Is there a function to detect k-complexes for N2 detection?

[PhD-GOAT]

asking because there is a spindle detection implementation to detect N2 sleep, but not a k-complex! I wonder if there is a development plan.

[raphaelvallat]

Hi @PhD-GOAT,

You should be able to detect k-complexes with the yasa.sw_detect function: you just need to make sure that you limit the detection to N2 sleep.

[PhD-GOAT]

Good. But what if some PSGs don't detect the K-complex and give an error? Of course, we're running the whole time.

[raphaelvallat]

Sorry I'm not sure I understand the question. Have you seen errors when running the SW detection?

[PhD-GOAT]

Yes. I mean, if I apply a function that detects K complex for the whole time, it fails to detect in a few psg.

[PhD-GOAT]

And there seems to be a fatal error in detecting the kcomplex with the sw function. The kcomplex should have a positive peak followed by a negative peak, but the sw function starts with a negative peak. So I tried amp setting -, but it didn't work.

[raphaelvallat]

YASA slow-wave detection expects a large negative peak followed by a small positive peak, not the other way around. In that case you might be better off not using YASA and trying to find another detection method for K-complexes. Manual scoring might be the most accurate method.