Where communities thrive


  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
People
Repo info
Activity
  • Aug 18 15:40
    RyszardS commented #3025
  • Aug 17 14:37
    RyszardS commented #3025
  • Aug 17 06:06
    RyszardS opened #3025
  • Aug 16 14:55
    c4franklinjr edited #3024
  • Aug 16 14:53
    c4franklinjr opened #3024
  • Aug 09 17:49
    LadyLadedah commented #3023
  • Aug 09 17:45
    andyrozman commented #3023
  • Aug 09 17:43
    andyrozman commented #3023
  • Aug 09 17:41
    LadyLadedah commented #3023
  • Aug 09 17:40
    andyrozman commented #3023
  • Aug 09 17:39
    andyrozman commented #3023
  • Aug 09 17:29
    LadyLadedah edited #3023
  • Aug 09 17:29
    LadyLadedah opened #3023
  • Apr 04 18:20
    lee-b commented #2291
  • Mar 31 11:10
    haiweihw closed #3022
  • Mar 31 11:10
    haiweihw opened #3022
  • Mar 31 11:08
    haiweihw closed #3021
  • Mar 31 11:08
    haiweihw opened #3021
  • Mar 07 22:09
    Webifi commented #2291
  • Mar 07 20:46
    nicoleloop commented #2291
Djuric
@dircsem

Are there any plans to support the new smart watches from Apple und Samsung which can measure blood glucose with AAPS in the future? I know they are not yet available. But if so...? Thanks

Is that officially announced that this watches will measure blood glucose?

teleriddler
@teleriddler

@dirceusemighini No, it is just speculation at the moment but given the ekg function they added last time and the massive market for Type 2 diabetics and amatuer/professional athletes looking to measure blood sugar it would make sense if the technology is feasible. I would guess that it will exact enough for those groups but I would not expect the values to be exact enough to loop with. Raman spectroscopy for blood glucose applications is not that refined yet and the leaked image is shown with the larger prototype on a pig, not a human.

https://en.wikipedia.org/wiki/Raman_spectroscopy
https://www.europeanpharmaceuticalreview.com/article/77451/raman-spectroscopy-as-a-blood-glucose-monitoring-tool/

dvdv
@dvdv_gitlab
I think raman spectroscopy can be accurate enough, but at the cost of size
Djuric
@dircsem

@dirceusemighini No, it is just speculation at the moment but given the ekg function they added last time and the massive market for Type 2 diabetics and amatuer/professional athletes looking to measure blood sugar it would make sense if the technology is feasible. I would guess that it will exact enough for those groups but I would not expect the values to be exact enough to loop with. Raman spectroscopy for blood glucose applications is not that refined yet and the leaked image is shown with the larger prototype on a pig, not a human.

https://en.wikipedia.org/wiki/Raman_spectroscopy
https://www.europeanpharmaceuticalreview.com/article/77451/raman-spectroscopy-as-a-blood-glucose-monitoring-tool/

Thanks @teleriddler I'm searching for a watch to buy to see my bg, and my thoughts was that it wouldn't be that precise, but maybe it can evolve faster than the other sensors.

teleriddler
@teleriddler

Understandable, I have also been looking at watches as well and to have a general idea of where you are would be good but when I am already wearing a CGM that is probably more accurate I would assume just use XDrip to get the readings to the watch, however would always be good to have a backup if the sensor fails.

@dvdv_gitlab correct, the science is sound and exact but miniaturization has been a problem in the past. Who knows maybe Samsung and Apple have figured something out the others could not. Google dropped their glucose watch project a while back, I am guessing they hit a similar barrier with shrinking the package. The device sitting on the pig in the picture did not seem to reflect a wearable watch that a consumer would buy but there is not a timestamp on the image so if could be older than we think. Hoping for the best regardless and interested to see what comes in 2021.

To break through the limitations that arose in prior studies, Samsung’s researchers developed an off-axis Raman spectroscopy system that enables the direct observation of glucose Raman peaks from in vivo skin. Utilizing this system, they demonstrated one of the highest prediction accuracies among non-invasive technologies. The research team also developed an innovative technique for reducing the effects of movement artifacts on glucose sensing in Raman spectroscopy.

Understandable, I have also been looking at watches as well and to have a general idea of where you are would be good but when I am already wearing a CGM that is probably more accurate I would assume just use XDrip to get the readings to the watch, however would always be good to have a backup if the sensor fails.

@dvdv_gitlab correct, the science is sound and exact but miniaturization has been a problem in the past. Who knows maybe Samsung and Apple have figured something out the others could not. Google dropped their glucose watch project a while back, I am guessing they hit a similar barrier with shrinking the package. The device sitting on the pig in the picture did not seem to reflect a wearable watch that a consumer would buy but there is not a timestamp on the image so if could be older than we think. Hoping for the best regardless and interested to see what comes in 2021.

