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%.

The topic of this master's thesis is the analysis of ECG signals using wavelet transform. In the introductory chapters there is a brief description of heart anatomy, the emergence and spread of potentials, which evocating activities of myocardium. There is an overview of techniques used for ECG signals analysis and explanation of ECG curve diagnostic importance. Work also containts an ECG signal analysis common procedure explanation and different approaches brief overview. The main part of this work is an application detecting significant intervals in the ECG signal, developed in Matlab. In several chapters the detection procedure is described in more details and gave reasons for chosen methods. In the last chapter there is a preview of several signals as a result of developed application, together with evaluation of the tests carried out at the CSE database. Detector sensitivity was quantified over 99,10%.

The aim of diploma thesis named “QRS detection using wavelet transform” is to simplify and accelerate the work of doctors. This can be achieved by using application for QRS detection, which can use one of four proposed algorithms. Application shows basic informations about inserted electrocardiogram. Part of this program is a graphical window with displayed record and with coloured marks on detected QRS complexes. By another algorythm are marks color-coded due to accurancy percentil of every detected complex. This program is designed for a several hours record from Holter ECG monitoring.

The thesis provides an analysis of the ECG signal, focusing on the possible types of interference occuring in the signal. These types of interference are closely examined and their realization in the MATLAB programming environment is characterized. The SPIHT algorithm is introduced and the possibilities of its use are discussed. Futher, the application of this algorithm to the ECG signal containing individual types of interference generated by means of the designed interference generator is described and the impact of the SPIHT algorithm on the specific types of signal interference is analyzed. Finally, the obtained results are evaluated.

This thesis deals with method of blood vessels segmentation from retinal images acquired by fundus camera. There is explored possibility of using wavelet transform as fast outline segmentation. The thesis includes study problems of preprocessing input image and decomposition of image using 2D DWT. Furthermore there is explored possibility of parametrical images thresholding that ensue from application of 2D DWT. There are designed algorithms for cleaning off artifacts from rough vessel map of blood vessel structures. The realization of algorithm was solved in programming environment MATLAB. There was created a user control interface in graphic application GUIDE, for easy control of whole segmentation process. In conclusion of thesis is proceeded the discussion of segmentation results for images from DBME database and quantitative evaluation of results for DRIVE database images.

This thesis deals myopotential denoising of ECG signals with using wavelet transform. There was used wavelet denoising subsequently wiener wavelet filtering. In both cases were found the most suitable coeficients for the best denoising. It is meant mainly settings suitable parameters for ideal filtration setting value of threshold, number of decomposition level, selection of thresholding and type of filter. These parameters are tested on real signals. Denoising is realized in Matlab version R2009b.

This master's thesis deals with electrocardiogram and its distortion, which causes errors, due to which a doctor can make a wrong diagnosis. In the theoretical part electrocardiogram is briefly described, further theory of wavelet transform, introduction to the signal compression and some information about ECG delineation. Methods used for the signal evaluation and therefore determination of the distortion size are described in the main theoretical part. This thesis contains known methods and also a method of own design based on the ECG delineation. Some of the described methods are checked in the practical part. These methods are tested for two different compression algorithms, which cause various distortion. In the final part distortion was tested on the influence on setting the wrong diagnosis.

The work deals with the wavelet transformation, focusing on wavelet transforms with discrete time (DTWT). The practical part is focused on the implementation of redundant packet DTWT and its use in the filtration of ECG signals. The main part of the work is to design wiener filter that uses redundant packet DTWT, designed to eliminate interference myopotentials of ECG signals. The actual solution is implemented in Matlab. Testing is performed on the library CSE using noise model myopotentials used to noising original signals. For optimum parameters designed filter is used the genetic algorithm (GA). The work is carried out comparing the proposed filter redundant packet DTWT a variant of redundant dyadic DTWT.

This master's thesis deals with an analysis of principles of ECG signals detection and delineation. The theoretical part describes heart electrophysiology and electrocardiography basics. Next, the most important QRS detection and ECG delineation algorithms are introduced. Especially the wavelet transform methods are described. In the practical part proper delineation algoriythm was realized. It was tested on the standard CSE database, then it was modified on data of isolated rabbit heartsand the results are published in the conclusion.

The aim of the study was focused on problem of electrograms recording with using voltage sensitive dyes. Study of electrograms recording was extended with thema of optical detectors suitable for measurement and noise analysis. Elimination of noise and disturption was made in program Matlab; lowpass filtration and wavelet transformation was used. Application for electrograms analysis was developed in latter part of this work.