No. In general, I like Kubios, I think it's a great resource, but I do not use the figure it provides for this.

There are several reasons why, but briefly: a) respiratory *variability* isn't listed, you just get a mean, b) the method used I think is derived from RS wave height, and, in my experience, this is less accurate than more recent methods, and c) if you want to calculate RSA you need a respiratory time series, not just a mean.

The methods I use now are modified from this excellent paper by Lazaro et al which, if your hardware filters aren't altering your signal morphology too much, works very well for me:

https://www.ncbi.nlm.nih.gov/pubmed/24110587

It is an extremely simple method to program and implement in any language, and I have compared it favourably to other forms of direct respiratory measurement for my own curiosity (as did the authors).

Kubios, although in my opinion is one of the best HRV analysis software, does not provide information about respiration directly. If you are going to estimate respiratory rate (and it's variability) from the ECG signal, search with "ECG-derived respiration" and have a look at this paper as well:

https://www.ncbi.nlm.nih.gov/pubmed/25194875

As mentioned by James, pay attention that band-pass filtering is of importance for ECG-derived respiration.

In the first place you will not find respiratory rate variations from second to second. In view of the duration of one breath the timescale is much slower here (by at least a factor of 4-5) than in HRV. Second, what you do want is a time-frequency analysis or a wavelet analysis of your HRV signal. Then you might be able to track changes in the respiratory peak.

Programs to do this can be found in the Matlab user files. Kubios does not do this, as far as I know.

Would anyone be able to recommend a device to measure respiratory rate (so we can understand the respiratory spectrum and compare it with HRV spectrum)? Thank you in advance.

Gunjan Y Trivedi I suggest using chest belt or air flow sensor in recording respiration rate. I have been using both of these from Biopac Systems.

Best bet will be to use MATLAB. However, this will require writing a programme for your specific analysis de Novo which may take months if you have to learn MATLAB. It's the simplest way I could think of however.

Olá Maria,

Estás bem?

Eu uso Kubious e há uma rotina que não é complexa!

Bj

Lucimere