Feature request - Add Basal input and tracking

[Kustomwireless]

Is it possible to add a second type of insulin input so that Basal could be input tracked and graphed.

Thanks for all the work you do and for a great APP

[jstevensog]

Add an MDI, I second this request.

…

On 13 Feb. 2018 06:49, "Kustomwireless" ***@***.***> wrote: Is it possible to add a second type of insulin input so that Basal could be input tracked and graphed. Thanks for all the work you do and for a great APP — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#302>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AIQs8zrG2LLlbL_3cl-FXBOoX3Z5WQ27ks5tUJXMgaJpZM4SCufA> .

[jamorham]

Is this for Lantus or something else?

[jstevensog]

Lantus for me. I'm sure there are other MDI basal insulins though. It could be an option in settings. MDI yes turns off all loop/temp basal entry, and just provides a basal entry instead. Cheers On 14 Feb. 2018 09:19, "JamOrHam" <notifications@github.com> wrote: Is this for Lantus or something else? — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#302 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AIQs88-BrfYg8r-gCBFdtR1UtNzym83Hks5tUgp4gaJpZM4SCufA> .

[Kustomwireless]

My wife uses Levemir now but it would great if we were able to select any of the common insulin's and set onset delay time and duration. She is also MDI

Thank you for looking into this and all your hard work.
chuckk

Is this for Lantus or something else?

[WanaGo]

Would be great to have the ability to enter different insulins each with customisable onset delay and duration, I agree. Some people use Regular (Actrapid) for protein bolusing, then Fast Acting like Novorapid/Apridra etc for food, then Lantus for long acting, and some even then do Inter Muscular shots for stubborn highs which the duration is about half what it is subcutaneous, but much faster acting. Would be awesome to be able to enter what you used and have xDrip plot its estimates based on those.

[lee-b]

There's no need to name the insulins, or even have onset/duration initially. "Simply" having a "basal" measurement that's shown on the graph like the boluses, but as a straight line until changed would be great for pump users, at least, and probably great for everyone as a starting point. Adding insulin profiles that change the (basal) IOB and delivery rate could be done later. Main thing is, it would let you track basal changes, which can happen frequently with keto.

[Kustomwireless]

Really need to be able to enter Onset and Duration time because it is so different for each person.

[Kustomwireless]

What is required to get someone to work on this?

[andywilson]

I'd also find this very useful. I've used Lantus, Levemir and Toujeo, and each has its own insulin profile and duration. Happy to help test if needed.

[filipedanunes]

It would be great to have this feature.

[tremor]

That would be great, and maybe a drop-down list of 2-3 types of insulin that is often used in the treatment input field...

[Kywalh]

Hi all
Any news on this request ?
I also find this very useful to understand the mechanisms !!!

[DavitBS]

It would be great to have visual basal tracking as it can be shown in Nightscout.

[ghost]

1+ for basal insulin also from my side.

[hrasky]

I agree, basal recording and prediction would be great

[michelle197602]

Another vote from me, I'm a pump user so constant line of basal dripping in

[st0ber]

I would like to see basal and tbr.

[Kywalh]

what's tbr ?

[st0ber]

temporary basal rate

[YYurchuk]

Manually correct Onset and Duration time of insulins profiles will be very, very helpfull.

[psonnera]

Having the possibility to add insulin duration (3/9/20) or dual choice (fast/basal) in the treatment would be great.

[DunePlodder]

I would be happy to simply be able to record my MDI basal injections in xDrip (& Nightscout would be good). Then everything is recorded in one place - currently I keep a separate log just for my Levemir injections.

[n3bbf]

I would also really appreciate the ability to log and track my basal injections. Have always logged it in as a comment but graphing it on the display would allow me to better visualize what is happening (and when).

[candelanaj]

1+ for basal insulin also..

[yepee1]

Personnaly, I add a note using the Treatment Note icon. I type the Insulin name and the quantity. In my case I write: Tresiba 17 units.

Then It's possible to track theses entries using the Note Search button.

Hope this will help someone :)

[stukov4]

Hi, +1 for basal insulin also..
Is this happening or is it just a thing a bunch of people wants but there are not real news?

[AnaHarr]

This would be very useful. Using Lantus as Basal on MDI for our daughter.

[filipedanunes]

Is this for Lantus or something else?

Lantus here.
Can you tell us if there is any chance of this to be developed in a near future? :)
It would bring joy to a lot of people, theres any chance we can crowdfund this?

