Evaluation of a semiautomatic software tool for left ventricular function analysis with 16-slice computed tomography |
| |
Authors: | Marc Dewey Mira Müller Florian Teige Bernd Hamm |
| |
Affiliation: | (1) Department of Radiology, Charité, Medical School, Humboldt-Universität zu Berlin, Schumannstr. 20/21, 10117 Berlin, Germany;(2) Department of Radiology, Charité – Universitätsmedizin Berlin, Humboldt-Universität zu Berlin, Schumannstraße 20/21, PO Box 10098, 10117 Berlin, Germany |
| |
Abstract: | The purpose of the study was to evaluate a semiautomatic analysis tool for assessing global left ventricular myocardial function with multislice computed tomography (MSCT). We examined 33 patients with MSCT using 16×0.5 mm detector collimation and magnetic resonance imaging (MRI) on a 1.5-T scanner. MSCT data were analyzed using semiautomatic volumetric analysis software (ANET, CSCF-001A, Toshiba). This software tool automatically creates endo- and epicardial contours that can be manually corrected on all short-axis slices at all reconstructed time points within the cardiac cycle, based on a contour-detection and density-threshold algorithm. All global left ventricular function parameters assessed with the semiautomatic MSCT software were highly correlated with the results of MRI. Bland-Altman analysis showed minor systematic overestimation of end-diastolic (10.7 ml) and end-systolic volumes (5.6 ml) and underestimation of ejection fraction (2.1%) with MSCT as compared with MRI. The post-processing time was moderately but significantly longer with the MSCT software (15.9±2.8 min) than necessary for MRI (14.0±2.5 min, P<0.01), mainly as a result of the longer time required for uploading of the MSCT datasets, which were on average 54 times larger (1.3 GByte). In conclusion, it appears feasible to accurately assess global left ventricular function with MSCT in a reasonable post-processing time using a semiautomatic software tool. |
| |
Keywords: | Myocardial function Computed tomography Heart Ventricles Magnetic resonance MSCT CT MRI |
本文献已被 PubMed SpringerLink 等数据库收录! |
|