FreeStyle Libre 2 blood glucose sensor

As yet, according to medina, the causes of CT1 are not known.
Genetically, it is possible to have one gene that increases the risk of contracting the disease (my child does not have this gene and has contracted the disease).
From information from medical staff, the incidence among children has increased significantly after 2020, which is Covid's time.
CT1 is an autoimmune disease, like so hasimoto, psoriasis.

I don't know how it works in Poland, but in some countries the recycling consists of sending the used ones to the manufacturer or distributor free of charge.
.

clubber84 wrote:

I once wrote in a topic about the functions of smartwatches that the blood sugar measuring function included in them is a fraud a mile away,

.
Unfortunately the layman on the subject of blood sugar measurement is you.
Please read an overview of the methods of measuring sugar "through the skin" https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7085605/

Of course "through the skin" methods have a catalogue of difficulties to overcome and measuring directly from the blood is simpler. But the risk of infection, scarring etc is not always worth the precision.

CosteC wrote:

clubber84 wrote:

I once wrote in a topic about the functions of smartwatches that the blood sugar measurement function included in them is a fraud a mile away,

.
Unfortunately the layman on the subject of blood sugar measurement is you.
Please read an overview of methods for measuring sugar "through the skin" https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7085605/

Of course "through the skin" methods have a catalogue of difficulties to overcome and measuring directly from the blood is simpler. But the risk of infection, scarring etc is not always worth the precision.

Research is of course being done, as are trials with high-temperature superconductors, cold nuclear fusion and time travel. But the fact that scientists believe they have moved some elementary particle back in time by a few picoseconds has no bearing on the consumer market for time machines.

However, if you link any reliable article about commercially available devices that actually work measuring glucose levels through the skin, I'd love to read. My father would be happy to wear some sort of bracelet instead of an implanted sensor.

zgierzman wrote:

you have linked some reliable article about the devices available on the market that actually work measuring glucose levels through the skin

.
You've been given a very reliable, cross-cutting article about advances in this field. If you can't get enough please: https://www.nature.com/articles/s42255-024-01016-9 Source: Nature. Not some Polish pop-up promoted by Chernk.
E.g. Stanford is working on something even more interesting https://cheme.stanford.edu/smart-contact-lens-monitors-blood-sugar

Commercially available solutions also exist, e.g.: https://afontechnology.com/glucowear/ or https://www.medicalexpo.com/prod/mediwise/product-105186-686902.html although I personally cannot assess the accuracy.
They are certainly inferior to invasive solutions, because they have to. But there are already some on the market.

I think it's my first post on the electrode after 16 years of registration O.o

It just so happens that I have been a long-time user of this system and I have completely mixed feelings, especially when it comes to conditions completely unrelated to technical ones. As I have a lot to do with electronics on a daily basis, the biggest magic for me is precisely the enzyme electrode, which is the only part of the device that is 'connected' to the body. The rest seems to be relatively graspable electronics: a precision ADC, probably temperature-compensated (the plastic housing is even thinner in the area of the primer NTCa), measurements stored inside the chip, which we can read via NFC and, more recently, via BLE.

The most interesting thing is precisely this BLE. For the past year or so, with the original manufacturer's app on my phone, I have been able to look at the screen and find out my sugar levels. Up to this point, the way BLE worked was completely incomprehensible to me: the BLE datagrams contain all the information (encrypted, the keys can be obtained by scanning NFC) about the measurements of the last few minutes, but were only used to display a push notification stating that the sugar level was too high or too low. To find out the numerical value of the measurement I had to perform an NFC scan....

There were a couple of guys who wrote unofficial apps that ran all communication over BLE and this was probably the motivator for the manufacturer to make such functionality, after X amount of time, appear in the official app. I suspect that otherwise, such functionality would only have appeared in the next generation of the device, which would have been correspondingly more expensive, well, because it would have had new functionality.
As far as I'm concerned, this is definitely a good thing: the built-in controller from Texas has memory for 8 hours of measurements, so in order for the doctor to issue the next reimbursement order, I have to have a certain % of time covered by measurements in the official app in the previous period. You slept more than 8 hours: what a bummer, another % of time covered by measurements fell out, because after all, it was impossible to enlist after BLE.

tmf wrote:

Then produce and earn...

