This thesis presents the implementation of machine learning algorithms for the classification of electrocardiographic (ECG) signals using the MATLAB programming environment. After an overview of biomedical signals and an in-depth description of the electrocardiographic signal, the work focuses on the automatic classification of ECG signals for the identification of arrhythmias and cardiac abnormalities. Various machine learning algorithms were used, including Support Vector Machine (SVM), K-Nearest Neighbors (KNN), Decision Tree and Random Forest. The data comes from the MIT-BIH Arrhythmia and PTB Diagnostic ECG databases, which are widely recognized in the biomedical field. Through a rigorous analysis of the performance of the algorithms, the thesis demonstrates that machine learning techniques can improve the accuracy of automated cardiac diagnosis, contributing to the advancement of wearable technologies and clinical monitoring systems.
Questa tesi presenta l'implementazione di algoritmi di machine learning per la classificazione di segnali elettrocardiografici (ECG) utilizzando l'ambiente di programmazione MATLAB. Dopo una panoramica sui segnali biomedici e una descrizione approfondita del segnale elettrocardiografico, il lavoro si concentra sulla classificazione automatica dei segnali ECG per l'identificazione di aritmie e anomalie cardiache. Sono stati utilizzati vari algoritmi di apprendimento automatico, tra cui Support Vector Machine (SVM), K-Nearest Neighbors (KNN), Decision Tree e Random Forest. I dati provengono dai database MIT-BIH Arrhythmia e PTB Diagnostic ECG, ampiamente riconosciuti nel settore biomedico. Attraverso una rigorosa analisi delle prestazioni degli algoritmi, la tesi dimostra che le tecniche di machine learning possono migliorare l'accuratezza della diagnosi cardiaca automatizzata, contribuendo al progresso delle tecnologie indossabili e dei sistemi di monitoraggio clinico.
IMPLEMENTAZIONE DI ALGORITMI DI MACHINE LEARNING PER LA CLASSIFICAZIONE DI SEGNALI ECG
LAUDADIO, ELISABETTA MARIA
2023/2024
Abstract
This thesis presents the implementation of machine learning algorithms for the classification of electrocardiographic (ECG) signals using the MATLAB programming environment. After an overview of biomedical signals and an in-depth description of the electrocardiographic signal, the work focuses on the automatic classification of ECG signals for the identification of arrhythmias and cardiac abnormalities. Various machine learning algorithms were used, including Support Vector Machine (SVM), K-Nearest Neighbors (KNN), Decision Tree and Random Forest. The data comes from the MIT-BIH Arrhythmia and PTB Diagnostic ECG databases, which are widely recognized in the biomedical field. Through a rigorous analysis of the performance of the algorithms, the thesis demonstrates that machine learning techniques can improve the accuracy of automated cardiac diagnosis, contributing to the advancement of wearable technologies and clinical monitoring systems.| File | Dimensione | Formato | |
|---|---|---|---|
|
ELISABETTA MARIA LAUDADIO TESI.pdf
File consultabile da postazioni biblioteca
Descrizione: Tesi di Laurea
Dimensione
3.34 MB
Formato
Adobe PDF
|
3.34 MB | Adobe PDF |
I documenti in UNITESI sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/20.500.14253/66825