|
|
Analysis of stabilometric signals in frequency domain
Netopil, Ondřej ; Hejč, Jakub (referee) ; Kozumplík, Jiří (advisor)
This work deals with the metods frequency and time frequency analysis of stabilometric signal. In the introroduction is described theory about posturography and posturographic measurment. The work contains describtion of stabilometric parametrs in time domain (1D and 2D parametrs) and in frequency domain. The aim is create review of basic metods used to processing and preprocessing of stabilometric signals and comparing this methods . In work is realized ferquency analysis used Frourier transfrmation and Burg method and time-frequency analysis used Short time Frourier transformation and Wavelet transformation. One part of program is aimed on comparison of this methods.
|
|
|
Analysis of High-Frequency ECG and Mechano-Electric Coupling in Isolated Heart
Novotná, Petra ; Ronzhina, Marina (referee) ; Hejč, Jakub (advisor)
Tato magisterská práce se zabývá analýzou vysokofrekvenčních složek záznamu EKG z pohledu mechano-elektrické vazby u izolovaného králičího srdce. První částí této práce je literární rešerše na zadané téma zahrnující informace o vzniku a šíření akčního potenciálu na chemické i elektrické úrovní i mechano-elektrické zpětné vazbě. Dále práce obsahuje kapitolu zabývající se tématikou průměrování signálu jako techniky ke zvýšení poměru signál-šum při analýze vysokofrekvenčních složek. V praktické části práce jsou získané poznatky aplikovány na dlouhé záznamy EKG z izolovaných králičích srdcí. Zahrnut je popis perfuzního systému podle Neelyho a jeho užití při experimentech. V realizaci byla data z 15 izolovaných králičích srdcí podrobeno analýze zkoumající přítomnost reakce systému na změnu vstupních parametrů (afterloadu a preploadu) v případě tlakově objemových dat. Výsledky byly vztaženy ke stejně získaným hodnotám z HF ECG. Dohromady tvoří popis mechano-elektriké reakce srdce na hemodynamické podněty. Výsledky byly statistisky testovány a vyhodnoceny.
|
|
|
Automatic Delineation of Multi-lead ECG Signals
Veverka, Vojtěch ; Smital, Lukáš (referee) ; Hejč, Jakub (advisor)
This semester thesis is focused on automated measurement of ECG signal. The theoretical part describes the rise and options ECG signal. Furthermore, the issue is staged principal components analysis, whose output is used as input signal for seasons. They describe the basic methods used in measurement to ECG signal. The practical part is designed in measurement algorithm for ECG signal that has been tested on basic CSE database. The results are discussed in the conclusion.
|
|
| |
|
|
Analysis of Ventricular Repolarization Parameters
Abbrent, Jakub ; Novotná, Petra (referee) ; Hejč, Jakub (advisor)
This bachelor’s thesis deals with the analysis of ventricular repolarization parameters on experimental ECG records. In the beginning of the theoretical part there are included information about heart electrophysiology, fundamental principle of ECG and cellular basis of the T-wave formation. Next chapter is focused on methods used for the analysis of ventricular repolarization, especially spatial parameters including principal component analysis (PCA). Then, in the thesis, there is described the database of experimental ECG signals created from isolated rabbit hearts. In the practical part of this bachelor’s thesis, there are implemented spatial parameters on experimental ECG records. Implementation of algorithms is performed after initial data preparation. Then, there is performed analysis of relation between spatial and hemodynamic parameters and the relation is evaluated by statistical analysis.
|
|
|
Reference signals in intracranial EEG: implementation and analysis
Uher, Daniel ; Hejč, Jakub (referee) ; Ronzhina, Marina (advisor)
The idea of a artifact-free brain activity recording has been circling around the scientific world for a few decades. Parasitic phenomenons and unwanted components may significatntly complicate the analysis of intracranial electroencephalographic (iEEG) recordings. However, with the rise of modern technology, new methods for precise removal of noise artifacts started to emerge. Here we use the concept of virtual reference signals for the elimination of such unwanted components. In this work, the algorithms for reference signal estimation using common average based method as well as more recent methods based on independent component analysis (ICA) were realized and evaluated on a variety of iEEG data. It was found that the ICA-based algorithms allow obtaining more accurate estimation of the reference signal as compared to the average-based one. Finally, all the methods were implemented into a open-source Python package đť‘źđť‘’đť‘“đť‘ đť‘–đť‘”, which is publicly available, easy to install and ready to use.
|
|
| |
|
|
Advanced classification of cardiac arrhythmias in ECG
Sláma, Štěpán ; Hejč, Jakub (referee) ; Ronzhina, Marina (advisor)
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.
|
|
|
Optimization of a Deep Neural Network Label Encoding in a Multi-Label Problem.
Zaťko, Martin ; Novotná, Petra (referee) ; Hejč, Jakub (advisor)
The aim of the diploma thesis is to propose a method of deep learning for the classification of arrhythmias from ECG recordings and to compare the effect of coding its outputs on the overall quality of the model. A 1D convolutional neural network was selected and methods of label coding using one-hot coding, ordinal coding, the method using an autoencoder and the word embbeding method were tested and compared on it. The obtained results show that the use of the word embbeding method can increase the classification capacity of the proposed network.
|
|
|
Generative Adversial Network for Artificial ECG Generation
Šagát, Martin ; Ronzhina, Marina (referee) ; Hejč, Jakub (advisor)
The work deals with the generation of ECG signals using generative adversarial networks (GAN). It examines in detail the basics of artificial neural networks and the principles of their operation. It theoretically describes the use and operation and the most common types of failures of generative adversarial networks. In this work, a general procedure of signal preprocessing suitable for GAN training was derived, which was used to compile a database. In this work, a total of 3 different GAN models were designed and implemented. The results of the models were visually displayed and analyzed in detail. Finally, the work comments on the achieved results and suggests further research direction of methods dealing with the generation of ECG signals.
|
guest ::