.
I realise the logic of this sentence: R&D costs money, customer service costs money, certification costs a lot and a thousand other things cost money too, but this price seems to be mainly due to a rather heavily regulated market: you need a lot more resources and documents than a simple CE to be able to bring this to market. It's a pity that there isn't much of a legal possibility for me to buy such a device on my own responsibility without all those receipts. I realise that quite a few people wanting to save money would do so, and then the courts would be flooded with a wave of lawsuits claiming that the sensor cheated me and killed me. So bad and so bad.

Anyway, it seems to me that the manufacturer has a lot of margin: in my case, in a large proportion of cases, the sensor is not able to work for the claimed 14 days. Most often it "breaks down" when:
- I have a fairly low sugar level for an extended period of time, and then I eat something sweet: the level starts (not in every case) to rise rapidly (i.e. probably some voltage drops or rises rapidly).
- I have very low sugar levels for a long time (so probably some tension is permanently low or high).

By making a complaint I will almost 100% end up sending another sensor, which I can use new 14 days with the requirement to return the previous "faulty" one. I don't think I have ever encountered such a complaints policy.
At least since I use the original sensor "reader", which is a small device with NFC with which I can scan the sensor and see the result on the screen. When I used my phone, complaints were rejected because, as is generally known, the BLE receiver certainly messed up the sensor. In official language, however, this is using an incompatible smartphone. The manufacturer has a list of "supported devices" that includes most iPhones, quite a few Samsungs and a few other models. My phone is outside this list. The previous phone was on this list and, interestingly, 8/10 attempts to scan NFC ended in a connection error, but did not result in a rejected claim. The new one is off the list, no connectivity issues, but there are issues with complaints. Either way, the average 'spoilage' frequency, for each of the three devices (official reader, supported phone and unsupported phone), is unlikely to have changed significantly.

Marcin125 wrote:

This is what I did not understand....
The choice is simple: either a human life is saved or the whole world is saved.

.
Well, exactly both.

For reasons I don't know, in some countries, the manufacturer is obliged to collect (I'm not sure if they recycle, but it seems to me that with the right scale of collection, it could be worthwhile) used devices, but not in Poland. As a curiosity: there are not even really precise guidelines on how one should deal with a used sensor, as it is a composite of medical waste and used electronics. Once under a Facebook post:

Abbott FreeStyle wrote:

The used FreeStyle Libre 2 sensor should not be disposed of with municipal waste. Separate collection of used electrical and electronic equipment is required in EU countries.
⚠️ Note: Sensors contain batteries that cannot be removed, so do not incinerate them.
Please remember to dispose of them properly! 🌿🔧🌍

<br/span>
Someone asked how to dispose in such a case and got in response:

Abbott FreeStyle wrote:

Welcome ***, according to the User's Guide the sensors must not be disposed of with municipal waste. In the countries of the European Union, separate collection of used electrical and electronic equipment is required. As the sensors may have been exposed to body fluids, they can be wiped down before disposal, for example with a cloth moistened with a mixture of one part household bleach and nine parts water. Should you have any further questions, we remain at your disposal. Best regards! 😊

.

The kit itself for 'mounting' the sensor in the body is also interesting. The kit consists of two parts: one is the electronics with the applicator, which does not appear to be sterile, and the other is the electrode with the needle that inserts it into the body.
This is what the electronic part and its applicator look like:

.

This is what the electrode and its transport packaging looks like.

.

And this is what the rubbish looks like after the sensor has been installed.



I leave it to the reader's own judgement as to whether this could somehow be handled better. Leading perhaps, however, to the answer I would like to get: Abbott in the West launched a third version of the device some 2/3 years ago (a fourth version has already been announced for 2022) and boasts, among other things:

Abbott wrote:

. The FreeStyle Libre 3 System was designed to be more sustainable for the environment with a 41% reduction in plastic use and 43% reduction in carton paper.1 The new sensor design aligns with Abbott's continued commitment to waste reduction.

How is it that we have version 2 all the time? That's a good question, too, but since our National Health Service subsidises 70% or 80% of the price all the time, maybe it's not worth launching a new product for the time being since business is booming. All the more so as Freestyle Libre 2 is the only glucose monitoring system that is, in a way, unconditionally subsidised: other systems can only be reimbursed to a person who, for example, has a problem with not having low blood sugar (or other additional attractions that may be associated with diabetes), which may simply be dangerous. FL2 users are actually only required to use the sensors quite sensibly.

