Deep Learning based Ultrasound Computed Tomography for Real-Time Construction

Qinhan Gao, Mohamed Almekkawy

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Conventional Ultrasound Computed Tomography (UCT) techniques, which rely on iterative optimization methods, are hindered by their high computational demands, making real-time processing impractical. To address this issue, we introduce an innovative deep learning-based solution that directly addresses the mapping between the received ultrasound signals and Speed of Sound (SoS) distribution images. Our approach employs a Convolutional Neural Network (CNN) that incorporates both 1D and 2D convolutions to process the input signals. Simulation data were generated using the k-wave open-source acoustic wave solver to simulate the wave equation. Remarkably, our experimental outcomes show the effectiveness of the proposed neural network in generating highly accurate SoS images, all achieved within a significantly shorter timeframe than traditional methods.

Original languageEnglish (US)
Title of host publicationIUS 2023 - IEEE International Ultrasonics Symposium, Proceedings
PublisherIEEE Computer Society
ISBN (Electronic)9798350346459
DOIs
StatePublished - 2023
Event2023 IEEE International Ultrasonics Symposium, IUS 2023 - Montreal, Canada
Duration: Sep 3 2023Sep 8 2023

Publication series

NameIEEE International Ultrasonics Symposium, IUS
ISSN (Print)1948-5719
ISSN (Electronic)1948-5727

Conference

Conference2023 IEEE International Ultrasonics Symposium, IUS 2023
Country/TerritoryCanada
CityMontreal
Period9/3/239/8/23

All Science Journal Classification (ASJC) codes

  • Acoustics and Ultrasonics

Fingerprint

Dive into the research topics of 'Deep Learning based Ultrasound Computed Tomography for Real-Time Construction'. Together they form a unique fingerprint.

Cite this