This paper deals with the classification of ECG measured from isolated rabbit hearts during the experiment with repeated ischemia. Classification features were calculated using the methods of heart rate variability analysis. The results were statistically evaluated. Heart rate variability parameters were calculated using Kubios HRV, other calculations were performed in MATLAB. Artificial neural network was created to classify the analyzed parameters to specific groups.

This master‘s thesis deals with the use of higher order cumulants for classification of cardiac cycles. Second-, third-, and fourth-order cumulants were calculated from ECG recorded in isolated rabbit hearts during experiments with repeated ischemia. Cumulants properties useful for the subsequent classification were verified on ECG segments from control and ischemic group. The results were statistically analyzed. Cumulants are then used as feature vectors for classification of ECG segments by means of artificial neural network.

The thesis describes design, algorithmization and realization of graphical application for recording EKG and PPG signal using LabJack UE9 tool in Matlab program, it also describes subsequent deposition of recorded signals and their processing, such as optional selection, cropping and filtering. Furthermore there are described types of filters, methods for detecting basic parameters of EKG and PPG signals and methods for detecting R waves and Systolic peaks. Based on detection of those parameters, algorithms for computing average heart rate and finding arrhythmias were designed and tested. Last part of the thesis includes an evaulation which compares values detected by designed algorithms with values from public database which includes reference annotation.

This work focuses on a theoretical explanation of heart rhythm disorders and the possibility of their automatic detection using deep learning networks. For the purposes of this work, a total of 6884 10-second ECG recordings with measured eight leads were used. Those recordings were divided into 5 groups according to heart rhythm into a group of records with atrial fibrillation, sinus rhythms, supraventricular rhythms, ventricular rhythms, and the last group consisted of the others records. Individual groups were unbalanced represented and more than 85 % of the total number of data are sinus rhythm group records. The used classification methods served effectively as a record detector of the largest group and the most effective of all was a procedure consisting of a 2D convolutional neural network into which data entered in the form of scalalograms (classification procedure number 3). It achieved results of precision of 91%, recall of 96% and F1-score values of 0.93. On the contrary, when classifying all groups at the same time, there were no such quality results for all groups. The most efficient procedure seems to be a variant composed of PCA on eight input signals with the gain of one output signal, which becomes the input of a 1D convolutional neural network (classification procedure number 5). This procedure achieved the following F1-score values: 1) group of records with atrial fibrillation 0.54, 2) group of sinus rhythms 0.91, 3) group of supraventricular rhythms 0.65, 4) group of ventricular rhythms 0.68, 5) others records 0.65.

Ztrátová komprese signálů EKG je užitečná a v současnosti stále se rozvíjející oblast. Stále se vyvíjí nové a nové kompresní algoritmy. V této oblasti ale chybí standardy pro hodnocení kvality signálu po kompresi. Existuje tedy sice mnoho různých kompresních algoritmů, které ale buď nelze objektivně porovnat vůbec, nebo jen zhruba. V oblasti komprese navíc nikde není popsáno, zda mají na výkon kompresních algoritmů vliv patologie, popřípadě jaký. Tato dizertační práce poskytuje přehled všech nalezených metod pro hodnocení kvality signálů EKG po kompresi. Navíc bylo vytvořeno 10 nových metod. V rámci práce byla provedena analýza všech těchto metod a na základě jejích výsledků bylo doporučeno 12 metod vhodných pro hodnocení kvality signálu EKG po kompresi. Také je zde představen nový kompresní algoritmus „Single-Cycle Fractal-Based (SCyF)“. Algoritmus SCyF je inspirován metodou založenou na fraktálech a využívá jednoho cyklu signálu EKG jako domény. Algoritmus SCyF byl testován na čtyřech různých databázích, přičemž kvalita signálů po kompresi byla vyhodnocena 12 doporučenými metodami. Výsledky byly porovnány s velmi populárním kompresním algoritmem založeným na vlnkové transformaci, který využívá metodu „Set Partitioning in Hierarchical Trees (SPIHT)“. Postup testování zároveň slouží jako příklad, jak by měl vypadat standard hodnocení výkonu kompresních algoritmů. Dále bylo statisticky prokázáno, že existuje rozdíl mezi kompresí fyziologických a patologických signálů. Patologické signály byly komprimovány s nižší efektivitou a kvalitou než signály fyziologické.

The thesis deals with possibilities of using wavelet transform in the field of surface electrocardiogram (ECG) signals denoising and ECG signals measuring. Several algorithms have been used to detect and estimate T-wave alternans (TWA), such as spectral method (SM), Poincaré Mapping (PM) or correlation method (CM). T-wave alternans, also called repolarization alternans, is a phenomenon appearing in the electrocardiogram as a consistent fluctuation in the repolarization morphology on every-other-beat basis. Electrical TWA has been recognized as a marker of electrical instability, and has been shown to be related with patients at increased risk for ventricular arrhytmias. Presence of TWA has been reported in a wide range of clinical and experimental situations including long QT syndrome, myocardial infarction, angina pectoris, acute ischemia, etc. Projected methods of detection TWA are realized in Matlab software, and they are experimentally verified on real ECG signals from the European ST-T Database.

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 thesis deals with the biological data averaging applied to a periodical and repetitive signal, specifically to an ECG signals. There were used signals from MIT-BIH Arrhythmia database and ÚBMI database. Averaging was realized with constant, floating and exponential Windows, where was used the method of addition of the filtered residue. This method is intended to capture the slow variations from the input to the output signal. The outcomes of these methods can be used as a basis for further work, or function as an example of principled methods. Methods and its outcomes were created in Matlab.

The resume of this master´s thesis is to introduce reader into principal component analysis (PCA), namely, the use of PCA for analysis of ECG. This method allows to reduce quantity of the data without loss of useful information. That is why PCA is widespread for preprocessing of the data for further classification, which this thesis also deals. Data available at the Department of Biomedical Engineering at the University of Technology in Brno were used in this work. All the methods were realized using Matlab.

The aim of this work is to study and describe the principles and method of delineation of ECG signals. Learn and describe about method of cluster analysis. In this work was created and described three different methods of delineations of ECG signals. Created algorithms were tested on complete CSE database. With cluster analysis were combine created methods. The obtained results from realized methods and combined method were compared with others known methods. At the end of this work is evaluate efficiency of detection of combined method.