eduzhai > Applied Sciences > Engineering >

Switchable Deep Beamformer

  • king
  • (0) Download
  • 20210507
  • Save

... pages left unread,continue reading

Document pages: 10 pages

Abstract: Recent proposals of deep beamformers using deep neural networks haveattracted significant attention as computational efficient alternatives toadaptive and compressive beamformers. Moreover, deep beamformers are versatilein that image post-processing algorithms can be combined with the beamforming.Unfortunately, in the current technology, a separate beamformer should betrained and stored for each application, demanding significant scannerresources. To address this problem, here we propose a { em switchable} deepbeamformer that can produce various types of output such as DAS, speckleremoval, deconvolution, etc., using a single network with a simple switch. Inparticular, the switch is implemented through Adaptive Instance Normalization(AdaIN) layers, so that various output can be generated by merely changing theAdaIN code. Experimental results using B-mode focused ultrasound confirm theflexibility and efficacy of the proposed methods for various applications.

Please select stars to rate!


0 comments Sign in to leave a comment.

    Data loading, please wait...