FDA Approves Wearable "Artificial Pancreas"
kkleiner writes "The FDA has approved a device that acts as an "artificial pancreas", which both continuously monitors a patient's glucose levels and injects appropriate amounts of insulin when needed. When blood-sugar levels become low, the device from Medtronics warns the wearer and will eventually shut down. The MiniMed 530G looks to offer an on-the-go solution for the growing number of people suffering from Type 1 diabetes who have to test their blood and inject insulin throughout the day. The company plans to improve the device to make a fully automated version down the road."
The current the pumps do not adjust the insulin levels. Instead the user and the doctor program it to pump different amounts at different times. The current sensors just log the blood sugars to the pump for easier downloading and analysis at the doctors office.
It is for type I diabetics, whose distinguishing characteristic is that they can't make more than trace amounts of insulin.
The integrated glucometer.
As has been covered before, airport full body scanners tend to kill medical devices like this. People have had devices like these, along with pacemakers and other equipment die after being subjected to high energy bursts of EMI; which is exactly what airport scanners do. While the goverment claims they're phasing these out, they are still in the field -- high power portable x-ray and 'mwave' scanners that are being used at customs checkpoints, or on unsuspecting civilians on the road. And then there's those pesky aircraft carriers that carry gigawatt radar scanners that on several occasions have locked people in their cars, garages, etc., due to EMI when they were passing by.
All of this kind of unregulated and largely unmonitored technology poses a very real danger to technology like this; And with so many people having diabetes, this could mean that entire towns' worth of diabetics drop dead while the government claims "it's a mystery why everyone with implantable medical devices died after we irradiated them..."
My point is; The laboratory environment these things were designed (and approved) in is very different from the environment they're going to be used in. And there's no evidence the FDA has taken this into consideration from what's provided here. Indeed, they have a poor track record of having an impartial approval process; I do not believe that 'FDA Approved' means much more than 'Scientology Approved' these days -- but this is to be expected when the FDA's income is derived directly from the companies' whose products they approve -- companies literally pay for approval. Anywhere else, this would be a clear conflict of interest. But when it comes to the safety of our food, drugs, and medical supplies... it's business as usual.
#fuckbeta #iamslashdot #dicemustdie
Currently, I'm using a Dexcom continuous glucose monitor, and an Omnipod insulin pump. The advantage of being able to automatically turn down one's basal rate is an advantage, yes. I do this manually for myself, based on my Dexcom readings. But it isn't all that your pancreas does. If your blood sugar is diving too quickly, you have to supplement with sugar orally to make up for the fact that your pancreas isn't secreting hormones to make your liver release stored glycogen, or you may go too low and pass out. Often if I engage in unexpected exercise (moving boxes, changing a tire, spontaneous run) shortly after bolusing for a meal, my sugar will crash because my body becomes more responsive to the insulin I've taken, and once I've taken it, I can't un-take it. Kills spontaneity.
Your pancreas also supplies you with insulin automatically based upon your blood sugar fluctuations... this product doesn't. You have to manually calculate your mealtime boluses and make the pump give it too you.
This bionic pancreas is the technology I'm excited about, and can't wait for it to come to market. It automatically calculates and releases both insulin AND glucagon in measured amounts to keep your blood sugar levels as close to normal as I've ever seen.
Having worked at Medtronic, but not at the Minimed division, I would guess that it doesn't really have an OS. The HMS Plus and Magellan devices didn't contain a RealTime OS or anything similar and the Magellan was originally programmed by a pacemaker engineer before I got on the project, so they use C to make embedded software, but there's no underlying OS like VxWorks, etc.
My brother has a normal insulin pump. They work by continuously pumping in "fast acting" insulin into you (the basal rate), if you eat a meal, you have to calculate by hand the amount of extra insulin needed and press buttons on the unit to deliver the required amount. And yes, it knows that at different times of the day, you need more insulin than others. This is totally separate from the slow acting insulin that type 2 users sometimes take an injection of once or twice a day. If for whatever reason, the insulin delivery doesn't work properly, he'll start to have problems quickly, under a couple hours.
