I was wondering how the calculation of a value for "Stress" is performed. There is no galvonic skin response being measured. Only heart rate and movement. Just curious!
Actually heart rate variability. While it correlates with "stress", fatigue and well-being, we don't know how the values (0-100) correspond to real world values as we don't know the algorithm. But consider it a within person measure with values relative to your own measurements.
I was wondering how the calculation of a value for "Stress" is performed. There is no galvonic skin response being measured. Only heart rate and movement. Just curious!
In simple terms, the science comes down to an analysis of beat-to-beat changes in your heart rate, known as the heart rate variability (HRV) phenomenon. Because your heart function is regulated by the autonomic nervous system, it is possible to observe interplay between the sympathetic (fight-or-flight) and parasympathetic (rest-and-digest) branches of the ANS system.
I think it gets a little tricky in the sense that what's being measured there is "physiological stress" - which isn't quite the same as "perceived stress." In other words, it tells you if your body is experiencing stress, not if you, necessarily, feel stressed - although these two things do overlap often.
HermanB, Thank you for the info and for the participation in the board. Any chance you could give some insight to some additional firstbeat features? For instance, i'd love to know, is altitude (not elevation change) is taken into account when measuring Vo2 Max via the 5x? My workouts are all starting above 5400 feet.
It might make sense to have an "Ask Firstbeat" thread!
In simple terms, the science comes down to an analysis of beat-to-beat changes in your heart rate, known as the heart rate variability (HRV) phenomenon.
Is there any chance you can tell us a bit about the maths of the metric? I'm doing some research in HRV (medicine rather than sport) - it seems that time domain parameters, specifically ln RMSSD, are the main ones in sport - and it strikes me that an optical sensor might not be accurate enough for some of the high frequency stuff? Just interested to see how Firstbeat metrics match up with the ones I'm familiar with!
When it comes to using optical sensor data for HRV "stuff," a) these sensors have improved quite a bit in the past few years, even taking HRV-analysis needs into consideration and b) it often comes down to knowing when the data is accurate enough to be used, and when it isn't.
Firstbeat does host an annual HRV Summit year, so that might be of interest. It hasn't been officially announced, but I believe next year will be in London (May?). You might keep an eye open for that.
Looks interesting - for those who don't want to dig into the paper it appears to combine RMSSD and frequency - domain parameters into (unfortunately) a "black box" algorithm (presumably a trade secret) which spits out the stress metric. On the plus side, it does have the option of refusing to produce a number if the data quality isn't high enough - and I know optical sensors are good enough for HRV at rest (though they may overestimate respiratory effects, which might actually be a good thing depending on what you're using it for). Certainly the Firstbeat measure hits the mark of plausibility for me - it seems to categorise my days correctly! Once I get back into high - stress training I'll see how it matches against a morning reading of lnRMSSD. I'll keep an eye out for the Summit, thanks!
