Studies on Approaches to Artifact Suppression in Ultrasound-Based Statistical Parametric Imaging

Ultrasound grayscale imaging (i.e., B-mode) is an important image-based diagnostic tool, having a good ability to describe the tissue morphology. Under the influences from system characteristics and operators, grayscale imaging cannot be used for tissue characterization in clinic. To resolve this problem, we have developed the ultrasonic Nakagami imaging based on the Nakagami statistical distribution model to complement the conventional B-scan. The ability of the Nakagami image to reflect the difference of the scatterer properties between normal and abnormal tissues has been demonstrated by our previous studies. However, according to our preliminary findings, the artifact problem would occur in the Nakagami image in some conditions, including the noise-induced cavity artifact in the Nakagami image of the anechoic areas and the far-field artifact due to the beam diffraction of the transducer. The above artifact problems result in information error in the Nakagami image, strongly affecting the performance of using the Nakagami image to classify benign and malignant tissues in clinical situations. Artifact suppression is a key point to keep developing the Nakagami image technique for future clinical applications. Therefore, the goal of this proposed two-year research project is to implement the artifact-free Nakagami imaging technique. In the first year, we will develop the noise-assisted correlation algorithm (NCA) as a strategy to suppress the cavity artifact. Computer simulations will be carried out to explore the feasibility of the NCA method and to find the optimal condition to perform NCA. Phantom experiments will be used to evaluate the practical performance of the NCA to reduce the cavity artifact, and we will also study the influences of frequency, scatterer concentration, signal-to-noise ratio, and beamforming on NCA operation. In the second year, multi-focus Nakagami imaging reconstruction technique will be developed to improve the far-field artifact. At first, simulations are used to explore the relationship between the degree of beam diffraction and the information bias of the Nakagami image due to far-field artifact effect. Phantom measurements are used to explore the feasibility of the multi-focus Nakagami imaging to reduce the far-field and the effects of frequency, scatterer concentration, beamforming, and the number of focus on the multi-focus Nakagami imaging reconstruction. Finally tissue ablation experiments will be carried out to explore the performance of combining the NCA and the multi-focus Nakagami imaging to detect the tissue denaturation for evaluating whether the artifact-free Nakagami image has a better ability in tissue characterization. This research project is expected to contribute two SCI papers, two conference papers, and one patent application per year.

Project ID:PB10101-3628
External Project ID:NSC100-2628-E182-003-MY2