This thesis deals with the detection of QRS complexes from electrocardiograms using time-frequency analysis. Detection procedures are based on wavelet and Stockwell transform. The theoretical part describes the basics of electrocardiography, then introduces common approaches to time-frequency analysis, such as short-time Fourier transform (STFT), wavelet transform and Stockwell transform. These algorithms were tested on a set of electrograms from the MIT-BIH and CSE-MO1 arrhythmia database. For the CSE database worked best the method based on the wavelet transform with the filter bank Symlet4, with the resulting value of sensitivity 100 % and positive predictivity 99.86%. For the MIT database had the best performance the detector using the Stockwell transform with values of sensitivity 99.54% and positive predictivity 99.68%. The results were compared with the values of other authors mentioned in the text.

Tolkningen av EKG är en viktig metod vid diagnostisering av onormala hjärttillstånd och kan användas i förebyggande syfte att upptäcka tidigare okända hjärtproblem. Att enkelt kunna mäta sitt EKG och få det analyserat och presenterat på ett pedagogiskt sätt utan att behöva rådfråga en läkare är något det finns ett konsumentbehov av. Denna rapport beskriver hur en EKG-signal behandlas med olika algoritmer och metoder i syfte att detektera hjärtslag och dess olika parametrar. Denna information används till att klassificera varje hjärtslag för sig och därmed avgöra om användaren har en normal eller onormal hjärtfunktion. För att nå dit har en mjukvaruprototyp utvecklats där algoritmerna implementerats. En enkätundersökning gjordes i syfte att undersöka hur utdata från mjukvaruprototypen skulle presenteras för en vanlig användare utan medicinsk utbildning. Sju filer med EKG-signaler från MIT-BIH Arrhythmia Database användes för testning av mjukvaruprototypen. Resultatet visade att prototypen kunde detektera en rad olika hjärtfel som låg till grund vid fastställning om hjärtat slog normalt eller onormalt. Resultatet presenterades på en mobilapp baserad på enkätundersökningen. / The interpretation of the ECG is an important method in the diagnosis of abnormal heart conditions and can be used proactively to discover previ-ously unknown heart problems. Being able to easily measure the ECG and get it analyzed and presented in a clear manner without having to consult a doctor is improtant to satisfy consumer needs. This report describes how an ECG signal is treated with different algo-rithms and methods to detect the heartbeat and its various parameters. This information is used to classify each heartbeat separately and thus determine whether the user has a normal or abnormal cardiac function. To achieve this a software prototype was developed in which the algorithms were implemented. A questionnaire survey was done in order to examine how the output of the software prototype should be presented for a user with no medical training. Seven ECG files from MIT-BIH Arrhythmia database were used for validation of the algorithms. The developed algorithms could detect of if any abnormality of heart function occurred and informed the users to consult a physician. The presentation of the heart function was based on the result from the questioner.

This master’s thesis deals with the detection of QRS complex detection using zero crossing counts. QRS detection is an important part of the analysis of ECG signal. From the point of determining the R wave detection is based on the other waves and intervals necessary for the diagnosis of heart. This method is very effective even for very noisy signals. Implementation of the method in Matlab, and the success of detection is tested on the CSE and MIT-BIH database. The optimization algorithm is an optional value for the detector.

<p dir="ltr"> The widespread adoption of Convolutional Neural Networks (CNNs) in real-world applications, particularly on resource-constrained devices, is hindered by their computational complexity and memory requirements. This research investigates the application of pruning and quantization techniques to optimize CNNs for arrhythmia classification using the MIT-BIH Arrhythmia Database. By combining magnitude-based pruning, regularization-based pruning, filter map-based pruning, and quantization at different bit-widths (4-bit, 8-bit, 2-bit, and 1-bit), the study aims to develop a more compact and efficient CNN model while maintaining high accuracy. The experimental results demonstrate that these techniques effectively reduce model size, improve inference speed, and maintain accuracy, adapting them for use on devices with limited resources. The findings highlight the potential of these optimization techniques for real-time applications in mobile health monitoring and edge computing, paving the way for broader adoption of deep learning in resource-limited environments.</p>

