Forum: open-discussion

RE: peak frequencies and amplitudes within frequency bands [ Reply ] By: Constantino Antonio on 2019-05-06 16:12	[forum:46672]
Hi Adeline, We have detected an issue related with the computations of the PSD. If you want to get values closer to those of Kubios, you may use the following quick and dirty solution: delta = median(diff(HRVData$FreqAnalysis[[index]]$periodogram$freq)) CalculateEnergyInPSDBands(HRVData, index) * delta This will be fixed in next release (current release is 4.2.4) Regareds

RE: peak frequencies and amplitudes within frequency bands [ Reply ]
By: Adeline Lacroix on 2019-05-02 15:31

[forum:46669]

frequency analysis.PNG (16) downloads

Dear Constantino,

I understand what Aaron means as I have the same problem and would like single values for one frequency analysis for one subject(see the file attached which is an output from Kubios).

I tried the CalculateENergyInPSDBands function and I got some results but they differ totally from what I have with Kubios and I suppose that these values are refering to something else. I suppose they correspond to Energy in each frequency band, but I don't know how to have similar values that I have with Kubios. Could you please help me?

hrv.CHANI1 = CalculatePSD(hrv.CHANI1, 4, "ar")
Calculating Periodogram using AR modelling
> CalculateEnergyInPSDBands(hrv.CHANI1, 4)
[1] 240215.23 33358.92 197728.63 51271.80

Best Regards,

Adeline

RE: peak frequencies and amplitudes within frequency bands [ Reply ]
By: Aaron Godlaski on 2015-08-21 13:37

[forum:42485]

Constantino,

I am traveling this weekend, but will check it out soon.

I think the below calculation of PSD may work. To try and clarify, I am working with HR data from multiple individuals and need single values per person for statistical comparisons. These can be expressed in terms of frequency (Hz) and amplitude (PSD of a given component) as an absolute value ms^2 within a frequency band or as ms^2/Hz to denote power within a band. For example, the reference value for normal adult LF (ms^2) = 791+/-563, or LF power (ms^2/Hz) = 1170+/-416.

I hope this clarifies rather than confuses what I am trying to compute.

Thank you for the quick response.

Be well,

Aaron

RE: peak frequencies and amplitudes within frequency bands [ Reply ]
By: Constantino Antonio on 2015-08-21 10:40

[forum:42476]

Dear Aaron,

I am not sure if I understand you correctly... If what you want is to compute the total spectral energy in a concrete frequency band you may install the current beta version of the package

install.packages("RHRV", repos="http://R-Forge.R-project.org")

and do something like....

data(HRVProcessedData)
hr=HRVProcessedData
hr = CreateFreqAnalysis(hr)
hr = CalculatePSD(hr,indexFreqAnalysis = 1, spans=c(9))
CalculateEnergyInPSDBands(hr, indexFreqAnalysis = 1)

I hope this is what you wanted. If not, please ask again!;
Kind regards

peak frequencies and amplitudes within frequency bands [ Reply ] By: Aaron Godlaski on 2015-08-21 02:49	[forum:42474]
Does RHRV currently provide the ability to calculate specific values denoting peak frequency within a given spectral band, or the peak power as denoted by the amplitude of peak frequency in a given spectral band? (In a fashion similar to its output of time domain statistics.) Apologies if this is confusing, I can attempt to clarify if need be, I am new to this program. Thank you.