EVALUATION OF LEFT-VENTRICULAR VOLUME USING AUTOMATIC BORDER DETECTION IN CHILDREN - A COMPARISON WITH CONVENTIONAL OFF-LINE ECHOCARDIOGRAPHIC QUANTIFICATION

Citation
I. Hashimoto et al., EVALUATION OF LEFT-VENTRICULAR VOLUME USING AUTOMATIC BORDER DETECTION IN CHILDREN - A COMPARISON WITH CONVENTIONAL OFF-LINE ECHOCARDIOGRAPHIC QUANTIFICATION, Acta Paediatrica Japonica Overseas Edition, 40(3), 1998, pp. 226-231
Citations number
20
Categorie Soggetti
Pediatrics
ISSN journal
03745600
Volume
40
Issue
3
Year of publication
1998
Pages
226 - 231
Database
ISI
SICI code
0374-5600(1998)40:3<226:EOLVUA>2.0.ZU;2-#
Abstract
Background: Evaluation of the clinical usefulness of the on-line autom atic border detection system for determination of left ventricular vol ume in children in comparison to the conventional off-line method. Met hods: Eighty consecutive patients in whom clear images were obtained b y two-dimensional echocardiography were studied. Using the Hewlett-Pac kard Sonos 2500 with a 3.5 or 5.5 Mhz phased array transducer, all pat ients were studied in the apical four-chamber imaging plane for automa tic border detection and apical four-chamber and two-chamber imaging p lanes for manual tracing. Left ventricular end-diastolic and end-systo lic volumes were measured and compared using the bi-plane Simpson meth od. Results: Left ventricular end-diastolic volumes obtained by automa tic border detection correlated well but were slightly underestimated compared to those obtained by manual tracing (r = 0.98). Left ventricu lar end-systolic volumes obtained by automatic border detection also c orrelated well with those obtained by manual tracing (r = 0.96). Left ventricular ejection fractions compared favorably. However, left ventr icular volumes obtained using the classical Pombo M-mode echocardiogra phy showed poorer correlation with those obtained by manual tracing me thods. Conclusions: Automatic border detection is a promising method f or real-time estimation of left ventricular volume. In patients with g ood endocardial tracking, automatic border detection can be used for r outine studies of cardiovascular disease, even in children.