|
|
Sleep EEG analysis
Vávrová, Eva ; Kozumplík, Jiří (referee) ; Ronzhina, Marina (advisor)
The bachelor´s thesis is focused on analysis of sleep electroencephalograms based on extraction of chosen parameters in time and frequency domain. The parameters are acquired from segments of EEG signals coincident with sleep stages. The parameters used for automatic detection of sleep stages are selected according to statistical analysis. The program with a graphical user interface for selection, display and analysis EEG was created using Matlab.
|
|
|
Sleep stages classification
Nováková, Kateřina ; Ronzhina, Marina (referee) ; Potočňák, Tomáš (advisor)
This work deals with the basic description of polysomnography, sleep morphology and sleep stages. Furtherly, some methods to process electroencephalographic signals are mentioned. Those processing methods are mainly focused on sleep stage classification. The practical part deals with the realization of three classification algorithms using artificial neural networks and verifying the functionality of these methods. All algorithms are designed in Matlab. Feature vectors for individual methods are obtained using energy values, Welch's spectral analysis and Hilbert-Huang Transform. For classification three types of artificial neural networks were used - layer recurrent network, feedforward network and pattern recognition network. On the basis of feature vectors, the sleep is divided into three stages - wakefulness (W), sleep without rapid eye movements (NREM) and sleep with rapid eye movements (REM).
|
|
|
Sleep stages classification
Nováková, Kateřina ; Ronzhina, Marina (referee) ; Potočňák, Tomáš (advisor)
This work deals with the basic description of polysomnography, sleep morphology and sleep stages. Furtherly, some methods to process electroencephalographic signals are mentioned. Those processing methods are mainly focused on sleep stage classification. The practical part deals with the realization of three classification algorithms using artificial neural networks and verifying the functionality of these methods. All algorithms are designed in Matlab. Feature vectors for individual methods are obtained using energy values, Welch's spectral analysis and Hilbert-Huang Transform. For classification three types of artificial neural networks were used - layer recurrent network, feedforward network and pattern recognition network. On the basis of feature vectors, the sleep is divided into three stages - wakefulness (W), sleep without rapid eye movements (NREM) and sleep with rapid eye movements (REM).
|
|
|
Sleep EEG analysis
Vávrová, Eva ; Kozumplík, Jiří (referee) ; Ronzhina, Marina (advisor)
The bachelor´s thesis is focused on analysis of sleep electroencephalograms based on extraction of chosen parameters in time and frequency domain. The parameters are acquired from segments of EEG signals coincident with sleep stages. The parameters used for automatic detection of sleep stages are selected according to statistical analysis. The program with a graphical user interface for selection, display and analysis EEG was created using Matlab.
|
guest ::
