Reviews of the most recent applications of deep learning on ultrasound imaging applications are presented. Architectures of deep learning networks are briefly explained for medical imaging application categories of classification, detection, segmentation, and generation. Ultrasonography applications are then reviewed and summarized for image processing and diagnosis along with some representative study cases of breast, thyroid, heart, kidney, liver, and fetal head. Efforts on workflow enhancement are also reviewed with emphasis on view recognition, scanning guide, image quality assessment, and quantification and measurement. Finally some future prospects are presented on image quality Enhancement, diagnostic support, and improving workflow efficiency, along with remarks on hurdles, benefits, and necessary collaborations.
This paper defines the echo signal-to-noise ratio (eSNR) for pulse-echo systems that adapts to the effects of shift-varying impulse responses, spatiotemporal coding, and various beamformers. Measurement techniques using point targets or random scattering media can be interrelated for a broad range of experimental conditions through the eSNR. The eSNR definitions are also illustrated by comparing a spatial matched filter (SMF) beamformer to conventional dynamic receive focusing methods to evaluate performance based on resolution and sensitivity. Closed-form expressions are presented that predict eSNR gains from SMF approaches relative to other beamformers.
In this paper, we propose an efficient method for implementing bi-directional pixel-based focusing(BiPBF) based on a sparse array imaging technique. The proposed method can improve spatial resolution and frame rate of ultrasound imaging with reduced hardware complexity by synthesizing a large transmit aperture with sparsely distributed small subapertures. As the distance between adjacent subapertures increases, however, the image resolution tends to decrease due to the elevation of grating lobes. Such grating lobes can be eliminated in conventional synthetic aperture imaging techniques. On the contrary, the grating lobes of the sparse BiPBF scheme can not be eliminated, which is to be proven analytically in this paper. We also propose the condition and method for suppressing the grating lobes below -40dB, which can be achieved by placing the transmit focal depth at a near depth and properly selecting the subaperture distance in proportion to receive aperture size. The results of both the phantom and in vivo experiments show that the proposed method implements two-way dynamic focusing using a smaller number of subapertures, resulting in reduced system complexity and increased frame rate.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.