Had pretty good two weeks, plus I am now into starting my 2nd year of running DIY closed-loop systems (OpenAPS first, then Loop), so it’s time to post something! First, the obligatory Dex 14-day report
Over the past year, my 2-week SD has typically been in the 20-25 range, so 17 is exceptionally good for me. The (60-140) time in range is also above my typical of around 95%. Here is how a Nightscout distribution report looks for the same 2 weeks:
I like how it gives the number of readings and some interesting in-range and out-of-range statistics. Finally, for those of you who may be interested in details of how a closed-loop system (Loop) actually works in practice, here are my 3 typical days with Nightscout day-to-day reports showing bg curve, basals (light blue), boluses (dark blue bars), “carbs” (red bars), and some more daily statistics. Information overload, perhaps, but I like to occasionally take a look at these daily data-dense reports (a side note: auto-tuning can be done based on the data, thanks to the latest work by Scott and Dana (OpenAPS)). Note that the “carbs” shown include everything I need to bolus for, which includes actual carbs (around 60-70% of the number shown), as well as protein and fat.
Day A: Loop did essentially nothing overnight, which was then followed by relatively light meals during the day (for some small snacks I do not bolus at all). For a larger dinner, I did pre-bolus for fast carbs, and let the Loop high-temp for the rest of that meal automatically.
Day B: much more Loop action overnight. First it saved me from a low by low-temping from midnight to around 1:30am, and then bounced temp basals mostly up. A much larger lunch: pre-bolused too early, Loop helped correct this a bit by low-temping first, followed by high-temping for lots of relatively slow stuff I had for lunch. Dinner went pretty well.
Day C:
Lots of Loop overnight activity again - it was not quite able to low-temp enough to negate the tail end of a dinner bolus; but I stayed up working anyway, took minor (3g) carb correction after midnight, and let the Loop surf through the rest of the night. A much better large lunch than Day B, followed by several snacks and dinner through the end of the day: semi-automated surfing in action. Very happy to hit 100% TIR on days with lots of food!
Have fun with data!