FDA Approves Insulin Pump You Can Wear | Hacker News Reader

FDA Approves Insulin Pump You Can Wear

singularityhub.com

39 pointspwg41 comments

It should be noted that this appears to be a highly accurate insulin pump. That's a far cry from an artificial pancreas. If you have the death sentence known as "pancreatic cancer" you can't get your pancreas removed and have this take its place for example. Not saying it isn't important for diabetics, but let's not get carried away.

"appears to be a highly accurate insulin pump"

The article was a little vague on how well the control loop is closed, but it provided the impression that there was a feedback loop, as though a real engineer told the journalist there was a feedback loop and the article is the result of the liberal arts interpretive dance of the artists impersonation of the concept of a feedback loop trying to play to the masses who don't know what "feedback" is or Bode stability plots or whatever.

The creation of a feedback loop would make it much more of an "artificial pancreas" than a mere free-running pump or traditional injection.

I think JunkDNA's point was that the pancreas does a lot more than just regulate insulin.

What else does it do? (if you know... I'm curious)

Besides that, not anything significant (as far as we know).

http://answers.webmd.com/answers/1195061/can-you-live-withou...

> The pancreas is also the main source of enzymes for digesting fats (lipids) and proteins. http://en.wikipedia.org/wiki/Exocrine_pancreas

In fact, it's just a pump that shuts off if you don't acknowledge its alarms that you're having a low. It's also the next gen of their (known to be poor) CGM.

Source: Diabetic wearing their current gen.

OK.. What is going on here:

This is an incremental improvement in Medtronic's continuous glucose monitoring system/insulin pump combination therapy that has been available for 5+ years already.

Medtronic is touting this as 'artificial pancreas technology' (not specifically an artificial pancreas) because it has a feature now in the US where the pump can turn off if your blood sugar drops below a certain patient set threshold. This has been available for years already outside the US.

This is most definitely not an artificial pancreas..

http://www.hanselman.com/blog/ItsWAYTooEarlyToCallThisInsuli...

It's WAY WAY too early for them to have said this.

I asked their twitter about it and got this: https://twitter.com/thedamon/status/384420935685910529

Still think 'artificial pancreas technology' is an overstatement. Though Singularity hub is also guilty of writing a ridiculous article esp with that headline.

Edit: great write-up, Scott!

This is nonsense. They are totally overstating what they've done.

NOTE: I've been diabetic for 20 years and I wear both a CGM and a Medtronic Pump (the model before this new one).

I'm so mad about this I've written it up IN DETAIL: http://www.hanselman.com/blog/ItsWAYTooEarlyToCallThisInsuli...

Ugh.. Sensationalist headline of the day.

Article headline: we have an artificial pancreas

Article body: we do not have an artificial pancreas. we have a new pump with some incremental feature improvements over our last model.

In the 90s I invested in MiniMed, and sales exploded. In the early 2000s they were acquired by Medtronic. This was the goal of MiniMed and I'm glad to see it come to fruition.

My spontaneous thought when I saw this was "what took them so long?" Your comment makes me wonder even more. How come this is such a difficult problem?

I was wondering the same thing - a simple feedback loop can't be that hard to achieve can it?

Put in some limits in case of sensor or pump error and maybe allow the user to boost sugar for activities, but otherwise this seems to me like a very very simple thing to do.

What am I missing?

There have been several efforts at developing a blood glucose sensor that could work on the skin rather than with an invasive probe. None of those went anywhere.

I assume this means that this is a very difficult task. Believe me as a Type II Diabetic that sticks their finger several times per day something that could detect your blood glucose without blood would be worth a LOT.

I can't imagine what it would be worth to someone with Type I.

It's complex because of LAG. Even a finger stick shows you your blood sugar in the past. When you take insulin it ALSO has a lag, as long as 30 to 90 minutes.

Managing Type 1 diabetes is like controlling the Mars Rover.

Source: Type 1 for 20 years on a pump and CGM

Details: http://www.hanselman.com/blog/ItsWAYTooEarlyToCallThisInsuli...

it probably it isn't , the problem is that the FDA is strict and it takes a lot of time even before starting human trials and even then it might take 5+ years depending on what is being tested.

I've also heard of some drugs that take 30+ years to test for long term side effects ( that's assuming the medical company has the finances to support it for that long).

The blood sugar sensors have been the problem.

My wife is a Type I diabetic and 15 + year Minimed pump wearer. She (and many other diabetics) don't like the fact that with the CGM you have to have two infusion sites (one for the monitor and one for the pump). Until they can figure out how to do this with one infusion site I suspect it will not be highly desirable by diabetics.

