'Artificial Pancreas'

[Jo_the_boat]

Interesting.....
http://www.bbc.co.uk/programmes/articles/1KYpXw7078xsp5Lb8518Znt/artificial-pancreas

 

[abcd99]

Thank you for this interesting information. It really gives hope. My son 26 years old now is type 1 diabetic since last 3 months.The pain and shock that I have got as a mother is unexplainable. I am praying all the time for these wonderful people including little innocent children. May God do some miracle and may some remedy be found to get rid of this condition called diabetes.
Do you have any idea till when this will be available in market and how much it will be the cost.
Good day.

 

[SHmano]

Interesting.....
http://www.bbc.co.uk/programmes/articles/1KYpXw7078xsp5Lb8518Znt/artificial-pancreas

Wow a good news. How long will it take to come to India. What will be cost?

 

[Deleted Account]

Thank you for this interesting information. It really gives hope. My son 26 years old now is type 1 diabetic since last 3 months.The pain and shock that I have got as a mother is unexplainable. I am praying all the time for these wonderful people including little innocent children. May God do some miracle and may some remedy be found to get rid of this condition called diabetes.
Do you have any idea till when this will be available in market and how much it will be the cost.
Good day.

Whilst this is great news, don't forget the amazing people who live with type 1 diabetes: it is not a death sentence and should not restrict your son's life.
Just look at athletes (Steve Redgrave), actors (Halle Berry), politicians (do I need to type her name?), ...

I am not in their league but diabetes has not stopped me travelling including tracking gorillas in Uganda and walking up waterfalls in Venezuela, sailing, climbing, ski diving, working, dating, ... if anything it makes me more determined.

 

[first14808]

Hmm..

The system consists chiefly of pre-existing off-the shelf components: a blood glucose sensor and an insulin pump, which is controlled remotely by a mobile phone. The breakthrough has been to get them working together to monitor glucose levels closely, and respond with an accurate dose of insulin with the help of a specially-designed phone app.

That may make approvals fun. Hopefully they're just using the phone as a compute device, not relying on the 'cloud' and mobile networks being available. Otherwise it would be problematic if you wanted to be adventurous, or just happened to be in one of the many places where there's no signal. As it automagically adjusts insulin delivery every 10mins, there's potential for harm if the machine learning gets it wrong, or the network's down.

To me, this sounds like a hybrid of Libre/CGM, pump and trust in 3rd parties rather than a true artificial pancreas. It may offer benefits to people who struggle with managing on their own, but personally I'd want to be in control.

PS..

http://paediatrics.medschl.cam.ac.uk/about-us/people/senior-academic-staff/dr-roman-hovorka-2/

Components of Florence D2A automated closed-loop system comprising an Android phone (Galaxy S4, Samsung, South Korea) running control algorithm and communicating wirelessly with Dana Diabecare insulin pump (Sooil, Seoul, South Korea) and translator (Triteq, Hungerford, UK) with inserted Nav2 Receiver (Abbott Diabetes Care, Alameda CA, USA)

So hopefully it doesn't rely on the network, and being paediatrics, may be aimed at children who can't manage their own dosing.

 

[DCUKMod]

As I understand frorm our very own, geek-tastic @tim2000s - This is, in essence, already our there, in the form of closed loop system. His blog has a number of entries talking about it.

 

[abcd99]

Whilst this is great news, don't forget the amazing people who live with type 1 diabetes: it is not a death sentence and should not restrict your son's life.
Just look at athletes (Steve Redgrave), actors (Halle Berry), politicians (do I need to type her name?), ...

I am not in their league but diabetes has not stopped me travelling including tracking gorillas in Uganda and walking up waterfalls in Venezuela, sailing, climbing, ski diving, working, dating, ... if anything it makes me more determined.

Thank you so much dear Helen......you don't know how happy you have made me. I feel much lighter.........God bless you.

 

[first14808]

As I understand frorm our very own, geek-tastic @tim2000s - This is, in essence, already our there, in the form of closed loop system. His blog has a number of entries talking about it.

