[Eeglablist] Code with tutorial to find optimal cut-off parameter of ASR

[Velu Prabhakar Kumaravel]

Hello EEGLABers,

You might all know of the Artifacts Subspace Reconstruction (ASR)
algorithm, an efficient algorithm especially to deal with non-stereotypical
artifacts (prevalent in mobile EEG and/or developmental EEG). ASR has a
user-defined parameter (ASR cut-off, k), which is set to 20, by default.
This value should work in most cases, however, it would be valuable to find
the optimal value for your datasets.

In our Newborns EEG Artifacts Removal (NEAR
<https://urldefense.proofpoint.com/v2/url?u=https-3A__www.sciencedirect.com_science_article_pii_S1878929322000123&d=DwIBaQ&c=-35OiAkTchMrZOngvJPOeA&r=kB5f6DjXkuOQpM1bq5OFA9kKiQyNm1p6x6e36h3EglE&m=j1r5OrQy6HMNTNYtzCAuvh_-7ZoAcHj0JoPEHnmvWsG-tfFHZVxF5a1OUYVuDzGk&s=UqF8-uu_HoSLuhsrW2mpvLy3vaNc0wABmPRsjtrH2ag&e= >)
pipeline, we proposed a systematic grid search approach to find the optimal
k. I wrote a short tutorial which can be found in this link
<https://urldefense.proofpoint.com/v2/url?u=https-3A__github.com_vpKumaravel_vpKumaravel.github.io_wiki_How-2Dto-2Dfind-2Dthe-2Doptimal-2Dhyperparameter-2Dfor-2DArtifacts-2DSubspace-2DReconstruction-2D-28ASR-29-2Dalgorithm-2Dto-2Dclean-2DEEG-2Dartifacts-253F&d=DwIBaQ&c=-35OiAkTchMrZOngvJPOeA&r=kB5f6DjXkuOQpM1bq5OFA9kKiQyNm1p6x6e36h3EglE&m=j1r5OrQy6HMNTNYtzCAuvh_-7ZoAcHj0JoPEHnmvWsG-tfFHZVxF5a1OUYVuDzGk&s=5HGabx27vchRkelZqWRzWDTatD1X1sWTMWfFwDZm-F0&e= >along
with the EEGLAB compatible MATLAB code.

Hope this is useful and let me know your feedback in case you tried it out.

Best regards,

Velu Prabhakar Kumaravel, Ph.D. Student
Center for Mind/Brain Sciences,
University of Trento, Italy