Study on Phase Correction for USCT Echo Image by Sound-Speed Image With Different Resolutions and Noise Levels

Abstract

Ultrasound computed tomography (USCT) is a promising technique for breast cancer detection. It provides three modalities: sound-speed-image (SSI), attenuation image, and echo image. In these modalities, the echo image is appropriate for the early detection of breast cancer, because it has high resolution. However, echo image reconstruction generally employs an assumed (pre-determined) sound speed to calculate the arrival time of echo signal, the mismatch between the assumed and real sound speeds results in phase aberration of echo signal, and the phase aberration further leads to a decline in image resolution and contrast. To correct the phase aberration, SSI can be used to provide sound speed information instead of the assumed speed, but the resolution of and noise level in SSI may affect the correction performance and echo image quality. In this article, we explored the effect of SSI resolution and the noise level on the resolution and contrast of echo image. First, USCT raw data was obtained by both numerical and experiment studies. Second, echo images were reconstructed by synthetic aperture (SA) technique, and their phase aberration was corrected by multi-stencils fast marching (MSFM) using SSIs with different resolutions and noise levels. Finally, the resolution and contrast of echo images were quantitatively evaluated. Both numerical simulation and phantom experiment demonstrated that the quality of echo images deteriorates as the SSI resolution decreases and the noise level increases.

Keywords

• Echo image
• phase aberration correction
• sound speed image
• ultrasound computed tomography