Neat. One thing I thought from looking at the pic on Dr Hovorka's page is the <cough>beta</cough> pancreas is a bit cumbersome, especially if intended for small people. But from what they and tim2000's doing, looks like there's scope for integration. Especially if there's co-operation from vendors, but that's probably where regulatory hurdles would come in.

Looking further ahead, it might be possible to use a shunt and combine spectroscopic or diffraction based glucose sensing with insulin delivery. That would get around some of the challenges with enzyme base CGMs and testing and reduce the need (and cost) for consumables in the sensing side.

 

[DCUKMod]

Neat. One thing I thought from looking at the pic on Dr Hovorka's page is the <cough>beta</cough> pancreas is a bit cumbersome, especially if intended for small people. But from what they and tim2000's doing, looks like there's scope for integration. Especially if there's co-operation from vendors, but that's probably where regulatory hurdles would come in.

Looking further ahead, it might be possible to use a shunt and combine spectroscopic or diffraction based glucose sensing with insulin delivery. That would get around some of the challenges with enzyme base CGMs and testing and reduce the need (and cost) for consumables in the sensing side.

Well, I think if we consider how relatively recently mobile phones were the size of half a case of beer and weighed almost as much to when they were at their smallest (now, many moons ago, as they're getting bigger again as they get smarter), it feels like these things will develop rapibly, although whilst individual components are being utilised, as opposed to functionality integrated into each other any miniaturisation will likely be limited.

That said; we live in exciting times. There has probably never been a better time to be diabetic.

 

[NewTD2]

This is also interesting-

https://www.diabetes.co.uk/artificial-pancreas.html

 

[ringi]

The setup the BBC showed on TV needed the number of grams of carbs for the meal to be entered into the phone application before the meal was started. But it did seem to work out the correction factors etc itself. I don't know what it did about exercise.

When faster insulins (with a much shorter half-half) come onto the market, I expect we will see a lot of progress with "artificial pancreas" systems.

(But I keep thinking of the risk of DKA if the system goes wrong, and therefore I question why people are not using a small dose of long half-life insulin in a daily pen so as to protect against DKA if the system stops working.)

 

[first14808]

That said; we live in exciting times. There has probably never been a better time to be diabetic.

Indeed. One of the things that put me off medical electronics as a career was the challenge of getting enough power and compute capability into a small, easy to use package. Now, that's a lot easier. Our first year project was to design a remote controlled pacemaker.. Or rather work with a company that had launched one as a way to see how many ways we could potentially kill our patients. Or just manage 'simple' things like the body's remarkable ability to reject, or otherwise burger up things we wanted to implant... Which I guess is the current challenge with sensing. I found this:-

By way of Tim's blog, and kept expecting it to say 'And now with your third hand..'. Current CGM sensors seem a bit.. fiddly, so not exactly suitable for some use cases, ie anyone with limited dexterity.

 

[TheBigNewt]

Indeed. One of the things that put me off medical electronics as a career was the challenge of getting enough power and compute capability into a small, easy to use package. Now, that's a lot easier. Our first year project was to design a remote controlled pacemaker.. Or rather work with a company that had launched one as a way to see how many ways we could potentially kill our patients. Or just manage 'simple' things like the body's remarkable ability to reject, or otherwise burger up things we wanted to implant...
.

I'm trying to think why anyone would benefit from a "remote controlled pacemaker". Having implanted a few hundred Medtronic pacers and defibrillators I'm not sure I'd want to remotely reprogram one. Yes, it's been possible for years to remotely interrogate them using a home transmitter (plugged into a land line usually). And yes a defibrillator that goes "boom" in the middle of the night will ring up a monitoring service in another state who will subsequently send an ambulance to the guy's house who got shocked in his sleep. But remote programming??? FWIW I think the "artificial pancreas" has been dangled in front of Type 1 diabetics so long we're getting tired of it to be honest. Yeah, I know, it can kinda sorta be done. Sorta. With a quasi-gerryrigged system that you gotta watch like a hawk. What we want is like a knee replacement. New pancreas without the rehab. Right? Fill 'er up with insulin and head to ColdStone!

 

[first14808]

I'm trying to think why anyone would benefit from a "remote controlled pacemaker". Having implanted a few hundred Medtronic pacers and defibrillators I'm not sure I'd want to remotely reprogram one.

