Causal Probabilistic Network and Power Spectral Estimation Used in Sleep Stage Classification

1997 ◽  
Vol 36 (04/05) ◽  
pp. 41-46
Author(s):  
A. Kjaer ◽  
W. Jensen ◽  
T. Dyrby ◽  
L. Andreasen ◽  
J. Andersen ◽  
...  
Keyword(s):  
Sleep Stage ◽  
Spectral Estimation ◽  
Interrater Agreement ◽  
Probabilistic Networks ◽  
Probabilistic Network ◽  
Spectral Bands ◽  
Power Spectral ◽  
Stage Classification ◽  
Sleep Stage Classification ◽  
The Brain

Abstract.A new method for sleep-stage classification using a causal probabilistic network as automatic classifier has been implemented and validated. The system uses features from the primary sleep signals from the brain (EEG) and the eyes (AOG) as input. From the EEG, features are derived containing spectral information which is used to classify power in the classical spectral bands, sleep spindles and K-complexes. From AOG, information on rapid eye movements is derived. Features are extracted every 2 seconds. The CPN-based sleep classifier was implemented using the HUGIN system, an application tool to handle causal probabilistic networks. The results obtained using different training approaches show agreements ranging from 68.7 to 70.7% between the system and the two experts when a pooled agreement is computed over the six subjects. As a comparison, the interrater agreement between the two experts was found to be 71.4%, measured also over the six subjects.

The Theory of Spectral Estimation of Signals and Generalized Kravchenko-Kotel'nikov-Levitan Theorems

PIERS Online ◽  
2009 ◽  
Vol 5 (4) ◽  
pp. 373-376
Author(s):  
Victor Filippovich Kravchenko ◽  
Dmitry V. Churikov
Keyword(s):  
Spectral Estimation

Signal Processing Enhanced Fiber Vibration Sensors

2013 ◽  
Vol 543 ◽  
pp. 302-305
Author(s):  
Daniele Tosi ◽  
Massimo Olivero ◽  
Alberto Vallan ◽  
Guido Perrone
Keyword(s):  
Signal Processing ◽  
Optical Fibers ◽  
Adaptive Filters ◽  
Spectral Estimation ◽  
Cost Effective ◽  
Fiber Sensors ◽  
High Frequencies ◽  
Vibration Sensors ◽  
Signal Processing Algorithms ◽  
Plastic Optical Fibers

The paper analyzes the feasibility of cost-effective fiber sensors for the measurement of small vibrations, from low to medium-high frequencies, in which the complexity of the measurement is moved from expensive optics to cheap electronics without losing too much performance thanks to signal processing algorithms. Two optical approaches are considered: Bragg gratings in standard telecom fibers, which represent the most common type of commercial fiber sensors, and specifically developed sensors made with plastic optical fibers. In both cases, to keep the overall cost low, vibrations are converted into variations of the light intensity, although this makes the received signal more sensitive to noise. Then, adaptive filters and advanced spectral estimation techniques are used to mitigate noise and improve the sensitivity. Preliminary results have demonstrated that the combined effect of these techniques can yield to a signal-to-noise improvement of about 30 dB, bringing the proposed approaches to the level of the most performing sensors for the measurement of vibrations.

Sign in / Sign up

Export Citation Format

Share Document