[filipedanunes]

I'd also find this very useful. I've used Lantus, Levemir and Toujeo, and each has its own insulin profile and duration. Happy to help test if needed.

Can help testing too

[Dainem21]

Since I started to use only Tidepool, I realy like to see thhis feature, too.
Thanks in advance.

[slinGitHub]

This would be a very important feature for me too +1. (Basal: Levemir/Bolus: Humalog)

[MariVargas87]

+1 Here too! (Basal: Tresiba / Bolus: Fiasp)

[redman8]

Basal please :) Humalog

[FreDiabetics]

Yes please Lantus as well :) would be great to see that feature

[k2s]

this ticket is obsolete ... there is active development in this pull request #973

it is working (I am using it) and Lantus is also included. you need to compile own version or wait until it is accepted in Nightly

[Son1k02]

@k2s fantastic news! Any instructions for those of us unsure how to compile a version?

[rudysroost]

I have the latest build, I cannot find an option to enable this. How do I start using it?

…

On Thu, Oct 10, 2019 at 1:17 PM k2s ***@***.***> wrote: this ticket is obsolete ... there is active development in this pull request #973 <#973> it is working (I am using it) and Lantus is also included. you need to compile own version or wait until it is accepted in Nightly — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#302?email_source=notifications&email_token=AC6WQNWKFAL5G2A7UEEAD7DQN5WVBA5CNFSM4EQK47AKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEA5JB5I#issuecomment-540709109>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AC6WQNVFQDYOENCH7CIVYC3QN5WVBANCNFSM4EQK47AA> .

[FreDiabetics]

this ticket is obsolete ... there is active development in this pull request #973

it is working (I am using it) and Lantus is also included. you need to compile own version or wait until it is accepted in Nightly

@k2s how can I compile that?

[k2s]

@Son1k02 @rudysroost @wagnefrede if you are interested in testing, please read and subscribe to that pull request comments

[mistermintsgh]

As well as the requests for the standard MDI basal to be implemented above, is it possible to include pump basal information in xDrip?

[lee-b]

Pump basal is already shown now, @mistermintsgh. It's the blue rectangles at the top. If you're not seeing those, check your settings, and that your particular pump is supported fully.

[mistermintsgh]

I don't have that @lee-b Presumably it's only available then for connected/loopable pumps? There's no way to display basal information for an offline pump?

[bennettnicholas]

I mean my pump puts in 1 unit every hour in, .1 increments. Is there a way to add "basal" support but you can just hard code it to track 1 unit every hour so the prediction curve is better?

[Nemz716]

Please can we have basal rates and the ability to add temporary basal rates be added to the app! Thank you 😄

[scgibbs]

Did anyone ever start on this?

[filipedanunes]

There's an on going work: #973

[LakeSideCitizen]

There's no need to name the insulins, or even have onset/duration initially. "Simply" having a "basal" measurement that's shown on the graph like the boluses, but as a straight line until changed would be great for pump users, at least, and probably great for everyone as a starting point. Adding insulin profiles that change the (basal) IOB and delivery rate could be done later. Main thing is, it would let you track basal changes, which can happen frequently with keto.

Well I use different basal, with different effect duration. To define just one wouldn't work for me

[Kustomwireless]

Why can it not be set up just like the orginal insulin entry where we can set onset and duration for the type of Insulin that we are using as Basal this has worked in GLIMP all along.

[dyjaks]

Would be great to have the ability to enter different insulins each with customisable onset delay and duration, I agree. Some people use Regular (Actrapid) for protein bolusing, then Fast Acting like Novorapid/Apridra etc for food, then Lantus for long acting, and some even then do Inter Muscular shots for stubborn highs which the duration is about half what it is subcutaneous, but much faster acting. Would be awesome to be able to enter what you used and have xDrip plot its estimates based on those.

This is what I use R for and this feature is helping a bunch. All I'd want is a way to input protein or have a second carb entry that I can change absorption time for. As is the prediction gets thrown off because it assumes I ate 30g carbs that will be absorbed fast when really it's 30g "carbs" that will absorb quite slow.

[gapolv]

Would be great to have the ability to enter different insulins each with customisable onset delay and duration, I agree. Some people use Regular (Actrapid) for protein bolusing, then Fast Acting like Novorapid/Apridra etc for food, then Lantus for long acting, and some even then do Inter Muscular shots for stubborn highs which the duration is about half what it is subcutaneous, but much faster acting. Would be awesome to be able to enter what you used and have xDrip plot its estimates based on those.

