Deep learning-based reconstruction of ultrasound images from raw channel data

Abstract

Purpose We investigate the feasibility of reconstructing ultrasound images directly from raw channel data using a deep learning network. Starting from the raw data, we present the network the full measurement information, allowing for a more generic reconstruction to form, as compared to common reconstructions constrained by physical models using fixed speed of sound assumptions. Methods We propose a U-Net-like architecture for the given task. Additional layers with strided convolutions downsample the raw data. Hyperparameter optimization was used to find a suitable learning rate. We train and test our deep learning approach on plane wave ultrasound images with a single insonification angle. The dataset includes phantom as well as in vivo data. Results The images produced by our method are visually comparable to ones reconstructed with the conventional delay and sum algorithm. Deviations between prediction and ground truth are likely to be related to speckle noise. For the test set, the mean absolute error is 4.23 ± 1.52 for the phantom images and 6.09 ± 0.72 for the in vivo data. Conclusion The result shows the feasibility of our approach and opens up new research directions regarding information retrieval from raw channel data. As the networks reconstruction performance is limited by the quality of the ground truth images, using other ultrasound reconstruction technique or image types as target data would be of interest.

A Convolutional Neural Network for Beamforming and Image Reconstruction in Passive Cavitation Imaging

“…The field of deep learning in medical ultrasound imaging is rapidly expanding with many applications such as clutter suppression in Doppler [1], super-resolution imaging [2,3], anomaly detection methods for breast ultrasound images [4], beamforming pre-steered, subsampled data [5], transcranial ultrasound imaging [6,7], minimum variance beamforming [8,9], and image reconstruction from raw channel data [10][11][12][13]. In the context of image reconstruction from raw channel data, CNNs have proven their capability to learn the reconstruction process without requiring explicit input information regarding receiver array geometry, a medium speed of sound, or the spatial discretization of the imaged region of interest (parameters which are typically essential for the standard delay and sum algorithm).…”