search

nightscout / AndroidAPS

Opensource automated insulin delivery system (closed loop)
https://wiki.aaps.app
GNU Affero General Public License v3.0
732 stars 1.78k forks source link

Evaluate new generation Eversense (labeled E3) and enable SMB #2121

Open ga-zelle opened 2 years ago

ga-zelle commented 2 years ago

I got this new Sensor type 27 days ago and its performance regarding deviation and smoothness is a quantum leap forward. It looks comparable to G6 which I tested a while ago.

I volunteer to work with whoever to assess and compare its real life characteristics and come up with a recommendation.

MilosKozak commented 2 years ago

is there a study or more users?

ga-zelle commented 2 years ago

In Germany the roll out startet 1.Aug. So we are getting more and in 6 moths the old XL version will be completely replaced in the market. The US startet even earlier.

The system is officially qualified to support treatment decisions without finger prick control.

ga-zelle commented 2 years ago

Extracts from the Precise Study are contained in the user manual which can be downloaded : https://www.eversensediabetes.com/sites/default/files/2022-10/LBL-4002-01-001%20Rev%20E_Eversense%20E3%20User%20Guide_mgdL_R2_web.pdf

See page 204ff.

4RK4N commented 2 years ago

im getting my first E3 in 11 days. its certified for use with official AID solutions like the omnipod 5 or dbgl1 system just like dexcom is. although its still too new for the makers of those systems to adopt it. i would open my nightscout site for select reviewers if needed. so they can check performance and smoothness. or provide any other info if neccessary.

ga-zelle commented 2 years ago

I am a firm believer in using rational criteria rather than "It feels good". Before we start diving into the data is there a set of criteria how to come to a conclusion?

I have an emulator for AAPS which can re-run any actual loop with different settings and compare results. That can be used to compare loop results with SMB enabled vs. disabled.

cO5Mo2 commented 7 months ago

is there a study or more users?

Hi, I am a physician myself and a type 1 diabetic for 38 years. There are indeed a few studies now that show the accuracy of the Freestyle Libre 3. I would be very grateful for the option to activate all automation features of AAPS while using FL3. Thanks for your amazing work!

https://pubmed.ncbi.nlm.nih.gov/38189290/

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10064376/

Oh sorry, wrong issue. I meant to post it in a freestyle Libre thread of course