Sponsored by
[ Events ]

Activity Search
Sort out
NCTS Short Course on Data Science
9:10 - 13:00, May 10, 2018 (Thursday)
Room 440, Astronomy-Mathematics Building, NTU
(台灣大學天文數學館 440室)
Automatic Sleep Stage Classification Based on Diffusion Maps
Gi-Ren Liu (National Cheng Kung University)


In this course, we will use the public Sleep-EDF database [1] to demonstrate how the sleep stages can be classified by two electroencephalogram (EEG) signals and diffusion maps. This course aims to answer the following problems:
(a) How to extract features from the EEG data based on the time-frequency analysis?
(b) After extracting features for each 30s epoch, how to measure their distance using the concept of Markov chain?
(c) How to assign a new coordinate for each feature vector based on the eigenvectors of the distance matrix (or the transition matrix) obtained from (B)?
(d) How to visualize the geometric structure of different sleep stages?
[1] Link of the Sleep-EDF database: https://www.physionet.org/physiobank/database/sleep-edfx/
[2] Hau-tieng Wu, Ronen Talmon, and Yu-Lun Lo. "Assess sleep stage by modern signal processing techniques." IEEE Transactions on Biomedical Engineering 62.4 (2015): 1159-1168.
[3] Gi-Ren Liu, Yu-Lun Lo, Yuan-Chung Sheu and Hau-Tieng Wu. "Diffuse to fuse EEG spectra - intrinsic geometry of sleep dynamics for classification." arXiv preprint arXiv:1803.01710 (2018).


back to list  
 (C) 2018 National Center for Theoretical Sciences