of 5.9 - 6.2. I could do so much better if I had a better understanding of how my body, specifically, reacts to certain foods, mealtimes, routines (exercise after eating), etc.

It would be super helpful to know (relative to other foods) how my body reacts to claimed low-carb foods. Is there a large spike (don't need to know the number) or is it a much more flat curve? How long in general does it take for the line to return to pre-meal levels? What does that trend look like over many months? Heck, I could even run a rudimentary and simple test to do comparative insulin response to a known amount of carbs to see if my insulin response is improving over time (using the period of the curve). I would love to get an alert that hey, we think your glucose level shot up a lot (don't care how much) so that I can remediate it through exercise then and there and avoid that food or timing going forward.

Really hoping the people in Medtech don't make perfect the enemy of good in this case. Although maybe what you listed would still be blockers for even getting general glucose curves. I've been planning on getting a CGM for at least a few months to achieve all of this, but it would be great to just have it in a watch or other simple wearable.

Agreed that the value of a CGM is in the change information, and that adding a CGM is probably the biggest quality of life increase for anyone with diabetes. Highly pro CGM if you can get it!

The issue with spectroscopic approaches is the amount of noise can be really hard to disentangle, to the point that you might get really unreliable trend information, where it might even be dangerous if you're making dosing decisions off it. And even if you aren't, getting incorrect trend information doesn't really help you any more than just not knowing it.

It sounds like you're a T2 diabetic, in which you're not taking insulin and high blood sugar over long periods of time is your risk profile.

Unfortunately T1 diabetics and insulin dependant T2 diabetics will get killed by a machine that tosses out numbers that can be considerably off to the point no manufacturer is going to make the risky product.

the whole point was the tech is not accurate at the moment, you're better off with a random number generator on your watch

Reply on news.ycombinator.com