Complex Convolutional Neural Networks for Ultrasound Image Reconstruction from In-Phase/Quadrature Signal.

A wide variety of studies based on deep learning have recently been investigated to improve ultrasound (US) imaging. Most of these approaches were performed on radio frequency (RF) signals. However, inphase/quadrature (I/Q) digital beamformers (IQBF) are now widely used as low-cost strategies. In this work, we leveraged complex convolutional neural networks (CCNNs) for reconstructing ultrasound images from I/Q signals. We recently described a CNN architecture called ID-Net, which exploited an inception layer devoted to the reconstruction of RF diverging-wave (DW) ultrasound images. We derived in this work the complex equivalent of this network, i.e., the complex inception for DW network (CID-Net), operating on I/Q data. We provided experimental evidence that the CID-Net yields the same image quality as that obtained from the RF-trained CNNs; i.e., by using only three I/Q images, the CID-Net produced high-quality images competing with those obtained by coherently compounding 31 RF images. Moreover, we showed that the CID-Net outperforms the straightforward architecture consisting in processing separately the real and imaginary parts of the I/Q signal, indicating thereby the importance of consistently processing the I/Q signals using a network that exploits the complex nature of such signal.