Deep learning based coronary artery motion artifact compensation using style-transfer synthesis in ct images

Sunghee Jung, Soochahn Lee, Byunghwan Jeon, Yeonggul Jang, Hyuk Jae Chang

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

2 Citations (Scopus)

Abstract

Motion artifact compensation of the coronary artery in computed tomography (CT) is required to quantify the risk of coronary artery disease more accurately. We present a novel method based on deep learning for motion artifact compensation in coronary CT angiography (CCTA). The ground-truth, i.e., coronary artery without motion, was synthesized using full-phase four-dimensional (4D) CT by applying style-transfer method because it is medically impossible to obtain in practice. The network for motion artifact compensation based on very deep convolutional neural network (CNN) is trained using the synthesized ground-truth. An observer study was performed for the evaluation of the proposed method. The motion artifacts were markedly reduced and boundaries of the coronary artery were much sharper than before applying the proposed method, with a strong inter-observer agreement (kappa = 0.78).

Original languageEnglish
Title of host publicationSimulation and Synthesis in Medical Imaging - Third International Workshop, SASHIMI 2018, Held in Conjunction with MICCAI 2018, Proceedings
EditorsOrcun Goksel, Ipek Oguz, Ali Gooya, Ninon Burgos
PublisherSpringer Verlag
Pages100-110
Number of pages11
ISBN (Print)9783030005351
DOIs
Publication statusPublished - 2018
Event3rd International Workshop on Simulation and Synthesis in Medical Imaging, SASHIMI 2018 Held in Conjunction with 21st International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2018 - Granada, Spain
Duration: 2018 Sep 162018 Sep 16

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume11037 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

Other3rd International Workshop on Simulation and Synthesis in Medical Imaging, SASHIMI 2018 Held in Conjunction with 21st International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2018
CountrySpain
CityGranada
Period18/9/1618/9/16

All Science Journal Classification (ASJC) codes

  • Theoretical Computer Science
  • Computer Science(all)

Fingerprint Dive into the research topics of 'Deep learning based coronary artery motion artifact compensation using style-transfer synthesis in ct images'. Together they form a unique fingerprint.

  • Cite this

    Jung, S., Lee, S., Jeon, B., Jang, Y., & Chang, H. J. (2018). Deep learning based coronary artery motion artifact compensation using style-transfer synthesis in ct images. In O. Goksel, I. Oguz, A. Gooya, & N. Burgos (Eds.), Simulation and Synthesis in Medical Imaging - Third International Workshop, SASHIMI 2018, Held in Conjunction with MICCAI 2018, Proceedings (pp. 100-110). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 11037 LNCS). Springer Verlag. https://doi.org/10.1007/978-3-030-00536-8_11