eduzhai > Applied Sciences > Engineering >

3D B-mode ultrasound speckle reduction using deep learning for 3D registration applications

  • king
  • (0) Download
  • 20210506
  • Save

... pages left unread,continue reading

Document pages: 10 pages

Abstract: Ultrasound (US) speckles are granular patterns which can impede imagepost-processing tasks, such as image segmentation and registration.Conventional filtering approaches are commonly used to remove US speckles,while their main drawback is long run-time in a 3D scenario. Although a fewstudies were conducted to remove 2D US speckles using deep learning, to ourknowledge, there is no study to perform speckle reduction of 3D B-mode US usingdeep learning. In this study, we propose a 3D dense U-Net model to process 3DUS B-mode data from a clinical US system. The model s results were applied to3D registration. We show that our deep learning framework can obtain similarsuppression and mean preservation index (1.066) on speckle reduction whencompared to conventional filtering approaches (0.978), while reducing theruntime by two orders of magnitude. Moreover, it is found that the specklereduction using our deep learning model contributes to improving the 3Dregistration performance. The mean square error of 3D registration on 3D datausing 3D U-Net speckle reduction is reduced by half compared to that withspeckles.

Please select stars to rate!


0 comments Sign in to leave a comment.

    Data loading, please wait...