His also has a blood meter which starts beeping if his insulin level falls below a certain level. What his pump doesn't do is automatically change the amount of insulin delivered on the fly. Any change in insulin delivery has to be programmed. If he eats an apple, he has to press buttons to dose himself. If his body chemistry changes and that basal rate needs to be adjusted, it has to be programmed. My understanding from him is that the blood glucose measurement isn't especially accurate, though I can't remember why.
This is just the next generation of those same components. The generation after this, expect to see a unit that does a lot more dosing automatically. I think the technology is there, we just need to clear the regulatory hurdles.
Typically insulin pumps deliver insulin in two modes: Basal and Bolus. While a bolus is a large injection provided as quickly as reasonable, the Basal is a rate of delivery which can be instructed to vary over the course of the day. I would imagine that the device described in the article likely organizes injection in this fashion, with the added feature that if your blood glucose spikes, it will react to that automatically.
I had an insulin pump for a number of years (from the same manufacturer that made the device in this article, in fact), so I am familiar with the usage. I, personally, had problems using it (I sweat too much for the catheter to stay in reliably), but I think that they're a great technology for those who can use them. This growing automation is certainly a good thing.
Learning about brewing beer, by brewing beer.
We've had insulin pumps for decades, and continuous glucose monitoring systems for many years. This is just a small iteration on top of that. The new thing? If the CGMS thinks your blood glucose is low, the pump is instructed to stop giving insulin.
This ain't an artificial pancreas by any measure - even the manufacturer says as much.
When MT says they plan an 'improved device', they mean it's ready now - just waiting for certification . . .
Alternatively, they've already identified potential design limitations but as the certification process takes longer than the development of a new model, they have opted to complete the certification process and begin getting a return on investment, while pursuing parallel development of a replacement model.
#fuckbeta #iamslashdot #dicemustdie
Also if the probe is the same as the one they currently sell, its crazy expensive, has to be replaced every 3 days and has an enormous needle. Too bad they don't mention any of that in the article.
Your problem is that you're looking to television to inform you. The purpose of tv is to entertain and sell. That's it.
It's probably cheaper than any other method of managing their condition like amputation, disability payments, and nursing homes.
Keeping diabetes from going from the "cheap to manage" to "terribly expensive" stage is probably, like most other healthcare things, a net savings once you get to even the medium term.
> I'm sure the scientists involved - who are much smarter than you, of that I have no doubt - thought of that. Dipshit.
Why aren't you on 4chan co-surfing with the Breaking Bad finale?
(-1: Post disagrees with my already-settled worldview) is not a valid mod option.
It's hard to make a closed-loop control because most continuous blood glucose monitors don't measure the blood, but the residual glucose in the intersitial fluid, and this lags blood glucose by several minutes, which can be a big deal, depending on the food type.
Blood sugar doesn't have a linear-time-invariant response to food input, different macronutrients can create different contours in blood sugar level over spans of time. Generally, a pump can't guess how many units to move in the bolus unless it knows specifically what you ate, it's not just a matter of dose, it's a matter of how long -- different foods require a more time-release bolus, sugars require a spike, all-at-once bolus.
Don't blame me, I voted for Baltar.
And go figure, you're posting this drivel on an article about a device specifically targeted at Type I (autoimmune) diabetics... A device which, among other benefits, will help a Type I diabetic manage weight better by enabling them to maintain tight bloodsugar control without excessive hypoglycemia incidents. (Common problem for a Type I - eat a meal, take insulin, and then exercise without properly reducing your mealtime insulin dose to compensate. End result - hypoglycemia which requires you to eat more. A CGMS makes it a lot easier to properly plan insulin deliveries without overcorrections/oscillations.)
retrorocket.o not found, launch anyway?