Decorrelated Compounding in Synthetic Transmit Aperture (STA) Ultrasound Imaging to detect low-contrast lesions

Speckles are ubiquitous in ultrasound images and have been used to track motion and characterize tissues in biomedical applications successfully. However, low signal-to-noise-ratio in ultrasound images due to speckles still severely limits the detection of low-contrast targets. Conventional Spatial Compounding (SC) or Frequency Compounding (SC) was used to reduce the speckles in ultrasonic imaging with limited success, usually at the cost of frame rate. The goal of this paper is to improve the performance of compounding in detecting low-contrast targets. We first generated a large number of correlated sub-images by using various sub-apertures and sub-bands in Synthetic Transmit Aperture (STA) ultrasound imaging. Then we decorrelated and normalized the sub-images by applying Singular Value Decomposition (SVD) to the sub-images. Lastly, the decorrelated sub-images were incoherently compounded to yield the de-speckled images. The speckle variance and contrast of targets in the reconstructed images were quantified. The Contrast-Noise-Ratio (CNR) improvement in simulation images using the proposed method is 1842% over the Delay-and-Sum (DAS) method. The feasibility of applying the proposed method to image a lesion created by high-intensity focused ultrasound was also demonstrated.