Naslov
Principal component analysis and visualization in optimization and personalization of lead's set for generation of standard 12-leads ECG

Autori
Tomašić, Ivan ; Skala, Karolj ; Trobec, Roman

Vrsta, podvrsta i kategorija rada
Radovi u zbornicima skupova, cjeloviti rad (in extenso), znanstveni

Izvornik
MIRRO 2008 / Biljanović, Petar ; Skala, Karolj ; - Opatija : Hrvatska udruga za informacijsku i komunikacijsku tehnologiju, elektroniku i mikroelektroniku - MIPRO, 2008, 307-313

ISBN
978-953-233-036-6

Skup
31st International convention MIPRO, Grid and Visualization System Conference

Mjesto i datum
Opatija, Hrvatska, 26.05.2008. - 30.05.2008

Vrsta sudjelovanja
Predavanje

Vrsta recenzije
Međunarodna recenzija

Ključne riječi
Component analyses; ECG devices; ECG visualisation

Sažetak
One of the most important requirements for personal and mobile ECG devices that can generate standard 12-leads ECG is minimizing number of electrodes. This paper proposes application of Principal Component Analysis (PCA) in process of determining minimal set of unipolar electrodes for generation of standard 12-lead ECG. PCA is multivariate statistical technique frequently utilized for ECG signal processing and dimension reduction. We introduce methodology that incorporates PCA and visualization to determine subset, of 31-multichannel ECG (MECG) leads set, which carries most of the system variability. For every patient a different subset of leads is defined. This subset, together with principal components, is then used to generate standard 12- leads ECG with minimum loss of diagnostic information.

Izvorni jezik
Engleski

Znanstvena područja
Elektrotehnika, Računarstvo, Kliničke medicinske znanosti

POVEZANOST RADA