eduzhai > Applied Sciences > Engineering >

Deep Network for Scatterer Distribution Estimation for Ultrasound Image Simulation

  • Save

... pages left unread,continue reading

Document pages: 11 pages

Abstract: Simulation-based ultrasound training can be an essential educational tool.Realistic ultrasound image appearance with typical speckle texture can bemodeled as convolution of a point spread function with point scatterersrepresenting tissue microstructure. Such scatterer distribution, however, is ingeneral not known and its estimation for a given tissue type is fundamentallyan ill-posed inverse problem. In this paper, we demonstrate a convolutionalneural network approach for probabilistic scatterer estimation from observedultrasound data. We herein propose to impose a known statistical distributionon scatterers and learn the mapping between ultrasound image and distributionparameter map by training a convolutional neural network on synthetic images.In comparison with several existing approaches, we demonstrate in numericalsimulations and with in-vivo images that the synthesized images from scattererrepresentations estimated with our approach closely match the observations withvarying acquisition parameters such as compression and rotation of the imageddomain.

Please select stars to rate!

         

0 comments Sign in to leave a comment.

    Data loading, please wait...
×