This project intends to do end-to-end ultrasonic imaging using convolutional-deconvolutional networks.
The inputs are cross-spectral-matrix computed from signal responses. These responses are acquired from Field II simulation. In the simulation, we use 64-element linear phased array to scan fake tissues, which are different categories of images. These images are also used as ground truth for the outpus of the network.