Abstract 17795: Study of Inter-Vendor Variability of 3D Echo Based Strain Measurements: Role of Vendor Specific and Vendor Independent Analysis Tools
Background: We studied the inter-vendor variability of 3D echo derived strain measurements.
Methods: Ten freshly harvested pig hearts were studied. Each pig heart was driven passively in a water tank by a calibrated pulsatile pump at different stroke volumes (20-80ml). Full volume 3D image loops were acquired at each pumped stroke volume by three major 3D Ultrasound systems (GE Vivid E9, Siemens SC 2000 and Toshiba Aplio Artida) keeping volume rate between 18-22/cycle. Image data from all systems was analyzed offline for circumferential strain at mid cavity level using vendor specific analysis programs (EchoPac for GE, WMT for Toshiba and VMM for Siemens), and a vendor independent analysis program (TomTec). Inter-vendor variability of strain measurements was determined using intra-class correlation (ICC).
Results: The ICC values using 1000 bootstrap samples are 0.5457392 (95% bootstrap CIs: [0.4445, 0.6390]) for vendor specific, and 0.793742(95% bootstrap CIs: [0.7182, 0.8465]) for vendor independent strain measurements. The lower bound of the 95% CI for ICC with vendor independent is higher than the upper bound of the 95% CI for ICC with Vendor specific. Higher intra-class correlation (ICC) indicates smaller inter-variability. The plots show no overlapping between the 95% CI for ICC between the two approaches: vendor specific (blue) vs. vendor independent (red).
Conclusions: Since higher intra-class correlation implies smaller inter-variability, we can conclude that vendor-independent strain measurements show smaller variability.
- © 2013 by American Heart Association, Inc.