Just to confirm, in your conversion from cartesian to polar coordinates, you are using the RHR as radius and the HRV as angle? I tried to confirm it in the code but my matlab days are long gone
I would be careful with following 100% the ithlete guide, I think that in this case it would not be the right thing to do. I have been testing with Iñigo for a while and not everything he says is applicable as it needs a post analysis. The ImReady4 application gives more interpretation information as it is indicating the load, training time, IF, duration, sDNN, etc…, at first glance we already have more information, ithlete probably see your baseline of more days through the database that has the athlete. My advice is to apply ImReady4’s advice and do an analysis of the data the app is giving us.
From center to the outside (+3 to -3sd) is the HRV.
From left to right (clockwise) is the RHR (-3 to +3SD)
Basically, you look from the right side of the square chart in a panoramic view.
If that’s understandable
This is awesome! Tx a stack for putting it together!
I am going to be adding support for scripts for custom wellness fields soon which might make it possible to do this sort of thing inside Intervals.icu … at least the calculations. Completely custom charts is another whole piece of work.
and made it a lot better as well, the API alone is awesome, now with the javascript fields as well, thats already far ahead of training peaks and wko in my opition.
ImReady4 brings a Xmas gift today
In version 4.00, two pages are displayed for each athlete
1 - the standard one available till now
2 - trend charts for rMSSD, SDNN and RHR
Trend is plotted for the last 30 days. The expected normal values range is defined by (ma +/- sd), being
ma = moving average of last 7 days
sd = moving standard deviation of last 7 days
Values out of range should be plotted in faded red, values within the range in faded green
If you find any difficulty reading the value on top of each bar, try to zoom in and drag the chart
Tried it there. No issues except like you mentioned that it might be difficult to read.
It’s very cluttered with the values shown on the graph. Zoom doesn’t work very well but I think if you omitted the values in the graph and just had the bars then it would be a much cleaner graphic.
You get away with the values in the rHR chart but not rMSSD or SDNN
I’m mobile now and can’t post a screenshot (never seem to be able to add a screenshot from a mobile to this forum)
A question re the graphing. Why a 7d SD?
Is not recommended we use 7 day average as a baseline and 60d +/- 0.75 SD for normalized?
RHR uses 30 +/- 1*SD for normalized?
Don’t worry about period for moving average and sd calculation
v4.00 is the framework for this upgrade. I will set these values later and will provide also ability to customize them by means of settings in config_athletes.m
Same for values on the charts, that you will be able to enable/disable
I have no more time right now, but I think it was worth posting the update like this.
Any additional feedback is welcome
Perhaps it could be interesting some kind of alert on the sDNN today value. I mean, you can get the “Go on! Train as planned” message, but you can have an unusual (too high or too low) sDNN value.
In 60d the data may be biased for several reasons. It is important that you have your own protocol, not what others tell you on the internet. You will be able to adjust better with fewer days. As for deviations each athlete has a data, hrv values can not be seen as something standard and if something personal for each one, it is important to understand this, there are athletes who are able to lower the hrv and work well and others who can not lower it much, my advice is that you adapt everything to each of your athletes, customizing everything.
Look how well the trends look at 30d, rmssd increases within its baseline, sdnn also and heartrate is decreasing and creating the trend on those specified days, this is good groundwork and an excellent fit.