Standard Acquisition(?)

[smoia]

Goes together with #23 .

One thing that we can do to facilitate the process for the user is to think about "standard cases" that could act as a blueprint for the acquisition of physiological data.

E.g.:

• Save the trigger in channel 1
• Use x sampling frequency
• Use y unit of measure
• ...

Keeping in mind that we want to be as flexible as possible, we can offer heuristics for particular cases that would work almost automatically. If we want to offer them, what would be good examples of "standard cases"?

[smoia]

Issue #36 is of interest!

[tsalo]

Is this a good place to ask about how people typically set up their triggers and recordings? I'm curious to know what the distribution of setups looks like. Perhaps a poll on Twitter would also help?

[smoia]

It definitely is! I'd ask @RayStick and @BrightMG @CesarCaballeroGaudes to answer because they have experience and very specific opinions about it!

[smoia]

What I can say on our side is that, using Biopac and AcqKnowledge, we have a feedback from the MRI that we store in a channel. I think this is a direct feedback, but @CesarCaballeroGaudes knows better the exact setup we use. I'm quite positive about the fact that we don't use psychopy or any other program to send the MRI trigger to the Biopac.

We also collect and store all the channels in Volts. I know that @RayStick set up their LabChart to automatically convert some of their unit of measure (e.g. gases in mmHg).

[RayStick]

I guess this topics relates to this documentation that we have already, but maybe it would be good to eventually expand on this documentation, to add more specific recommendations for experimental set-up. 



Here are my initial thoughts:



Our set-up (@BrightMG's lab at Northwestern) involves using AD Instrument’s Power Lab which reads in all of our inputs as analog signals (MR Pulse, CO2, O2, Pulse, Respiration). All of the peripheral recording devices connect straight to the PowerLab, displayed in volts, but the CO2 and O2 inputs go via a gas analyzer which in turn connects to the power lab. The transducers in this gas analyzer often come set-up so that the voltage outputs are linearly proportional to the concentrations of CO2 and O2 sampled. So in the labchart software you can set CO2 and O2 signals to voltage or percentage (and convert them later) or ask them to display in pressure units (mmHg) whilst recording. In our lab, we favor the latter - displaying in pressure units. And also, we check the calibration of our system now and again, and re-calibrate, as sometimes there seems to be an offset that accumulates over time (not sure why though). So the AD Instruments software and hardware allows you to re-calibrate if you have two gas samples with known CO2 and O2 concentrations - one can be normal air, and for the other one we use a small calibration gas cylinder with set amount of CO2 and O2 concentrations in.

For the MR trigger pulse - this connects to directly to our PowerLab to record volume triggers. It has a splitter function so the same signal can be sent to a PC that is displaying stimuli, so that you can cue the onset of your stimuli with the onset of the scan starting.

As to “standard cases” I’d need to think (and maybe discuss with the lab) about clear recommendations and work out which things we do are specific to us, and which things would relate to more general recommendations. But things I can say now:


• Saving the trigger as channel 1 - definitely good idea
• Make one continuous physiological recording, leaving at least 30 seconds between scans