|
|
Analysis of sleep EEG signal
Ježek, Martin ; Kozumplík, Jiří (referee) ; Rozman, Jiří (advisor)
Cílem této práce byl vývoj programu pro automatickou detekci arousalu v signálu spánkového EEG s použitím metod časově-frekvenční analýzy. Předmětem studie bylo 13 celonočních polysomnografických nahrávek (čtyři svody EEG, EMG, EKG a EOG), tj. celkově více než 100 hodin záznamu. Jednalo se o část dat z dřívějších výzkumných prací expertní lékařky v problematice spánku Dr. Emilie Sforzy, Ženeva, Švýcarsko, která rovněž poskytla základní hodnocení těchto dat. V záznamech bylo celkem označeno 1551 arousal událostí. Pro usnadnění výběru konkrétní metody časově-frekvenční analýzy byla následně vytvořena sada nástrojů pro vizualizaci jednotlivých signálů a jejich různých časově-frekvenčních vyjádření. S ohledem na závěry vizuální analýzy, charakter signálu EEG a efektivitu výpočetních metod byla pro analýzu vybrána waveletová transformace s mateřskou vlnkou Daubechies řádu 6. Jednotlivé svody EEG byly dekomponovány do šesti frekvenčních pásem. Z takto odvozených signálů a signálu EMG byly následně stanoveny ukazatele možné přítomnosti události arousalu. Tyto ukazatele byly dále váhovány lineárním klasifikátorem, jehož hodnoty vah byly optimalizovány pomocí genetického algoritmu. Na základě hodnoty lineárního klasifikátoru bylo rozhodnuto o přítomnosti události arousalu v daném svodě EEG – arousal byl detekován, jestliže hodnota klasifikátoru překročila danou mez na dobu více než 3 a méně než 30 vteřin. V celém záznamu pak byl arousal označen, byl-li detekován alespoň v jednom ze svodů EEG. Následně byly odvozeny míry senzitivity a selektivity detekce, jež byly rovněž základem pro stanovení fitness funkce genetického algoritmu. Pro učení genetického algoritmu byly vybrány první čtyři záznamy. Na základě takto optimalizovaných vah vznikl program pro automatickou detekci, který na celém souboru 13 záznamů dosáhl ve srovnání s expertním hodnocením míry senzitivity 76,09%, selektivity 53,26% a specificity 97,66%.
|
|
|
EEG Signal Analysis during the Stroop Test
Tolaszová, Eva ; Roman,, Robert (referee) ; Sekora, Jiří (advisor)
Master’s thesis deals with the measurement of biological signals for the effect of psychological burden. To monitor this effect was elected Stroop test, which is in the psychology used to detect disorders of attention and concentration. EEG and ECG signals during Stroop test were obtained using the EEG recording systém, in the context of research evoked potentials. As a part of the work it has been designed custom application for analyzing and interpreting data and statistical analysis by t-test.
|
|
|
Diagnosing Parkinson's disease from analysis of speech recording
Vymlátil, Petr ; Trzos, Michal (referee) ; Lněnička, Jakub (advisor)
This thesis is focused on diagnosing Parkinson’s disease from analysis of speech recording. Introduction of this work deals with description of voice production mechanism, it’s basic qualities and influence of hypokinetic dysarthria on speech. In next chapter, there is described voice signal and some methods of it’s preprocessing. Next part continues dealing with description of chosen individual symptoms, which are needed for PD diagnosing, followed by definition of chosen reduction methods and classifiers. There is a comparison of classify succes of naive bayes classifier, depending on chosen reduction method in last chapter of this work.
|
|
|
Tool for visualization and listening of vibration signals
Slavov, Jan ; Mancl, Vlastimil (referee) ; Čala, Martin (advisor)
This bachelor thesis deals with the development of a tool for visualization and listening of vibrational signals. The aim of the work is to find out the possibilities of the LabVIEW development environment in the area for which the application is intended, to further evaluate similar programs offered on the market and to create an application that will allow the user basic signal processing. The application also allows you to work with the myDAQ device and listen to signals through speakers, during which it will display the current playback position.
