|
|
Deep learning based QRS delineator
Malina, Ondřej ; Ronzhina, Marina (referee) ; Smíšek, Radovan (advisor)
This thesis deals with the issue of automatic measurement of the duration of QRS complexes in ECG signals. Special emphasis is then placed on the possibility of automatic detection of QRS complexes while exciting cardiac tissue with a pacemaker. The content of this work is divided into four logical units, where the first part deals with the heart as an organ. It describes the origin and spread of excitement in the heart, its possible pathologies and their manifestations in ECG recording, it also deals with pacing and measuring ECG recording during simultaneous pacing. The second part of the thesis contains a brief introduction to the topic of machine and deep learning. The third part of the thesis contains a search of current approaches using methods based on deep learning to solve the detection of QRSd. The fourth part deals with the design and implementation of its own model of deep learning, able to detect the beginnings and ends of QRS complexes from ECG recordings. It describes the data preprocessing implemented in the MATLAB programming environment. The actual implementation of the model was performed in the Python using the PyTorch and NumPy moduls.
|
|
|
Muscle noise filtering in ECG signals
Fedorov, Vasilii ; Smíšek, Radovan (referee) ; Smital, Lukáš (advisor)
This work deals with problematic of muscle noise filtration in ECG signals. It contains theoretical and practical parts. In theoretical part we first mentioned a topicality of ECG scanning and filtration. Then we got acquainted with the origin of ECG, it's properties, and types of noises, that typically occurring there. Further different known methods of linear and non-linear techniques in EMG filtration were discussed. After we got acquainted with wavelet transform and its possibilities practical part was carried out in environment MATLAB 2020b®. Wiener wavelet filter was implemented and supplemented by a threshold adaptive function. Parameters were optimized with brute force method in reduced range. The evaluation of the filter took place on a CSE database, where the results were compared with the authors of other methods. In result the filter shows good filtration capabilities and stability.
|
|
|
Human activity classification
Müller, Jakub ; Smital, Lukáš (referee) ; Smíšek, Radovan (advisor)
This bachelor's thesis describes daily activity classification using accelerometric data. The first theoretical part summarizes the basics about daily activity and benefits that we get from monitoring it. In the next part of theory the principles of accelerometer inner workings are described. The last part of theory is dedicated to explaining the basics of neural networks and SVM. The aim of the practical part was to find a suitable dataset from a publicaly shared database, containing daily activity accelerometric data and also to collect our own data. Then performing classification using our own algorithm, optimizing it and finally evaluating the results.
|
|
|
Deep learning based QRS delineator
Malina, Ondřej ; Hejč, Jakub (referee) ; Smíšek, Radovan (advisor)
This thesis deals with the issue of automatic measurement of the duration of QRS complexes in ECG signals. Special emphasis is then placed on the possibility of automatic detection of QRS complexes while exciting cardiac tissue with a pacemaker. The content of this work is divided into four logical units, where the first part deals with the heart as an organ. It describes the origin and spread of excitement in the heart, its possible pathologies and their manifestations in ECG recording, it also deals with pacing and measuring ECG recording during simultaneous pacing. The second part of the thesis contains a brief introduction to the topic of machine and deep learning. The third part of the thesis contains a search of current approaches using methods based on deep learning to solve the detection of QRSd. The fourth part deals with the design and implementation of its own model of deep learning, able to detect the beginnings and ends of QRS complexes from ECG recordings. It describes the data preprocessing implemented in the MATLAB programming environment. The actual implementation of the model was performed in the Python using the PyTorch and NumPy moduls.
|
|
|
Identification of Abnormal ECG Segments Using Multiple-Instance Learning
Šťávová, Karolína ; Smíšek, Radovan (referee) ; Hejč, Jakub (advisor)
Heart arrhythmias are a very common heart disease whose incidence is rising. This thesis is focused on the detection of premature ventricular contractions from 12-lead ECG records by means of deep learning. The location of these arrhythmias (key instances) in the record was found using a technique based on Multiple-Instance Learning. In the theoretical part of the thesis, basic electrophysiology of the heart and deep learning with a focus on the convolutional neural networks are described. Afterward, a program was created using the Python programming language, which contains a model based on the InceptionTime architecture, using which classification of the signals into the selected classes was performed. Grad-CAM was implemented to find locations of the key instances in the ECGs. The evaluation of the arrhythmia detection quality was done using the F1 score and the results were discussed at the end of the thesis.
|
|
|
Atrial fibrillation model
Ředina, Richard ; Smíšek, Radovan (referee) ; Ronzhina, Marina (advisor)
The aim of this master thesis is to create a 3D electroanatomical model of a heart atria, which would be able to perform atrial fibrillation. To control the model, the differential equations of the FitzHugh-Nagumo model were chosen. These equations describe the change of voltage on the cell membrane. The equations have established parameters. The modification of them leads to changes in the behavior of the model. The simulations were performed in the COMSOL Multiphysics environment. In the first step, the simulations were performed on 2D models. Simulations of healthy heart, atrial flutter and atrial fibrillation were created. The acquired knowledge served as a basis for the creation of a 3D model on which atrial fibrillation was simulated on the basis of ectopic activity and reentry mechanism. Convincing results were obtained in accordance with the used literature. The advantages of computational modeling are its availability, zero ethical burden and the ability to simulate even rarer arrhythmias. The disadvantage of the procedure is the need to compromise between accuracy and computational complexity of simulations.
|
|
| |
|
| |
|
|
Detection Of P Wave During Second-Degree Atrioventricular Block In Ecg Signals
Maršánová, Lucie ; Němcová, Andrea ; Smíšek, Radovan
Automatic detection of P wave during the second-degree AV block is the main condition for automatic detection of this pathology. This work deals with developing of the algorithm for P wave detection. The algorithm is appropriate for ECG signals with AV block as well as signals with other rhythm types (it does not produce false positive P wave detections). For P wave detection, the phasor transform is applied and several innovative rules are created. These rules are based on knowledge of heart manifestation during both physiological and pathological heart function. The proposed algorithm consists of four parts – filtration, QRS complex detection, application of rules, and P wave detection. The accuracy of the P wave detection algorithm is 99.74 % for signals with AV block, and 99.82 % for signals without any pathologies.
|
|
|
Automatic detection of stress using biological signals
Votýpka, Tomáš ; Kozumplík, Jiří (referee) ; Smíšek, Radovan (advisor)
Bachelor's thesis is focused on stress detection. This thesis defines the concept of stress, analyzes the appropriate biological signals for stress detection, presents databases of biological signals, that were used for stress detection and mentions methods of automatic stress detection. Then, a stress detection program was implemented in the MATLAB software environment. A freely available database of non-EEG signals was used to implement the program. Models classifying stress were created using 4 machine learning methods for binary classification and 3 machine learning methods for classifying 4 psychical states. Efficiency of the classification was summarized in the conclusion of this thesis.
|
guest ::
RSS feed.