Upotreba wavelet detektora u analizi P-vala elektrokardiograma nakon ugradnje aortokoronarnih premosnica

doctoral thesis

• Title (english): Application of wavelet detector in ECG P-wave analysis after coronary bypass grafting
• Author: Siniša Sovilj
• Mentor: Ratko Magjarević (mentor)
• Committee member: Stanko Tonković (predsjednik povjerenstva)
• Committee member: Velimir Išgum (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; (Department of Electronic Systems and Information Processing); Zagreb
• Defense date and country: 2006-07-07, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing; Information Systems
• Scientific / art field, discipline and subdiscipline: BIOMEDICINE AND HEALTHCARE; Clinical Medical Sciences; Surgery
• Universal decimal classification (UDC): 004 - Computer science and technology. Computing. Data processing ; 616 - Pathology. Clinical medicine

Abstract

Razvijena je metodologija dugotrajnog, višednevnog i kontinuiranog praćenja i analize P vala elektrokardiograma snimanog II odvodom nakon operacije ugradnje aortokoronarnih premosnica. Projektirani su QRS i P val detektori temeljeni na wavelet transformaciji te su im radne značajke verificirane na signalima standardne MIT-BIH baze elektrokardiograma. Mjerena osjetljivost detektora iznosi 99,70%, a pozitivna prediktivnost iznosi 99,73%. Brojni parametri P vala u vremenskoj i frekvencijskoj... More domeni bili su mjereni i izračunati za grupu od 48 pacijenata nakon operacije aortokoronarnog premoštenja. Korištenjem wavelet transformacije signala dobiven je dublji uvid u atrijsku aktivnost u multirezolucijskoj domeni. Svi mjereni parametri su usrednjeni za svaki sat snimljenog EKG-a nakon operacije. Učinjena je statistička i diskriminacijska analiza mjerenih parametara te su identificirani parametri koji najbolje diskriminiraju grupu pacijenata koja je razvila atrijsku fibrilaciju poslije operacije od grupe pacijenata koja to nije. Vrijednosti tih parametara smatrane su prediktorima poslijeoperativne atrijske fibrilacije. Predložen je model klasifikacijskog stabla za klasifikaciju pacijenata poslije operacije ugradnje aortokoronarnih premosnica temeljen na parametrima P vala. Postignut je visok iznos specifičnosti (96,55%), pozitivne prediktivnosti (85,71%) i negativne prediktivnosti (84,84%). Ukupna točnost predloženog modela iznosi 85%, a postignuta osjetljivost iznosila je 54,54%. Less

Abstract (english)

A method of long-term continuous monitoring and analysis of P waves in II standard lead electrocardiogram of patients after Coronary Artery Bypass Grafting (CABG) has been developed. The QRS and P wave wavelet detectors we designed for our analysis algorithm have been verified on signals from the MIT-BIH database. We measured the sensitivity of the detector 99.70% and the positive predictivity 99.73%. Numerous P wave parameters in time and frequency domain have been acquired and... More calculated for a group of 48 patients after CABG. We also used wavelet transformation in order to provide deeper insight into atrial activity in multiresolution domain. All measured parameters were averaged for each hour of the recorded ECGs after the surgery. Statistical and discriminant analysis of the measured P wave parameters has been made in order to identify those parameters which discriminate the patients which developed the atrial fibrillation after the surgery from those who have not developed it during hospitalization. The values of the measured P wave parameters were considered as predictors of atrial fibrillation after CABG. We proposed a classification tree model of atrial fibrillation prediction based on P wave parameters. We have obtained high specificity (96.55%), positive predictivity (85.71%) and negative predictivity (84.84%) and overall accuracy (85%) of the model but the achieved sensitivity of the model was 54.54%. Less

• Keywords: atrijska fibrilacija; elektrokardiogram (EKG); QRS kompleks; wavelet transformacija; usrednjeni P val; aortokoronarno premoštenje; diskriminantna analiza; klasifikacijsko stablo
• Keywords (english): atrial fibrillation; electrocardiogram (ECG); QRS detection; wavelet transformation; averaged P wave parameters; coronary artery bypass grafting (CABG); discriminant analysis; classification tree
• Language: croatian
• URN:NBN: urn:nbn:hr:168:359050
• Study programme: Title: Electrical Engineering and Computing; Study programme type: university; Study level: postgraduate; Academic / professional title: Doktor znanosti elektrotehnike i računarstva (Doktor znanosti elektrotehnike i računarstva)
• Type of resource: Text
• Extent: str. 85 ; 30 cm
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2019-05-06 13:06:23