Deep Learning for Tumor Margin Identification in Electromagnetic Imaging

Amir Mirbeik, Negar Ebadi Tavassolian

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

This paper proposes a deep learning technique which accurately divides the electromagnetic images of cancer-affected tissues into two regions of tumorous and normal sections. This capability will enable the visualization of the border of the cancerous tissue for the surgeon prior to or during the excision surgery. We formulate deep learning from a perspective relevant to electromagnetic image reconstruction. A recurrent auto-encoder network architecture is presented. The effectiveness of the algorithm is demonstrated by segmenting the reconstructed images of an experimental tissue-mimicking phantom. The structure similarity measure (SSIM) and mean-square-error (MSE) of the reconstructed result are approximately 0.94 and 0.04 respectively, while the values obtained from conventional frequency-domain reconstruction methods can hardly overcome 0.35 and 0.41 respectively.

Original languageEnglish
Title of host publication2023 IEEE International Symposium on Antennas and Propagation and USNC-URSI Radio Science Meeting, AP-S/URSI 2023 - Proceedings
Pages1873-1874
Number of pages2
ISBN (Electronic)9781665442282
DOIs
StatePublished - 2023
Event2023 IEEE International Symposium on Antennas and Propagation and USNC-URSI Radio Science Meeting, AP-S/URSI 2023 - Portland, United States
Duration: 23 Jul 202328 Jul 2023

Publication series

NameIEEE Antennas and Propagation Society, AP-S International Symposium (Digest)
Volume2023-July
ISSN (Print)1522-3965

Conference

Conference2023 IEEE International Symposium on Antennas and Propagation and USNC-URSI Radio Science Meeting, AP-S/URSI 2023
Country/TerritoryUnited States
CityPortland
Period23/07/2328/07/23

Fingerprint

Dive into the research topics of 'Deep Learning for Tumor Margin Identification in Electromagnetic Imaging'. Together they form a unique fingerprint.

Cite this