This is what I use R for and this feature is helping a bunch. All I'd want is a way to input protein or have a second carb entry that I can change absorption time for. As is the prediction gets thrown off because it assumes I ate 30g carbs that will be absorbed fast when really it's 30g "carbs" that will absorb quite slow.

Exactly! Same issue here, do you have any thoughts or suggestions? Do you know any other app that could handle this? Thank you!

[tolot27]

PR #1388 adds more insulin types including basal insulins. Visualization of basal insulin is not included to avoid touching the business logic.

[simonpickering]

I would also like to have basal insulin considered in the calculations, having read most of the thread in PR #1388 (much of which would probably be better in this PR re relevance) I have at least one concern/query, which is that the algorithm needs to know or be able to estimate a basal glucose generation rate in order to do anything useful. Does anyone have any thoughts/papers/etc. with which to look at this - I'm guessing one could start with a flat curve and then automatically populate it by looking at periods with no food and/or bolus insulin in the system.

I definitely think it's worth a go (along with some automation of insulin sensitivity adjustment though the day/based on activity/previous activity), and I've started writing some MATLAB code to post-process my exported data to see whether any of this might work.

My reason for wanting this is that I'm on basal/bolus and my insulin sensitivity is very different overnight and during the day (and my basal insulin is a Lantus type ~24h action). My sensitivity also changes significantly depending on how much exercise I do (as well as the usual illness, etc.), and I also get the feeling there's something else in the mix which is more short term - perhaps muscle glycogen restoration after high intensity/long duration exercise (e.g. riding bike a reasonable distance).

I guess some of this could be "fixed" by my using a basal insulin which lasts ~12h and injecting different quantities for day and night, but I am where I am for the time being and it would be useful to have this functionality to understand/predict things better.

I guess one way of avoiding the arguments would be to allow the end user to simply select the "business logic" they desire (from a list) and everyone can then be happy with whichever they want to use.

[tolot27]

I have at least one concern/query, which is that the algorithm needs to know or be able to estimate a basal glucose generation rate in order to do anything useful.

A new algorithm would change business logic which is highly debatable. The adjustment of the basal rate is completely separate from bolus insulin and discussed independently in all CSII studies I know of. It is also highly dependent on the individual. Implementing algorithms to address this is a huge effort.

Does anyone have any thoughts/papers/etc. with which to look at this - I'm guessing one could start with a flat curve and then automatically populate it by looking at periods with no food and/or bolus insulin in the system.

The basal rate is highly dependent on age and day time. Renner et al. developed some standard curves, see https://www.diabetologie-online.de/_storage/asset/1916029/storage/kirchheim:article-lightbox/file/334471441/16215326.jpg for instance. I've this data extracted, already.

I recommend to just implement an input option for the amount of injected basal insulin and displaying it as a single data point in the graph.

[simonpickering]

Thanks for the bar chart data :)

To your latter point, this is already available in XDrip, but it does nothing to the algorithm calculating predicted blood glucose (which is the interest I have). So perhaps this is the wrong place to discuss it if this issue/thread has already been resolved, and I should therefore open a new issue.

I agree that implementing algorithms is an effort, but one must start somewhere - I completely understand the desire to separate the basal and bolus calculations for the sake of simplicity of analysis, however as the graphs you provided show, unless you are on a pump you cannot account for the basal effects completely so they need to be taken into account when calculating boluses (which is presumably why the CSII studies don't need to do this as they assume an effective basal profile).

I can also well imagine that the basal requirement curves are quite different for different people, which indicates to me a good reason to look at whether at least some of this data can be automatically extracted from observed responses. Certainly from my point of view (not on a pump) that's my interest. Perhaps I am the only one who is not on a pump and is interested in this, who knows.

While I understand the reluctance to alter the "business logic" (for the sake on consistency and on the assumption that it works for most people), it would enhance the platform (IMO) if one were able to select different "business logic" implementations to handle what you wish, whether that's the addition of GI/GL, protein:fat:carb ratios or basal insulin, etc, even if these eventually turn out to not be as effective as the standard "business logic". The unfortunate alternative is that people will fork the application in order to implement their particular flavour of "business logic", which would be a shame for overall productivity.

[Navid200]

This issue, based on its title, was about adding the ability to enter basal insulin.
That capability already exists now. Hence, I am closing this issue.

If you like something else, please find a related open issue and add your suggestion to it.

If you like to bounce ideas, please use the support channels. Many developers frequent those as well.
https://www.facebook.com/groups/xDripG5
https://gitter.im/jamorham/xDrip-plus