Deep Generative Models to Simulate 2D Patient-Specific Ultrasound Images\n in Real Time

We present a computational method for real-time, patient-specific simulation\nof 2D ultrasound (US) images. The method uses a large number of tracked\nultrasound images to learn a function that maps position and orientation of the\ntransducer to ultrasound images. This is a first step towards realistic\npatient-specific simulations that will enable improved training and\nretrospective examination of complex cases. Our models can simulate a 2D image\nin under 4ms (well within real-time constraints), and produce simulated images\nthat preserve the content (anatomical structures and artefacts) of real\nultrasound images.\n