fear-induced HPR: pulse oxymeter

[amjh23]

Dear experts,
Is it possible to calculate the fear-induced HPR from pulse oxymeter data recorded during an fMRI experiment, or does it not make sense because of the typically lower accuracy of the pulse oxymeter data? If it can be done, are the preprocessing steps the same as with real ECG? Our pulse oxymeter data also contains time stamps of R-peaks. Can we import these to PSPM, or should we rather apply PSPM's heartbeat detection algorythm to the waveform data?
Many thanks for your help!

[dominikbach]

It's certainly possible to quantify fear-conditioned bradycardia from PPG data - see figure 5, panel "experiment 4" in https://doi.org/10.1111/psyp.12778

Preprocessing consists in finding the pulse time stamps (PPG) or R peaks (ECG) - from then on, it's the same pipeline.

You can certainly import pulse time stamps into PsPM - select "heart beat" as channel type in the matlabbatch GUI (or "hb" in direct function call).

Whether pulse detection of your device, or of PsPM, is superior, is hard to say. If it's a clinical device it probably uses a validated and high-quality algorithm. On "good" data, all algorithms will perform fairly well, you could visually compare the algorithms on some artefactual data stretches to get an idea of which one is better.

[amjh23]

Thank you very much for this extremely helpful answer!

[amjh23]

Dear experts,
I have another question regarding the PPG data: Is it possible to edit a PPG-derived heart beat channel by hand, e.g. in the ECG editor? I have some very faulty PPG-based HB data (many missed R-peaks), which I could improve via visual inspection, However, when I tried to open my PPG wave and HB channel in the ECG editor, I got the following error message:

_"Index exceeds the number of array elements. Index must not exceed 0.

Error in pspm_ecg_editor>load_data_file (line 866)
handles.data_chan = ecg_chans(sel_ecg_chan);

Error in pspm_ecg_editor>pbChangeFile_Callback (line 787)
load_data_file(hObject, handles, fn);

Error in gui_mainfcn (line 95)
feval(varargin{:});

Error in pspm_ecg_editor (line 55)
gui_mainfcn(gui_State, varargin{:});

Error in matlab.graphics.internal.figfile.FigFile/read>@(hObject,eventdata)pspm_ecg_editor('pbChangeFile_Callback',hObject,eventdata,guidata(hObject))

Error while evaluating UIControl Callback."_

I am using PSPM 6.1.2 on Windows 10.

Could I work around this problem by (wrongly) defining the PPG wave as an ECG wave?

Many thanks for your advice!

[dominikbach]

To load PPG data into the ECG editor you need to upgrade to PsPM 7.