Sciact
  • EN
  • RU

Encoder neural network in 2D acoustic tomography Full article

Journal Applied and Computational Mathematics
ISSN: 1683-3511
Output data Year: 2024, Volume: 23, Number: 1, Pages: 83-98 Pages count : 16 DOI: 10.30546/1683-6154.23.1.2024.83
Tags Acoustic Tomography, Ultrasound, Deep Learning, Neural Networks, Coefficient Inverse Problem.
Authors Prikhodko A.Yu. 1 , Shishlenin M.A. 1,2,3 , Novikov N.S. 1,2,3 , Klyuchinskiy D.V. 2
Affiliations
1 Novosibirsk State University
2 Institute of Computational Mathematics and Mathematical Geophysics
3 Sobolev Institute of Mathematics

Funding (1)

1 Russian Science Foundation 19-11-00154

Abstract: We investigate deep learning approach in 2D dynamic ultrasound acoustic tomography. The mathematical model of acoustic tomography is described by a first-order hyperbolic system PDE and is based on conservation laws. This model guarantees us that the training sets of dynamic data are close to the physical solution. We train a neural network consisting of an encoder and a decoder with this data (they contain only one inclusion) and associate the data with a velocity coefficient. Numerical results show that we recover not only single inclusions, but also homogeneities consisting of two inclusions.
Cite: Prikhodko A.Y. , Shishlenin M.A. , Novikov N.S. , Klyuchinskiy D.V.
Encoder neural network in 2D acoustic tomography
Applied and Computational Mathematics. 2024. Т.23. №1. С.83-98. DOI: 10.30546/1683-6154.23.1.2024.83 WOS Scopus РИНЦ OpenAlex
Dates:
Published print: Apr 15, 2024
Published online: Apr 15, 2024
Identifiers:
Web of science: WOS:001195457700003
Scopus: 2-s2.0-85187486568
Elibrary: 66215788
OpenAlex: W4392395418
Citing:
DB Citing
Web of science 1
Scopus 3
Altmetrics: