Clutter reduction in echocardiography with short-lag spatial coherence (SLSC) imaging

Abstract

Clutter, a problematic noise artifact in echocardiography, appears as a diffuse haze that obscures endocardial borders and inhibits accurate diagnoses. Several approaches are available to reduce clutter in cardiac images, yet difficult-to-image patients still exist. We have recently developed a novel imaging method, termed short-lag spatial coherence (SLSC) imaging, that has demonstrated potential to reduce clutter in simulated and experimental data. With this technique, images are created from the same individual channel signals used to form B-mode images, but instead of applying a conventional delay-and-sum beamformer, the data are cross-correlated to measure and display differences in spatial coherence. This technique was applied to in vivo cardiac images. Individual channel signals were acquired to form matched B-mode and SLSC images of the left ventricle in fourteen human volunteers. The contrast and contrast-to-noise ratio (CNR) of the ventricle and the signal-to-noise ratio (SNR) of the endocardium were measured in the same locations in matched B-mode and SLSC images. In SLSC images created with a short-lag value equivalent to 16% of the transmit aperture, contrast and CNR was improved by 9±7 dB and 0.4±0.2, respectively, in the SLSC images. The average SNR of the endocardium was 1.7±0.4 in the SLSC images and 1.8±0.4 in the B-mode images. The presented approach demonstrates a new method for reducing clutter in cardiac images.