Doppler myocardial imaging vs. B-mode grey-scale imaging: A comparative in vitro and in vivo study into their relative efficacy in endocardial boundary detection

Abstract

Doppler myocardial imaging (DMI) is a new ultrasound imaging modality in which colour Doppler algorithms are adapted to visualise the myocardium. It allows measurement of regional intramyocardial velocities and quantification of intramural left ventricular function. However promising the technique is, to date the accuracy of endocardial boundary detection by DMI has not been validated. As Doppler velocity estimation is based on measurement of phase shift rather than signal strength, the technique is relatively independent of chest wall attenuation. In the current study, a series of in vitro and in vivo studies was performed to compare standard B-mode grey-scale imaging (GSI) and DMI techniques in endocardial boundary detection. In vitro, the minimum and maximum volumes of a single-chamber tissue-mimicking phantom were calculated using both imaging techniques. In vivo, left ventricular end-diastolic (ED) volume and end-systolic (ES) volume indices were measured from GSI and DMI images in a group of 40 volunteers. All images were obtained in the freeze-frame mode with the Doppler display turned on and off so that simultaneous DMI and GSI information was obtained. In vitro, the limits of agreement between the minimum volume of the phantom and the minimum volume measured by GSI and DMI was 4% and 3%, respectively. For maximum volumes, limits of agreement were 3% for GSI and 2% for DMI. In vivo, the limits of agreement between the two imaging techniques in volume measurements were 6 mL (9%) for ED and 4 mL (11%) for ES. The comparison of the endocardial boundary detection by GSI vs. DMI showed DMI to be significantly superior: ED (72 ± 16% vs. 85 ± 8%, respectively; p < 0.05) and ES (71 ± 13% vs. 88 ± 7%, respectively; p < 0.05). The results of the study show that: (1) in vitro, based on two-dimensional algorithms, DMI provides as accurate volume measurements as GSI; and (2) in vivo, there is a very good agreement of left ventricular volume measurements between GSI and DMI. However, the endocardial boundary is more reliably displayed and visually easier to detect using DMI than GSI.