Image reconstruction for electrical impedance tomography based on spatial invariant feature maps and convolutional neural network

Delin Hu, Keming Lu, Yunjie Yang

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

Abstract

Data-driven methods are attracting more and more attention in the field of electrical impedance tomography. Many learning-based tomographic algorithms have been presented and investigated in the past few years. However, few related studies pay attention to the symmetrical geometrical structure of tomographic sensors and the possible benefits it may bring to learning-based image reconstruction. Aiming to this, we propose the concept of electrical impedance maps, which can better reflect the nature of geometry of tomographic sensors and have similar properties to images. Then we design a fully convolutional network to build the relationship between electrical impedance maps and conductivity distribution images. The effectiveness and performance of our method is evaluated by both simulation and experimental datasets with different conductivity distribution patterns.
Original languageEnglish
Title of host publication2019 IEEE International Conference on Imaging Systems and Techniques (IST)
PublisherIEEE
Pages1-6
ISBN (Electronic)978-1-7281-3868-8
DOIs
Publication statusPublished - 27 Feb 2020
Event2019 IEEE International Conference on Imaging Systems and Techniques - ABU DHABI, United Arab Emirates
Duration: 8 Dec 201910 Dec 2019
https://ist2019.ieee-ims.org/

Conference

Conference2019 IEEE International Conference on Imaging Systems and Techniques
Abbreviated titleIST 2019
Country/TerritoryUnited Arab Emirates
CityABU DHABI
Period8/12/1910/12/19
Internet address

Fingerprint

Dive into the research topics of 'Image reconstruction for electrical impedance tomography based on spatial invariant feature maps and convolutional neural network'. Together they form a unique fingerprint.

Cite this