Minimizing Image Quality Loss After Channel Count Reduction for Plane Wave Ultrasound via Deep Learning Inference

Abstract

High-frame-rate ultrasound imaging uses unfocused transmissions to insonify an entire imaging view for each transmit event, thereby enabling frame rates over 1,000 fps. At these high frame rates, it is naturally challenging to realize real-time transfer of channel-domain raw data from the transducer to the system back-end. Our work seeks to halve the total data transfer rate by uniformly decimating the receive channel count by 50% and, in turn, doubling the array pitch. We show that, despite the reduced channel count and the inevitable use of a sparse array aperture, the resulting beamformed image quality can be maintained by designing a custom convolutional encoder-decoder neural network to infer the radiofrequency (RF) data of the nullified channels. This deep learning framework was trained with in vivo human carotid data (5 MHz plane wave imaging; 128 channels; 31 steering angles over a 30° span; 62,799 frames in total). After training, the network was tested on an in vitro point target scenario that was dissimilar to the training data, in addition to in vivo carotid validation datasets. In the point target phantom image beamformed from inferred channel data, spatial aliasing artifacts attributed to array pitch doubling were found to be reduced by up to 10 dB. For carotid imaging, our proposed approach yielded a lumen-to-tissue contrast that was on average within 3 dB compared to the full-aperture image, whereas without channel data inferencing, the carotid lumen was obscured. When implemented on an RTX-2080 GPU, the inference time to apply the trained network was 4 ms which favors real-time imaging. Overall, our technique shows that, with the help of deep learning, channel data transfer rates can be effectively halved with limited impact on the resulting image quality.1

Deep learning assisted sparse array ultrasound imaging

“…According to the Huygens principle, the pitch of linear array ultrasound transducer is required to be close to one wavelength to form a single wave front from wave fronts from each element [ 3 ]. When the pitch is much larger than the wavelength, the interference of ultrasound waves from each element would generate unwanted grating lobes and degrade the image quality, which is the limitation of sparse array imaging [ 4 ]. In this study, a predictive CNN model was introduced to assist in restoring the sparse array imaging and improving the image resolution.…”

“…These dimensions are chosen to prevent imaging artifacts due to grating lobes, which are unwanted additional beams that can lead to spatial aliasing artifacts such as false echoes or image smearing. Such artifacts can compromise the image quality and accuracy of diagnostic information derived from the image [ 4 ]. Ultrasound transducers for 2D imaging usually have over one hundred transducer elements [ 5 ], and 2D array transducers for 3D imaging may have several thousands of elements [ 6 ].…”

“…Deep learning technologies have also been proposed to improve the image quality of sparse array ultrasound imaging [ 4 , 19 ]. The resolution and SNR of sparse array images via deep learning were comparable to traditional dense array imaging [ 20 ].…”

“…However, these existing works are only demonstrated on simulations or imaging phantoms with fixed structures rather than real tissue [ 21 , 22 ]. Other works showed in vivo/ex vivo experiments were performed but with a pitch reduction of only two-fold ( S1 Table ) [ 4 , 19 , 23 ].…”

Branched Convolutional Neural Networks for Receiver Channel Recovery in High-Frame-Rate Sparse-Array Ultrasound Imaging

“…Methods that can enable high-frame-rate receiver channel reduction with capability to recover missing raw RF data would help to integrate high-frame-rate techniques into compact ultrasound systems. Indeed, our group has previously developed a deep-learning-based RF recovery system that can recover a full set of plane-wave RF channel data from only half of receiving channels [18]. This framework can enable a system to operate with fewer receiving channels, while providing access to a full set of data during ultrasound image formation.…”