RRI & RRI_Time as input to hrv()

[danibene]

Describe the solution you'd like I would like the HRV functions to be able to accept interbeat intervals and their corresponding timestamps as input rather than the peak samples. This seems to be a recurring issue (e.g. https://github.com/neuropsychology/NeuroKit/issues/682, https://github.com/neuropsychology/NeuroKit/issues/667). I also think it would add more flexibility in terms of how to preprocess the interbeat intervals before HRV analysis (outlier removal & treatment: https://github.com/neuropsychology/NeuroKit/issues/541, interpolating & detrending: https://github.com/neuropsychology/NeuroKit/pull/680).

How could we do it? I did this for a project where I had missing data & therefore the timestamps of the interbeat intervals were needed to properly interpolate in hrv_frequency(). I also made some other changes to the HRV analysis functions to account for the missing data (but I think the input format is probably better to address first on its own). Here is the current state of that branch: https://github.com/danibene/NeuroKit/blob/feature/adjust_for_missing_intervals/neurokit2/hrv/hrv_time.py#L148-L170

There, I added extra arguments rri_time and data_format but in retrospect, it would probably be cleaner if there were no extra input arguments and if the input type could just be automatically detected from the keys of a dictionary/columns of a dataframe ("RRI", "RRI_Time")?

Let me know if you are interested in this :)

[stale[bot]]

Is this still relevant? If so, what is blocking it? Is there anything you can do to help move it forward?

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs.

[DominiqueMakowski]

Seems I totally missed this (guess the stale bot can be useful), what's the status of this?

[danibene]

guess the stale bot can be useful

who knew ^^

@DominiqueMakowski I'm happy to submit a PR for this, checking whether the peaks argument in hrv() is a dict with RRI and RRI_Time as keys rather than adding any extra arguments like data_format? https://github.com/neuropsychology/NeuroKit/blob/5cd089e9d82e6f7fb98d2028e69840aee08f8ab3/neurokit2/hrv/hrv.py#L29-L32

[DominiqueMakowski]

Agree to check the peaks argument, best is to keep the number of arguments tight when possible