I guess it depends what you mean by 'reprogramming'. Which in our case, was understanding and making sure there were enough checks and safety features that it only did what it was supposed to. The company we were working with was Sorin, and in the early '90s, I think most of their competitors only paced a set rate. And from a quick look at wiki, think Sorin's was a 'VVIR' mode so users could dial up their heart rate if they wanted to be more active. Operators (ok, cardiologists) would have their own config mode & the two had to be kept very seperate. It was quite fascinating going through all the physical and technical challenges though. Oh, and ethics. When we touched on marketing, we thought of a secondary use, ie being able to crank it up to 140bpm and get a cardio workout while watching tele on the sofa. That would be unlikely to gain approval for some reason..

 

[TheBigNewt]

Yeah rate responsive pacing was early 90's. Then the next big thing was "mode switching", to keep the thing from pacing the ventricle at the upper rate limit if the dude went into a.fib. Then biventricular pacing (3 leads).

 

[donnellysdogs]

A consultant offered me a trial asap of artificial pancreas... after getting to see my consuktant that knows me... my consultant agreed with me that the algorythyms in the AP could not cope with my problems with my body rejecting cannulas...

Please dont think that this gadget consists of "off the shelf components" either.. if they were off the shelf, the nhs would be queuing up and handing them out to everyone. The standard pump alone before tubes, cannulas and insulin is what? £3,000. All it is basically getting is different algorithms... thats it.. same cannulas and tubes and sensors...

It is not an "off the shelf product" it will be deemed by NICE or NHS as only available for the really needy...... probably under exceptional cases.

 

[first14808]

Yeah rate responsive pacing was early 90's. Then the next big thing was "mode switching", to keep the thing from pacing the ventricle at the upper rate limit if the dude went into a.fib. Then biventricular pacing (3 leads).

At the risk of sounding old and flatulent, kids these days.. That was something we wanted to do, but my development box then was a state of the art 486DX50 (that's 50Mhz CPU) with I think 4MB of RAM. And another large box with some DSP and A/D boards. Now, I have 4.2Ghz CPU with 6 cores, 32GB RAM, and another 24GB on my graphics cards. A cheap 'smart' phone comfortably outperforms what I had then.

And challenge was fitting everything into a neat little titanium case that could be worn comfortably by the patient, and a practical battery life. If we'd evolved to be digital rather than analogue, it would make engineer's lives so much easier.

I suspect there are similar challenges with a sim-pancreas. From looking into CGMs, some don't look that continuous. Think one I saw said it stored 40 readings per day, so average once every 36mins. Nyquist might say that's not often enough, especially with fast acting insulin. That I suspect is something Dr Hovorka's algorithms and machine learning deals with.

But for the youngsters.. the man/machine interfacing's an interesting career to get into, and a lot more fun than C# or SQL coding for corporates. Plus if you can crack in vivo glucose measuring, there's the potential of an implantable solution. I think there are other systems that allow injecting into implanted implants for medication top-ups, so that part's sorted. Give or take patents..

 

[mentat]

OpenAPS and Loop already do this and anyone can use them (if they are prepared to take the risk and do the mountain of learning necessary). I used OpenAPS myself for a few months and it was a big relief overnight.

I'm not using it anymore because I finally figured out why my diabetes has been all over the place for 7 years, and treated the underlying problem, so MDI is sufficient now.

 

[abcd99]

As I understand frorm our very own, geek-tastic @tim2000s - This is, in essence, already our there, in the form of closed loop system. His blog has a number of entries talking about it.

OpenAPS and Loop already do this and anyone can use them (if they are prepared to take the risk and do the mountain of learning necessary). I used OpenAPS myself for a few months and it was a big relief overnight.

I'm not using it anymore because I finally figured out why my diabetes has been all over the place for 7 years, and treated the underlying problem, so MDI is sufficient now.

Please can you tell me what do you mean by MDI......as I told in my first post that this subject is complete new for me. It was only 3 months before that my 26 years old son was found having Type 1 diabetes.

 

[mentat]

MDI = multiple daily injections, nothing special. By the way, I was diagnosed at 22 and I can understand what a shock it must be to you.