|
|
|
Order Analysis
Honc, Lukáš ; Pikula, Stanislav (referee) ; Beneš, Petr (advisor)
This master's thesis deals with order analysis. The first part of the thesis describes common methods for order analysis and methods for processing tacho and vibration signal. The second part contains a brief description of some open source tools for order analysis focusing on Sound and Vibration Measurement Suite (SVMS) package for LabVIEW by National Instruments company. The main purpose of the thesis has been designing and realization of own library for order analysis as a plugin for development tool LabVIEW. In the library, there are implemented methods for order analysis including basic functions for processing vibration and tacho signal. Their brief description with manual for its usage is in the third part of the thesis. In the last chapter, implemented functions are compared with those, which are implemented in SVMS by National Instruments, using both simulated and real data.
|
|
|
Simulator for Passive Multi-Static Radar using WiFi/WiMAX
Sládek, Ondřej ; Kubíček, Michal (referee) ; Šebesta, Jiří (advisor)
This master’s thesis deals with the concept of passive multistatic radar. The radar system exploits WiFi or WiMAX transmitters as the source of radiolocation signal. The transmitters are considered non-cooperative. The master’s thesis evaluates limitations arising from utilization of WiFi or WiMAX signals. A Matlab simulator was created as a part of the thesis, which was used to verify the basic idea behind this concept. Based on the results of real-life simulations, conclusions are suggested towards a possible application of WiFi/WiMAX radar.
|
|
| |
|
|
Design of testing device for piezocomposite sensors
Zlámal, Radim ; Chalupa, Jan (referee) ; Kšica, Filip (advisor)
Presented bachelor’s thesis describes the design and the following assembly of a device used for testing MFC sensors. In the first part, fundamental theoretical findings regarding main properties of the sensors are listed, including their description and comparison with other piezoelectric materials. Available methods for measuring a response from the sensor are mentioned sequentially by means of applied stress (tension/compression, bending, torsion). The second part focuses on realization of the device based on the chosen physical principle for loading the tested specimen, to which the sensor is firmly attached. Construction of the device is designed and adjusted to meet the input requirements and boundary conditions for bending. Both hardware and software are selected to simulate the specific form of mechanical loading. The last part involvesstatic measurements of deformations on the specimen using a charge amplifier, as well as modelling of various dynamic cycles sensed by utilizing a load resistor. Voltage signal from the sensor is converted to strain, which is further transformed into deflection for measurement verification, and finally, the functionality of the entire device is evaluated.
|
|
|
Train Identification System at Railway Switches And Crossings Using Advanced Machine Learning Methods
Krč, Rostislav ; Vorel,, Jan (referee) ; Plášek, Otto (referee) ; Podroužek, Jan (advisor)
This doctoral thesis elaborates possibilities of automatic train type identification in railway S&C using accelerometer data. Current state-of-the-art was considered, including requirements stated by research projects such as S-Code, In2Track or Turnout 4.0. Conducted experiments considered different architectures of artificial neural networks (ANN) and statistically evaluated multiple use case scenarios. The resulting accuracy reached up to 89.2% for convolutional neural network (CNN), which was selected as a suitable baseline architecture for further experiments. High generalization capability was observed as models trained on data from one location were able to classify locomotive types in the other location. Further experiments evaluated the effect of signal filtering and denoising. Evaluation of allocated memory and processing time for pre-trained models proved feasibility for in-situ application with regard to hardware restrictions. Due to a limited amount of available accelerometer data, distribution grid power demand data were utilized for further refinement of the proposed CNN architecture. Deep multi-layer architecture with regularization techniques such as dropout or batch normalization provides state-of-the-art performance for time series classification problems. Class activation mapping (CAM) allowed an explanation of decisions made by the neural network. Presented results proved that train type identification directly in the S&C is possible. The CNN was selected as optimal architecture for this task due to high classification accuracy, automatic filtration, and pattern recognition capabilities, allowing for the incorporation of the end-to-end learning strategy. Moreover, direct on-site application of pre-trained models is feasible with respect to limitations of in-situ hardware. This thesis contributes to understanding the train type identification problem and provides a solid theoretical background for future research.
|
|
|
Tool for visualization and listening of vibration signals
Slavov, Jan ; Mancl, Vlastimil (referee) ; Čala, Martin (advisor)
This bachelor thesis deals with the development of a tool for visualization and listening of vibrational signals. The aim of the work is to find out the possibilities of the LabVIEW development environment in the area for which the application is intended, to further evaluate similar programs offered on the market and to create an application that will allow the user basic signal processing. The application also allows you to work with the myDAQ device and listen to signals through speakers, during which it will display the current playback position.
|
guest ::
