[Eeglablist] Question Welchparse function

[Arnaud Delorme]

Dear Diede,

> I am working with the new ROIconnect toolbox in eeglab (
> https://urldefense.com/v3/__https://github.com/arnodelorme/roiconnect__;!!Mih3wA!C1n9l2EneLA7LAvTmlHLLJZnIhXJdy0aq0AoIp6sLoOlvCeShy5mDgqTf_pcb7xMm01QXwD4HVyez2HuyV8Z$  ) and now trying to calculate the
> ROI activity where the Welch's method is used to calculate the power.
> I am getting the error in the welchparse function: The length of the
> segments cannot be greater than the length of the input signal.

ROIConnect works on longer data epochs for now. 

> It is not possible for me to increase the number of time points in my
> epochs (188 timepoints in total, while I would need 500 timepoints for the
> script to work).
> Would it be possible to adapt the length of the segment?

Everything is possible. I will discuss it with the team.

> Or is there another solution to make the script work? We thought about
> merging epochs together or multiplying the data points to increase the
> length of the input signal, would that be an option?

The epoch length is relatively arbitrary, although there might be issues with data stability for shorter epochs.

ROIconnect is planned to be released in 2023. It is difficult to use now because
- There is almost no documentation
- There are no stats across subjects

We are planning to remedy that, but it takes time.

This is the reference article recently submitted
https://urldefense.com/v3/__https://www.biorxiv.org/content/10.1101/2022.10.05.510753v2__;!!Mih3wA!Cn7eXjRpSOzSGCcVHdohmvxVkLE0Ve4SCBYvOhVK6p2LhSm96copJP_jY4Yw6iXu0uV0go_sOx7jJ4Dzavgf3W0a$  

Arno