Pronalaženje karakterističnih dijelova elektrokardiograma vremensko-frekvencijskom analizom

scientific master's theses

• Title (english): Detection of ECG Segments by Means of Time-Frequency Analysis
• Author: Miroslav Jeras
• Mentor: Ratko Magjarević (mentor)
• Committee member: Vedran Bilas (predsjednik povjerenstva)
• Committee member: Ratko Magjarević (član povjerenstva)
• Committee member: Vladimir Medved (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; (Department of Electronic Systems and Information Processing); Zagreb
• Defense date and country: 2003, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Electrical Engineering; Electronics
• Universal decimal classification (UDC): 621.3 - Electrical engineering

Abstract

Razvijena je metoda za detekciju P-vala u EKG signalu u stvarnom vremenu pomoću valićne transformacije. Metoda se sastoji od dva dijela: algoritma za detekciju segmenata EKG signala s pretraživanjem i algoritma za detekciju P-vala u stvarnom vremenu. Algoritam s pretraživanjem koji ne radi u stvarnom vremenu se koristi za prikupljanje statističkih podataka o trenutnom obliku P-vala. Na osnovi te analize predviđa se oblik P-vala, kojeg drugi algoritam koristi za detekciju P-valova u signalu odmah nakon njihovog završetka. Oba algoritma koriste diskretnu dijadsku biortogonalnu valićnu transformaciju za transformaciju signala i mjerenje karakterističnih parametara valova u EKG signalu. Uspješnost algoritma je ocijenjena pomoću signala iz MIT/BIH Arrhythmia Database i ezofagijskih signala izmjerenih na životinjama. Rezultati ispitivanja za površinski EKG pokazuju da se algoritam može uspješno primjenjivati na signalima površinskog i ezofagijskog EKG-a kod kojih je amplituda P-vala dovoljno veća od amplitude smetnji. Kod ezofagijskog signala postiže se veća pouzdanost detekcije, posebno kod implantacije ezofagijske sonde u blizini pretklijetke. Prikazani se algoritam zbog toga može koristiti za pronalaženje optimalnog položaja ezofagijske sonde za snimanje EKG-a ili transezofagijsku stimulaciju srca.

Abstract (english)

A method for real-time detection of P-wave in ECG signal using the wavelet transform has been developed. The method consists of two parts: a search-back algorithm for ECG segment detection and a real time P-wave detection algorithm. The search-back algorithm for P-wave detection is used for collecting statistical data on current P-wave shape. Based on that analysis the prediction of P-wave shape is calculated and another algorithm detects the P-wave in real time using that data. Both algorithms use discreet dyadic biorthogonal wavelet transform for signal transformations and for measuring characteristic parameters of waves in the ECG signal. The algorithm was evaluated on surface ECG signals from the MIT/BIH Arrhythmia Database and esophageal ECG signals recorded on experimental animals. The results for the surface ECG show that the algorithm can be successfully used on signals where the amplitude of P-waves is high enough compared to the amplitude of noise. Real-time P-wave detection is more successful in the esophageal ECG, especially when the esophageal lead is implanted close to the atrium. Therefore, this algorithm can be used in optimal positioning of esophageal leads for transesophageal ECG recording and cardiac stimulation.

• Keywords: EKG; detekcija P-vala u stvarnom vremenu; valićna transformacija; ezofagijski EKG
• Keywords (english): ECG; real-time P-wave detection; wavelet transform; esophageal ECG
• Language: croatian
• URN:NBN: urn:nbn:hr:168:076127
• Study programme: Title: Postgraduate master programme in electrical engineering; Study programme type: university; Study level: postgraduate; Academic / professional title: Magistar znanosti elektrotehnike (Magistar znanosti elektrotehnike)
• Catalog URL: https://lib.fer.hr:8443/cgi-bin/koha/catalogue/detail.pl?biblionumber=29156
• Type of resource: Text
• Extent: 86 str. ; 30 cm
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2019-09-11 12:18:44