Open lukemorse opened 5 years ago
I agree with that this is a quite manatory feature, which most people would need. However i'm not quite sure how we can implement this feature properly.
One potentional implementation:
I am also not quite sure yet how to implement the calibration method. Anyone some suggestions for algorithms which can be of use?
I found this in the forum: http://forum.choosemuse.com/t/favorite-algorithms-for-eeg-features-analysis/487/4
Calculating the mean and standard deviation is easy enough, but he says he feeds it into a leaky-integrator. I looked it up and it seems a bit outside of my skill set.
On Sat, Jan 26, 2019 at 9:35 PM Sacation notifications@github.com wrote:
I agree with that this is a quite manatory feature, which most people would need. However i'm not quite sure how we can implement this feature properly.
One potentional implementation:
- Add a Calibrate(seconds) method to the Muse class.
- Add a signaladdress for Eeg_calibrated (or per frequency range, like SignalAddress.Delta_Calibrated, enz.) This allows subscribing to the calibrated signals.
I am also not quite sure yet how to implement the calibration method. Anyone some suggestions for algorithms which can be of use?
— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/LuukEbenau/SacaDev.Muse/issues/6#issuecomment-457835907, or mute the thread https://github.com/notifications/unsubscribe-auth/AQ15Ma3sGiFW1GuLisp3iVBecK9qyXIuks5vHGeVgaJpZM4aUDSU .
It's also not really inside my own toolset, but perhaps I (or someone else) can make an initial suboptimal version of an calibration method, which can be improved by someone more experienced.
Is your feature request related to a problem? Please describe. No.
Describe the solution you'd like It would be great to have a calibration feature similar to what Muse has done with their own app.