Ultrasound synthetic aperture imaging with SpecularNet for enhanced needle visualization.
Authors
Affiliations (2)
Affiliations (2)
- Department of Electrical Engineering, National Taiwan University of Science and Technology, #43, Section 4, Keelung Road, Taipei, 106, Taiwan. [email protected].
- Department of Electrical Engineering, National Taiwan University of Science and Technology, #43, Section 4, Keelung Road, Taipei, 106, Taiwan.
Abstract
Ultrasound imaging has been routinely used for needle guidance due to its real-time capability and cost-efficiency. However, conventional ultrasound with fixed-angle transmission often suffers from reduced visualization of oblique needles when the specular echoes are reflected away from the receiving transducer. Synthetic transmit aperture (STA) beamforming with delay and standard deviation (DASD) relies on the variation of echo magnitude among various incident angles to construct the image. It can be further integrated with the specular information of the needle shaft predicted by convolutional neural networks (CNN) models to improve the image contrast in needle detection. The proposed SpecularNet estimates the orientation and the specular probability of the needle from pixel-wise STA channel data. The proposed specular STA-DASD was validated using experiments. Results indicated that, for needle orientations between 10° and 40°, the specular STA-DASD improved the needle-to-speckle ratio (NSR) by 11.0 dB and the needle-to-artifact ratio (NAR) by 18.9 dB compared to the conventional STA-DASD. Compared to standard specular beamforming, the inference to specular information can be efficiently performed in SpecularNet, while both the NSR and NAR of the corresponding specular STA-DASD also improve by another 25-30 dB. The efficacy of the proposed specular STA-DASD beamforming was demonstrated for enhancement of needle visualization. The specular information could be predicted using SpecularNet with high computational efficiency to guide the transmit/receive apodization and to suppress background noise and artifacts using image weighting.