Super-resolution microbubble localization in unprocessed ultrasound RF signals using a 1D dilated CNN
Nathan Blanken, Jelmer M. Wolterink, Hervé Delingette, Christoph Brune, Michel Versluis, Guillaume Lajoinie
We present a super-resolution ultrasound approach based on direct deconvolution of single-channel ultrasound radio-frequency (RF) signals with a one-dimensional dilated convolutional neural network (CNN). Data are generated with a physics-based simulator that simulates the echoes from a dense cloud of monodisperse microbubbles and captures the full, nonlinear response of resonant, lipid-coated microbubbles. The network is trained with a novel dual-loss function, which features elements of both a classification loss and a regression loss and improves the detection-localization characteristics of the output. The potential of the presented approach to super-resolution ultrasound imaging is demonstrated with a delay-and-sum reconstruction with deconvolved ultrasound data. The resulting image shows an order-of-magnitude gain in axial resolution compared to a delay-and-sum reconstruction with unprocessed element data.
Thursday 7th July
Poster Session 2.1 - onsite 15:20 - 16:20, virtual 11:00 - 12:00 (UTC+2)