aparek/mcsleep

Multichannel Sleep Spindle Detector for sleep EEG

McSleep: Multichannel sleep spindle detector for sleep EEG

This program detects sleep spindles simultaneously across all channels of a multichannel human sleep EEG using a multichannel transient separation algorithm. Please see the publication below for further details.

To run a quick demo, run the file demo.m To utilize parallel detection on an excerpt of EEG (30 minutes) use runSpindleDetection.m For the MASS database, use the file Mass_parallelSpindleDetection.m For a demo of paramter tuning method used in the publication, run parameterTuning.m

Permanent Contact: [email protected] [email protected]

For questions regarding the code, please email at the addresses above.

Please cite as: Multichannel Sleep Spindle Detection using Sparse Low-Rank Optimization A. Parekh, I. W. Selesnick, R. S. Osorio, A. W. Varga, D. M. Rapoport and I. Ayappa Journal of Neuroscience Methods, Vol. 288, pp. 1-16, Aug. 2017. (https://doi.org/10.1016/j.jneumeth.2017.06.004)

mcsleep's Issues

hyper parameter selection scrip

Dear Ankit,

First of all thank you for providing the code of your approach, that is of great value to reproduce your work.

Would you accept to add to the repository the script that enables to select the hyper parameters (lambda_2, c) on Mass dataset as explained in the paper ? This would allow me to reproduce your results on the same setting.

Thank you in advance,
Best regards,

Stanislas