I hope that it can evolve to be more than just a backup. Another barrier is that it will need to be approved by the health agencies (aka FDA) to be used?

dvdv
@dvdv_gitlab
That would not be a hindrance for DIY folks ;)
teleriddler
@teleriddler
@dvdv_gitlab good find, the off-axis approach must be yielding better results and saving space, I did not read that in the Korean article I found. Thanks for sharing.
Martin Esser
@TinoLL_gitlab
Pref2021_OV_Adv.png
Is this a bug? This percentage seems to influence the superbolus as well as the bolus. I didn't find another place to define the percentage of the superbolus. E.g. I want 100% bolus from what the bolus calculator calculated but only 80% for the super bolus. How can I do that, please? Thanks
szantos
@szantos
@TinoLL_gitlab
Currently superbolus is hardcoded with 2 hours of basal and the % (deliver this part of bolus wizard result) applies for everything.
szantos
@szantos
@TinoLL_gitlab see also nightscout/AndroidAPS#300
TinoLL
@TinoLL
Thank you
Mark Brighton
@seattlebrighton
Anyone else here using Amazfit band 5? I just got it working with xDrip+, with some code changes in case others are interested. I'd like to get HR data from xDrip into AAPS to be used for automation. Please. tell me if anyone is working on that or already done it. Thanks.
Mark Brighton
@seattlebrighton
Amazfit band 5 costs $35, has 5+ days battery, waterproof, shows BG from xDrip, alerts me when it sees I'm being active. If the active alert could also include an AAPS automation, that would eliminate lows when I kitesurf, bike, ski, etc. and forget to set a high target.
Djuric
@dircsem
Hi Mark, I don't have a band, but I want to buy an Amazfit watch. I'm also trying to get information from smartwatches and put it together at someplace available to public access.
What do you plan to do with HR dat? Will you try to predict BG variations using that info?
wwwired
@wwwired
@seattlebrighton & @dirceusemighini - i've just started using a Mi Band 4 and was hoping to do very similar. IF HR >115 THEN SET TEMP ACTIVITY TARGET or something similar. When i'm planning to exercise i always use the TT's, but it's those unexpected activities that i think might benefit from an early high TT as opposed to just watching the drop 5-10 minutes later.
Mark Brighton
@seattlebrighton
@dirceusemighini @wwwired - Yes, Ideally I'd like to automate an adjustment to AAPS tuning based on "exercise effort" (measured by HR or other inputs like steps or some combination), adjusting Temp Target BG, and possibly Carb Decay and Insulin Activity Duration. I think those make the most sense, biologically. For now, I'll be happy getting 80% of the benefit by using just HR to set a TT of 140 mg/dl and maybe setting Insulin Sensitivity up by 20/40/60%, depending on HR range. I updated xDrip-plus code to properly recognize Amazfit bands and that code already has some HR handlers in it. Tell me if you know anyone who might know how to get HR data from xDrip (or Google Fit if necessary) into AAPS, to allow automation to use it. If you have other skills or experience with this, I could use your help.
Penta
@Pentagit
What is the purpose of the FOOD tab which displays the food presets defined in nightscout? It only shows the foods but it is not possible to use foods for carb-entries or such things.
Djuric
@dircsem

@dirceusemighini @wwwired - Yes, Ideally I'd like to automate an adjustment to AAPS tuning based on "exercise effort" (measured by HR or other inputs like steps or some combination), adjusting Temp Target BG, and possibly Carb Decay and Insulin Activity Duration. I think those make the most sense, biologically. For now, I'll be happy getting 80% of the benefit by using just HR to set a TT of 140 mg/dl and maybe setting Insulin Sensitivity up by 20/40/60%, depending on HR range. I updated xDrip-plus code to properly recognize Amazfit bands and that code already has some HR handlers in it. Tell me if you know anyone who might know how to get HR data from xDrip (or Google Fit if necessary) into AAPS, to allow automation to use it. If you have other skills or experience with this, I could use your help.

Very interesting, do you have a lot of hipos? I never thought on correlating it, this seems to be a great thing to do, for those who does a lot of sports. The main idea is to suspend before BG start dropping when HR is high? A question I have is, how precise is the HR sensor of your band? It's mi band 5 of just band 5?

What is the purpose of the FOOD tab which displays the food presets defined in nightscout? It only shows the foods but it is not possible to use foods for carb-entries or such things.

Where is this FOOD tab? I saw this collection at my mongo, but I didn't see any data on it, and I have a doubt that what was this used for?