Ditto for my wife. She also does not like the 45 degree "bent" insertion approach because "it hurts more".

In addition to all the accurate assessments that this is very far from an artificial pancreas, I would like to point out that the glucose monitor being marketed by Minimed is a far cry behind even the best-in-class available from Dexcom (http://www.dexcom.com/dexcom-g4-platinum), and not near accurate enough to allow for closing the loop. Notably, a number of the farthest advanced artificial pancreas research projects are moving away from Minimed products-- for example: http://bionicpancreas.org

And you might mention that you also write on this pretty frequently :)

"Karmel Allison - Where is My Robot Pancreas?"

http://asweetlife.org/author/karmel/ http://asweetlife.org/karmel/blogs/science/ada-2013-live-con... http://asweetlife.org/karmel/blogs/type-1-blogs/updates-from... http://asweetlife.org/karmel/blogs/personal-blogs/in-chicago...

I moved off of the Medtronic for the Dexcom G4 for this very reason. The difference between 30 minute lag and 5 minute lag is SO noticable. My A1c is now 5.5 and I attribute that to the Dexcom, 100%.

I must say I recently was testing the new Enlite medtronic sensors and found them far more accurate than the older ones whatever they were called (I gave up on those quickly). Can't compare directly to dexcom but that's good to know!

Considering there are places where they will steal your purse which might be worth $3k. Imagine what the theft rate will be for artificial Organs that could be grabbed right off you.

No need to knock a person out and leave them in a hotel bathtub with their kidney's removed, you can just catch them on the subway.

Its a pity you got downvoted; discussion of the social issues relating to technology is one good definition of hard science fiction, which the world certainly has a lack of. Soft sci fi is that boring "search and replace" dreck applied to existing generes.

Another societal issue is militarized cops and the like beating people and ripping away their electronics, which we've been conditioned to feel is "normal", but its kinda queasy/gross to imagine cops physically ripping needles and stuff out of peoples bodies when they give a good ole american beatdown for jaywalking or driving while black or other similar criminal activities.

This is good news for diabetics, although I find it pretty depressing that something like a quarter of the US population needs something like this in the first place.

Before making sweeping statements about a disease you obviously don't understand please read up on the difference between type 1 and 2 diabetes, and what the causes and treatments are for both.

TL:DR - This pump would be used for type 1 diabetics. This type means your body does not produce insulin, so the treatment is to artificially inject it (via this device for example). It's not caused nor limiting to diet. Being unhealthy/overweight doesn't increase your chances of getting type 1. It's a genetic, juvinile onset disease.

Type II is the type you're most likely thinking of. The body becomes less sensitive to insulin, so you need to consume less carbs and generally eat healthier. This pump would not be used by type II diabetics.

Off topic: this site is painful to use on mobile with Javascript off.

Also, the pictures don't load if I turn images off.

jrockway, you've really changed. You weren't always this cynical. What happened?

Pretty sure I've always been this cynical.

Why can't we just fix one letter in the DNA and make diabetes not happen? Diabetes is stupid. All internal organ illnesses are stupid.

Still that's the best we can do atm. Still all those anti-GM people actively hurt our chances to get real treatments soon.

> Why can't we just fix one letter in the DNA and make diabetes not happen?

I'm sure that it's because of the big pharma conspiracy which tries to keep that technology secret so that they can make more money.

Because that's not how diabetes works

Wikipedia tells us that diabetes type I is usually genetic.

If you're susceptible, you might get it, might not; if you aren't, you aren't. Guess what, no point of carrying bad genes around.

The genetics are complex and not yet understood.

(I am a diabetic who does genetics)

Are you sure it's possible to understand them fully without actually doing some hacking? How would you be sure when you understand enough to start acting?

They do "hacking" etc in mouse models but it is expensive and time consuming. You might have about 700,000,000 genomic variants to choose from so we need to do other studies to try things.

For the time being one of the most effective means of studying a disease is to sequence a pedigree of individuals some of whom have the disease and analyze the inheritance patterns until you narrow it down to the point you can be sure.

You can also do studies where you just sequence a ton of people with and without the disease but those can be more problematic statistically because of the number of possible variants.

There are known workarounds that make diabetes-1 not fatal; "doing some hacking" (whatever you'd mean by that) would likely have a significant percentage risk of being fatal if you want do do anything that does something to your body at all without spending, say, less than a year.

"Above all, do no harm" makes for a very conservative approach of being really sure before risking actual action.

Reply on news.ycombinator.com