Tese de mestrado integrado. Bioengenharia. Área de Especialização de Engenharia Biomédica. Faculdade de Engenharia. Universidade do Porto. 2012

This master's thesis describes basics principles of QRS complex detection. It is focused on QRS detection using zero crossing counts method. There are described princips and program realization of this method. The other part is focused on genetic optimalization algorithm. There are presented obtained optimalization results on standard CSE and MIT-BIH database. The quality of the detector is compared with other authors. The optimalized QRS detector achieves comparable results with other authors. The part of the thesis is graphical user interface which supply view on modified ECG signal and detection results.

<p dir="ltr">Cardiac disease is one of the prominent reasons of deaths worldwide. The timely de-<br>tection of arrhythmias, one of the highly prevalent cardiac abnormalities, is very important<br>and promising for treatment. Electrocardiography (ECG) is well applied to probe the car-<br>diac dynamics, nevertheless, it is still challenging to robustly detect the arrhythmia with<br>automatic algorithms, especially when the noise may contaminate the signal to some extent.<br>In this research study, we have not only built and assessed different neural network models<br>to understand their capability in terms of ECE-based arrhythmia detection, but also com-<br>prehensively investigated the detection under different kinds of signal-to-noise ratio (SNR).<br>Both Long Short-Term Memory (LSTM) model and Multi-Layer Perception (MLP) model<br>have been developed in the study. Further, we have studied the necessity of fine-tuning<br>of the neural network models, which are pre-trained on other data and demonstrated that<br>it is very important to boost the performance when ECG is contaminated by noise. In<br>the experiments, the LSTM model achieves an accuracy of 99.0%, F1 score of 97.9%, and<br>high precision and recall, with the clean ECE signal. Further, in the high SNR scenario,<br>the LSTM maintains an attractive performance. With the low SNR scenario, though there<br>is some performance drop, the fine-tuning approach helps performance improvement criti-<br>cally. Overall, this study has built the neural network models, and investigated different<br>kinds of signal fidelity including clean, high-SNR, and low-SNR, towards robust arrhythmia<br>detection.</p>

Over the last century, physiological signals have been broadly analyzed and processed not only to assess the function of the human physiology, but also to better diagnose illnesses or injuries and provide treatment options for patients. In particular, Electrocardiogram (ECG), blood pressure (BP) and impedance are among the most important biomedical signals processed and analyzed. The majority of studies that utilize these signals attempt to diagnose important irregularities such as arrhythmia or blood loss by processing one of these signals. However, the relationship between them is not yet fully studied using computational methods. Therefore, a system that extract and combine features from all physiological signals representative of states such as arrhythmia and loss of blood volume to predict the presence and the severity of such complications is of paramount importance for care givers. This will not only enhance diagnostic methods, but also enable physicians to make more accurate decisions; thereby the overall quality of care provided to patients will improve significantly. In the first part of the dissertation, analysis and processing of ECG signal to detect the most important waves i.e. P, QRS, and T, are described. A wavelet-based method is implemented to facilitate and enhance the detection process. The method not only provides high detection accuracy, but also efficient in regards to memory and execution time. In addition, the method is robust against noise and baseline drift, as supported by the results. The second part outlines a method that extract features from ECG signal in order to classify and predict the severity of arrhythmia. Arrhythmia can be life-threatening or benign. Several methods exist to detect abnormal heartbeats. However, a clear criterion to identify whether the detected arrhythmia is malignant or benign still an open problem. The method discussed in this dissertation will address a novel solution to this important issue. In the third part, a classification model that predicts the severity of loss of blood volume by incorporating multiple physiological signals is elaborated. The features are extracted in time and frequency domains after transforming the signals with Wavelet Transformation (WT). The results support the desirable reliability and accuracy of the system.