@Andy_Drake thank you again for the insights. The app is camera only for now, infact most of the work was around artifacts, as you can imagine ah ah.
About the 3 minutes window, yes it can become configurable, and it’s the only way to get reasonably realible Respiratory rate, LF power, HF power, LF power (n.u.), HF power (n.u.), LF/HF ratio (5 minutes would be better).
BTW I want keep it simple and straightforward and quick: thia means camera and 1-minute only for now.
Thanks!
If I can be honest, I think simulated HRV which you are doing via PPG technique is unlikely to be useful for people who take HRV seriously.’
With Polar Verity (PPG) and H10 (true ecg via Polar SKD) there is variance by several ms and both are gold standard devices
well I do not want to be too negative so maybe my feelings on camera derived ‘HRV’ is misplaced?
have you tested simultaneously with another app. recording connected to a chest strap with a strong reputation for R-R interval accuracy? Do trends (even if rmSSD differ) correlate?
@Andy_Drake I think you are totally right. I think the same way, but the scope of the app is simpler.
No, I didn’t cross test the data, and as far as I know, also the best measurements tools and protocols are subjected to (massive sometimes) errors. HRV4training has some papers that “prove” camera reliability and I’m not here, nor have the deep knowledge required to discuss those studies.
My app wants to be simple and give you a ballpark idea of what’s going on. HRV4training via camera was useful in several ways for me, especially predicting illness some days in advance.
Basically, I agree with this one, so I don’t feel like over-complicate it:
Thanks!