Menu

CGM Sensor Emulation

In the near future, Type One Diabetes treatment technology will already be doing what you want it to do… sort of…

I have mentioned a copule of times I have an idea of how to be able to use the upcoming technologies that will bring T1D treatment a step forward…

We all know the limitations the companies have imposed on these devices:

— Closed. You have no way of knowing what they’re doing and why…

— Limited on customization. There is no way you can make the pre-defined treatment be more adapted to your personalized needs than the limits imposed by the company.

— Not interoperable. They all force the patient to use a set of technologies from the company decided by the device manufacturer.

Thinking about the upcoming technologies, I know one or more of the following will happen:

— Pumps will not allow us to remotely operate them. This is already a reality for all of us with diy closed-loop systems. We depend on super old technology, but you know, #wearenotwaiting and this:
https://www.wired.com/2016/10/hacking-car-pacemaker-toaster-just-became-legal

— Companies will probably impose encryption or other “””security””” features on the communication between sensors & pumps. I have not read / heard of anything related, but it seems the obvious move from their point of view.

After thinking and discussing, there is a huge opportunity for us to still be able to open up the new devices to customization… I think it could be with: CGM Sensor Emulation.

Technology will already be doing what you want it to do…

The upcoming hybrid closed-loop systems will already be doing what you want them to do… sort of…

The system will give you insulin when your blood glucose level is higher than what the company that created it wants it to be for you and it will stop giving you insulin when your blood glucose level is lower than what the company that created it wants it to be for you.

These decisions will be based on the readings from CGM sensors…

If you are able to emulate the signal sent from the sensor to the device, you could make the device do whatever you want it to do.

I just think there must be some considerations:

A) You must be able to emulate their insulin dosing algorithm to know exactly what the sent values will make the pump do. This could be done by testing and testing and testing so you can figure out the right way to do it and the limits.

B) You must be able to have a real blood glucose reading… the real sensor on your skin giving you the real values not the emulated ones. This real sensor is the one you would use for alerts and graphs and the stuff we are already doing.

C) If the sensor has encrypted communications, you must then be able to create a “simulated sensor site” which I imagine would be a device with a sensor connected to cables being manipulated by electricity in order for it to send the required blood glucose values to the pump… simulating the interstitial fluid resistance on our bodies.

“C” would be the most delicate part of the system. I have no idea how to build this, but I am sure someone in the community has the ability to do so. However, if the sensor communication with the pump is not encrypted, then you will be able to directly simulate the signal.

I can’t wait for the first new-generation devices to start hitting the market next year ?

Comments

comments