New AlphaHRV version

Indeed, the Firstbeat algorithm has little validity in the measurements, thanks to the collaboration of Bruce Rogers, we have been adjusting the algorithm, we have to think that we are fighting against the low processing capacity of the Garmin equipment, even so we have achieved something that can be worthwhile. We are currently developing an app that detects thresholds, Bruce Rogers commented on the high similarity of the respiratory rate with the behaviour of lactate and we have taken advantage of this to see the same behaviour in the respiratory rate, soon we are going to do massive tests, all with laboratory tests. We are financing all this with our own time and money, so we are going somewhat slowly due to lack of resources.

2 Likes

I see.
Could you provide one?

Tks

1 Like

do you mean you are using the Garmin Dual strap with the Polar H10 sensor?

1 Like

Yes. Works like a charm

2 Likes

We have not tested the Garmin HRM Dual heart rate monitor, but to check if it is valid you only have to enter Garmin Connect and check the Artifacts chart, in wko5 you can also check it by creating a chart and using the formula @artifacts.

1 Like

This is a lactate test and it is the same one we are using for the FR, if you do not want to reach 120%, up to 90% would be enough.

-15m 40% power=3s
-6m 50% power=3s
-6m 60% power=3s
-6m 70% power=3s
-6m 80% power=3s
-6m 90% power=3s
-6m 100% power=3s
-6m 110% power=3s
-6m 120% power=3s

Once you finish the test you stop it and record it, do not record cool down.

3 Likes

Overall, AlphaHRV looks like it is doubling down on breaths.
I’ve noticed that when I have too many artifacts it’s time to wash the band, mostly once a week is enough :slight_smile:
Looking at this graph of DFA a1 to time, would you say I have LT1 around green and LT2 around red? Is that what this is about?

1 Like

AlphaHRV does not duplicate respiration rate, as I said the measurements are totally on par with Movesense, it is not worth counting breaths as you do.
As for DFA-alpha-1 I think you do not have the concept very clear, the standard DFA=0.75 is used for first threshold and DFA=0.5 as second threshold, after more than 100 tests we have done, we can say that these values are valid for some and not for others, I do not think that any coach uses 2mmol for first threshold and 4 mmol for second using lactate. Besides in that image you have not calculated the linear regression and eliminated the part at the beginning, besides you have to change some parameters in Runalyze for that regression to work “well”. We stopped testing thresholds with DFA because first the test has to be very well done because it is affected by any movement, depending on the type of test is more reactive and finally with covid those DFA values have changed in many athletes, although there is scientific evidence I guarantee that not all have the thresholds in the “standard” values.

First image without covid and dfa at 0.75

Second image two months after passing covid

Same test, same conditions

Respiratory rate test with alphaHRV

Lactate

image

3 Likes

I have done some more testing with the latest version. An important thing to always remember is that the results for a1 and resp rate are always the average of a longer measurement window. For AlphaHRV, the window is 200 beats if I remember correctly. So if you want to verify a stable resp rate by counting your breaths, you should do that for at least 2 min and then compare to the result of the algorithm. It’s not quite instantaneous…

1 Like

That’s right, if you want to try with 300 beats contact me privately, depending on the device you can try it, I would give you a code to put in 300.

Not exactly. Data window for a1 and RR are not the same length. For a1, D=200 beats is the default value (configurable for some devices going into developer mode).
For RR calculation, default window is T=60s. You can modify this value, being limited by the beats selected for a1: (T/60)xHR<D
But you are right, a1 and RR are not instantaneous metrics, but average for some period

hi, I tried to perform this test today:

-used 6min protocol
-wahoo ticr with garmin epix watch
-ant+ - alphahrv; bt - garmiin epix been
-wahoo core trainer

i think my results are totally wrong :slight_smile:
but my question is,:
-at the beggining there is a countdown from zero to 100%. Have impression it is a base setup. What should i do during this countdown? sit relaxed, ride slowly, or start first step of the protocol?

-what should the artefacts show? what is a correct reading and what reading is wrong? my chart is just empty

-is there any difference between ant+ and bt? Should I use one over another to get better results?

In two days i will try again, so jope this time my reading will be batter :wink:

Maybe it counts breathing in and out separately?

No, it counts complete breathing cycles (inspiration+expiration)

1 Like

alpha-1 and RR in that chart are not good. Good quality RR-intervals data are required for this analysis, that probably your HRM is not providing. If you share the fit file I can check it.
Regarding your questions:

  • DFA analysis requires an amount of data to be carried out. The count to 100% referes to initial collection of data. Reaching 100% means that the app has enough RR data to start calculations. You can do what you want during this period, but you won`t have any alpha1 nor RespRate for that period, so I recommend not to start the test before 100% is reached
  • Artefacts (wrong data received from the HRM) should be as low as possible. It is assumed that >5% seriously affects alpha-1 calculation
  • ANT+ and BLE are equivalent and you should get same results regardless the communication protocol you select. However, if you want to natively save HRV in your device for further analysis in external soft (runalyze, aiendurance, …) you must take into account that Garmin uses ANT+ in a wrong way for that purpose. So, in that case, I recommend to use ANT+ for alphaHRV and BLE for the device
2 Likes

Hi,

Which app do you use to get this chart?

Thank you,

The app is in development, not ready yet

2 Likes

So, meanwhile no way to get a chart like that ?

Tks

thanks for quick response

I did another test with same result (so no point bothering you with fit file) but after that I switched to ble connection and there are some diferences:

  • not sure why but it takes much longer to get 100% (like 3x longer)
    -too tired to make another test today but readings are much lower alpha-1: 0.70 when resting
    -tomorrow will try again but probably need to get anothe betterr strap :confused:

The wahoo heart rate monitor does not work for anything related to HRV, so neither DFA nor respiration rate will give you correct data.

1 Like