Foxy7
@Foxy7
@seattlebrighton i was thinking about heart rate too recently, but as it's for my daughter I didn't think i could permanently get a watch on her. however i watched this video and realised that exercise/heart rate is way more complicated than i expected, and explains a few things historically i've not been able to work out before. https://www.diabettech.com/living-with-looping/living-with-looping-january-2021-the-replay/ scroll down to the box "Managing Exercise with a DIYAPS (or a commercial one!):"
rosaasor1
@rosaasor1
In the Actions screen, my Canula age displays: "-8327d-12h". Happened when I switched from my main Medtronic pump to backup. Doesn´t help to register a new pump site/insulin cartridge change. Believe an incorrect entry is registered somewhere. Using AAPS 2.8.2. Any advice?
Djuric
@dircsem
there seems to be a register in the future, I've had that with an incorrect bg reading and to fix it I needed to clean all aaps data. I don't know if this is the best approach to fix it, and if you do that, be aware that you will loose the data of battery change, sensor change, and everyting else.
dvdv
@dvdv_gitlab
Combining the food tab with the bolus calculator would be really great.
Or perhaps some way to allow EasyFPU to communicate with AndroidAPS so that both carbs and ecarbs are added based on the EasyFPU calculations
Djuric
@dircsem
Does EasyFPU write data to nightscout?
dvdv
@dvdv_gitlab
Nope. Entirely independent app
Great app tho
Djuric
@dircsem
Does this food tab, is something new?
wootmaster
@wootmasterslick
@rosaasor1 @dirceusemighini Cant you remove the specific entry from the treatments tab?
Mayan-t
@Mayan-t
Hello. When I try to generate APK, I get this error: "Failed to find build tools revision 29.0.2". I have SDK platform Android 9.0 installed. Android Studio v 4.1.2.
Andy Rozman
@andyrozman
@Mayan-t You will need to install those tools. If you google it, you will get correct instruction on how to do this. But short instruction is you need to click on SDK Manager icon, and then update SDK tools... If you are in window, there is SDK tools tab there.
Penta
@Pentagit
@dirceusemighini I don't know if it's new. APS docs have one entry for the tab (https://androidaps.readthedocs.io/en/latest/Configuration/Config-Builder.html#food). It sais it's for viewing food only.
Mayan-t
@Mayan-t
@andyrozman I followed wiki precisely, so I have installed SDK tools (Android 9.0 - 28, also Android 10 - 29). I tried to uninstall/install SDK tools, also deleted the project and cloned again. No luck.
Djuric
@dircsem

@dirceusemighini I don't know if it's new. APS docs have one entry for the tab (https://androidaps.readthedocs.io/en/latest/Configuration/Config-Builder.html#food). It sais it's for viewing food only.

Ah, thanks @Pentagit I thought that this was a nightscout tab. I saw this collection at nightscout database and didn't knew if there was someone writing to it. I was planning to track and see, if I can stabilish any correlation between food and post meal BG. If it's true I could schedule some post meal bolus to "flatten the bg curve", in my case altough I do carb counting, I have some post meals skyrockets, due to fat/protein.

Andy Rozman
@andyrozman
@Mayan-t Can you check out the project again and if you are asked to upgrade something, say no... This might fix the problem... Before you do that go into File menu and use "Invalidate caches and restart"
Mayan-t
@Mayan-t
@andyrozman Thanks, but it didn't help. I had to completelly uninstall Android Studio and all folders, install and it worked.
Why are two SMB settings (High temp-target raises sensitivity and Low temp-target lowers sensitivity) grayed out (cannot be enabled/disabled)? They were grayed on my previous version (2.6.1.4) and they also are grayed out after update to current 2.8.2 and import of settings.
Dave
@mistermintsgh
@Mayan-t Have you completed the objectives?
ArthurusDent
@ArthurusDent
@Mayan-t There was some kind of problem with theses two greyed-out options, maybe a safety concern.
Cybereu
@Cybereu
Its possible AndroidAPS accept treatment from Nightscout ?
ArthurusDent
@ArthurusDent
@Cybereu This has been disabled becaus of a bug in NSclient, I think. The bug is dangerous and can lead to overdosing of insulin. You can enable accepting treatments from Nightscout by enabling engineering mode but you should really know what you are doing. It is not clear yet, when the bug will be fixed.
mordechi-david
@mordechi-david
Peace
Peace
I started yesterday SMB and he is a bit too aggressive what will make him more gentle without changing profile because in the profile calculator he is good and I do not want to change it
Great download for all developers
It changes lives
teleriddler
@teleriddler

@mordechi-david It is in the documentation on SMB

https://androidaps.readthedocs.io/en/latest/Usage/Open-APS-features.html#super-micro-bolus-smb

https://androidaps.readthedocs.io/en/latest/Usage/Objectives.html

Objective 10: Enabling additional oref1 features for daytime use, such as super micro bolus (SMB)
You must read the SMB chapter in this wiki and chapter oref1 in openAPSdocs to understand how SMB works, especially what’s the idea behind zero-temping.
Then you ought to rise maxIOB to get SMBs working fine. maxIOB now includes all IOB, not just added basal. That is, if given a bolus of 8 U for a meal and maxIOB is 7 U, no SMBs will be delivered until IOB drops below 7 U.

A good start is maxIOB = average mealbolus + 3x max daily basal (max daily basal = the maximum hourly value in any time segment of the day - see objective 7 for an illustration)

min_5m_carbimpact default in absorption settings has changed from 3 to 8 going from AMA to SMB. If you are upgrading from AMA to SMB, you have to change it manually.

@Mayan-t What CGM are you using? If it is not a Dexcom CGM, then some of the SMB settings are disabled because the results of the CGM are not exact enough to be used with this feature. I use FreeStyle Libre 2 and SMB (even with Engineering mode turned on) is not completely functional due to this limitation.