skip to main content
DOE PAGES title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Low-dose x-ray tomography through a deep convolutional neural network

Abstract

Synchrotron-based X-ray tomography offers the potential of rapid large-scale reconstructions of the interiors of materials and biological tissue at fine resolution. However, for radiation sensitive samples, there remain fundamental trade-offs between damaging samples during longer acquisition times and reducing signals with shorter acquisition times. We present a deep convolutional neural network (CNN) method that increases the acquired X-ray tomographic signal by at least a factor of 10 during low-dose fast acquisition by improving the quality of recorded projections. Short exposure time projections enhanced with CNN show similar signal to noise ratios as compared with long exposure time projections and much lower noise and more structural information than low-dose fats acquisition without CNN. We optimized this approach using simulated samples and further validated on experimental nano-computed tomography data of radiation sensitive mouse brains acquired with a transmission X-ray microscopy. We demonstrate that automated algorithms can reliably trace brain structures in datasets collected with low dose-CNN. As a result, this method can be applied to other tomographic or scanning based X-ray imaging techniques and has great potential for studying faster dynamics in specimens.

Authors:
 [1];  [1];  [1];  [2];  [3];  [1]; ORCiD logo [4]
  1. Argonne National Lab. (ANL), Lemont, IL (United States)
  2. Georgia Institute of Technology & Emory Univ., Atlanta, GA (United States)
  3. Argonne National Lab. (ANL), Lemont, IL (United States); Univ. of Chicago, Chicago, IL (United States)
  4. Argonne National Lab. (ANL), Lemont, IL (United States); Northwestern Univ., Evanston, IL (United States)
Publication Date:
Research Org.:
Argonne National Lab. (ANL), Argonne, IL (United States)
Sponsoring Org.:
USDOE Office of Science (SC), Basic Energy Sciences (BES) (SC-22)
OSTI Identifier:
1421983
Grant/Contract Number:  
AC02-06CH11357
Resource Type:
Accepted Manuscript
Journal Name:
Scientific Reports
Additional Journal Information:
Journal Volume: 8; Journal Issue: 1; Journal ID: ISSN 2045-2322
Publisher:
Nature Publishing Group
Country of Publication:
United States
Language:
English
Subject:
60 APPLIED LIFE SCIENCES

Citation Formats

Yang, Xiaogang, De Andrade, Vincent, Scullin, William, Dyer, Eva L., Kasthuri, Narayanan, De Carlo, Francesco, and Gursoy, Doga. Low-dose x-ray tomography through a deep convolutional neural network. United States: N. p., 2018. Web. doi:10.1038/s41598-018-19426-7.
Yang, Xiaogang, De Andrade, Vincent, Scullin, William, Dyer, Eva L., Kasthuri, Narayanan, De Carlo, Francesco, & Gursoy, Doga. Low-dose x-ray tomography through a deep convolutional neural network. United States. doi:10.1038/s41598-018-19426-7.
Yang, Xiaogang, De Andrade, Vincent, Scullin, William, Dyer, Eva L., Kasthuri, Narayanan, De Carlo, Francesco, and Gursoy, Doga. Wed . "Low-dose x-ray tomography through a deep convolutional neural network". United States. doi:10.1038/s41598-018-19426-7. https://www.osti.gov/servlets/purl/1421983.
@article{osti_1421983,
title = {Low-dose x-ray tomography through a deep convolutional neural network},
author = {Yang, Xiaogang and De Andrade, Vincent and Scullin, William and Dyer, Eva L. and Kasthuri, Narayanan and De Carlo, Francesco and Gursoy, