How to use your pipeline in a CTF form data

[wobunengwang]

Dear author, thank you very much for your elaborate codes and guidelines.
(1) I used to use fieldtrip to preprocess and reconstruct my CTF form MEG data. But i found that your codes were all preprocessed with osl-core toolbox. I am very eager to know how to transform a fieldtrip preprocessing pipeline into a osl-based pipeline?
(2) I have tried many solutions and finally could extract time-series matrix from fieldtrip reconstructed data, could you tell me whether the different preprocessing and reconstruction steps could be suitable for your HMM models?

[AJQuinn]

Hello - thanks for your interest in OSL.

1. OSL uses a lot of SPM functions internally, so it should work with any data set that can be loaded into an SPM object.
   This tutorial gives an example of how to load a CTF .ds directory into an SPM object (often called 'D'). Once you have the D-object, then the other preprocessing tools should work as normal. Note that you might have to change the chantypes in some function calls to match the CTF channel types. This preprocessing tutorial then gives an overview of which preprocessing tools can be used with OSL. This would be a good place to start if translating an existing analysis from fieldtrip.

2. We have a couple of example/recommended pipelines for HMM analyses. This paper (Quinn et al 2018) gives a detailed overview of preprocessing for HMM with accompanying code in OSL. This example script runs the full preprocessing and HMM analysis pipeline from this paper (Vidaurre et al 2018).

I'm sure you could do a similar preprocessing job in another toolbox such as fieldtrip, these papers and example scripts might be a good starting point