Wow, what an elaborate piece of writing I've come up with. .

Kubasaw wrote:

The FL2 users are actually only required to use the sensors quite sensibly.

.
And that's interesting, because I haven't been able to get a prescription for it and if I want to use that sensor, I have to pay full price. FL3 is available with us, but you can't download the app. You can download it for example in Germany and install it, but when the app senses that the tel is in Poland, it locks up.

Generally, the indication for reimbursement is 'intensive insulin therapy', i.e. at least 3 injections per day. This qualifies virtually all type 1 diabetics and probably some type 2 who are treated with insulin rather than tablets.

To get another reimbursement order this:
- You need to see your doctor within a certain period of time (4-6 months from your previous visit) (all in all unless there are super queues, it's manageable, it can also be a remote visit).
- Show that we have probably 75% of the time covered by measurements (with BLE, this is actually automatically done, but depends on the official app, because how do you show it otherwise?).
- Have 70% of that time from the previous point sugar between 70 mgdl and 180 mgdl or have a haemoglobin level below 7,5% or fulfil other therapy goals defined jointly with the physician (in particular, the third component of the alternative appears to be widely interpretable).

At this point, as far as I know, more conditions are unlikely to be present.

As to the causes of the onset of diabetes - I will not comment. I do not have sufficient knowledge.

As for the construction of electronics, costs and recycling:

Why do we have this and not another MSP here? Because it has RF, it's infernally cheap - before Covid to buy for ~30gr / piece when buying 100k. It has indestructible memory - it's on FRAM. It has a DAC, a pretty tolerable ADC. We can put it to sleep almost completely - including RAM, we write data for 8h in program memory, it can withstand 10^15 writes minimum.

chip for BT - it's economical and also a hell of a lot cheaper, pre Covid ~25gr / 100k.
Board about 1-2£ at 100k
Plastic case? 2-3£. I'm betting the most expensive is the housing for this.

So the price is only made up by these plastic inventions for applications etc. in general, battery replacement. Some programming, chipping it out of the case etc. will come out more expensive than manufacturing.

As for damaging the sensor - dropping to a low value and a sudden rise can lock it in programmatically, not physically damage the sensor. Likewise staying in a low glucose state for a long time. Some safety algorithms, it's better to 'break' a device than to let it show stupid things - which can endanger the life of the owner.

But the latter is just a guess, I don't know the exact working system of this 'needle'.

I wonder if it is possible to change the programme of such a uC? Does anyone perhaps have some "bulk" pieces? I'd be happy to send a label to the package:) .

That is, the cost of production is $5. And the selling price is 280£. x millions of units makes it a good business.

pawlik118 wrote:

That is, the cost of production is $5. And the selling price is $280. x millions of units makes it a good business.

How neatly you left out the cost of inventing the electronics, the cost of inventing the measuring electrode, the cost of clinical trials, the cost of setting up the production of something that is supposed to be sterile, among other things.

If your time is so cheap, how about I hire you to do the work? A bowl of potatoes will do?

This is probably produced in the hundreds of millions of units. So these additional input costs should quickly tend towards zero. Anyway, that's the manufacturer's risk and cost. That's what investment is all about. In addition, in this sector research and development is solidly subsidised by public money.
If someone thinks the price is right, imagine them throwing such a smartphone in the bin every 14 days:
.
When a new car model goes on sale it does not cost many times more than the previous model, even though the cost of developing it takes many years to pay off, or sometimes not at all. This is the manufacturer's problem, not the customer's.
The price is certainly also influenced (positively) by universal reimbursement.

Here an interesting article : Drugs should be cheaper. Research and development expenditure does not justify high prices .

"The world's 15 largest biopharmaceutical companies spent more on sales, general and administrative activities (including marketing) than on research and development between 1999 and 2018, and that most of the new drugs developed during this period offered little or no advantage over existing treatments."

"The authors note that most of these same companies also spent more on share buybacks than on R&D."

"In the US, the estimated net price of newly introduced prescription drugs has risen from an average of around $1,400 per year in 2008 to over $150,000 by 2021."

