Ultrasound Beamforming using MobileNetV2
Sobhan Goudarzi, Amir Asif, Hassan Rivaz
Abstract
In the past few years, deep learning has disrupted several low-level medical imaging tasks such as reconstruction of Computed Tomography (CT) and Magnetic Resonance (MR) images. In this work, we propose a novel deep learning-based approach for the low-level task of ultrasound image reconstruction from the pre-beamformed channel data. More specifically, we adapt MobileNetV2 to train a model that mimics Minimum Variance Beamforming (MVB). Results confirm that the proposed method takes much less time to reconstruct images with a quality similar to one achieved by applying MVB directly. The current paper is a part of our submission to Challenge on Ultrasound Beamforming with Deep Learning (CUBDL) announced by 2020 IEEE International Ultrasonics Symposium (IUS).