Unfortunately, with type 1 diabetes, a CGM system (Libre, Dexcom, or Medtronik) is unfortunately but almost mandatory if you want to live.
This unfortunately disturbs the law of the market, because demand is constant, so the price is not affected.
What's worse, it happens that availability is also limited and wholesalers sell sensors a month ahead max.

Macosmail wrote:

This is probably produced in the hundreds of millions of units. So these additional input costs should quickly tend towards zero.

I understand that you work in the medical sector and have extensive experience of such projects and are fully aware of how inefficient corporations are and how often scientific projects succeed.

Macosmail wrote:

Here is an interesting article : Medicines should be cheaper. Research and development spending does not justify high prices

.
I don't understand what you mean. Capitalism works like this. 100 million for R&D, 500 million for marketing, project costs of 800 million because there's still a 200 million bonus for management for success. If you then sell 10 million devices with a revenue (not price!) of £100 per unit it will barely pay for itself because you had to spend on marketing again.

Galareta wrote:

So the price is only done by those plastic inventions for applications etc. in general, battery replacement. Any programming, gouging it out of the case etc. will come out more expensive than manufacturing.

Definitely .

Galareta wrote:

What about damage to the sensor - dropping to a low value and a sudden rise can lock it out programmatically, not physically damage the sensor. Likewise remaining in a low glucose state for long periods of time. Some safety algorithms, it's better to 'break' the device as to let it show stupidity - which can endanger the life of the owner.

Definitely x2. I've had a couple of situations where the unofficial app continued to show measurements after a while when the official one showed sensor failure. This is the terminal state of this state machine. It is also fairly common knowledge that the sensor can last a bit longer than 14 days: all the end-of-life logic is done on the 'front-end' / mobile app side.

Galareta wrote:

I wonder if it is possible to change the program of such a uC? Anyone have some "bulk" pieces perhaps? I'd be happy to send a label to the pack:)

.
I have a dozen (dozens?) used pieces in a box collected for no great purpose. If you want one, I'd be happy to send it.

Kubasaw wrote:

I've had a couple of situations where the unofficial app continued to show measurements after a while, when the official one showed sensor failure. This is the terminal state of this state machine. It is also quite common knowledge that the sensor can last a bit longer than 14 days: all the end-of-life logic is done on the 'front-end' / mobile app side.

And when the sensor is damaged according to the official app, does it show correctly according to the unofficial app?

I expect the answer "depends" .

CosteC wrote:

I expect a "depends"

response.
Exactly so .

Kubasaw wrote:

CosteC wrote:

I expect a "depends"

response.
Exactly so

.
This is actually a good thing. Their algorithm correctly interprets sensor failures and is more likely to shut down rather than lie.
A very interesting topic.

I managed to take the sensor apart, unfortunately I don't have a microscope and can't read the IC, 1.5V battery.

.

klamot wrote:

unfortunately I do not have a microscope and cannot read the IC,

.
All you need to do is take a sharp photo at a good angle, it would be possible to read.

I took a picture again, but I have a poor camera on my phone, so you can't see anything further. I borrowed an eye magnifier from a watchmaker friend and managed to get a reading:

RF430
TAL160H
TI 428
AG0Q C

Has anyone tried to use these boards in their application? Is there communication, power, ADC ready, e.g. as a remote temperature or voltage measurement chip? Since they are rubbish after use? I have 2 people in my family using them myself, so their source I would have....

Jacek Rutkowski wrote:

Has anyone tried to use these boards in their application? Is there communication, power, ADC ready, e.g. as a remote temperature or voltage measurement circuit? Since they are rubbish after use? I have 2 people in my family using them myself, so their source I would have...

.
I don't know about Libras, but I've successfully reprogrammed Dexcoms one+ for my use, only there sits a more friendly NRF52 for me. There are contact points on the board that you can solder to or poke with test needles. There's also a hall sensor, so I was planning to convert this into a speed or cadence bike counter with BT communication, I even started a project on github for an alternative use for the residue, but on reddit I didn't find anyone willing to join in, and I don't have time to play around with it myself at the moment, and ended up figuring out the exposed pads and trying to program it. The disadvantage is the chip itself, which is on the board in a non-enclosed version, the silicon on top, so you have to protect it from light.

Does it give the temperature in the app too?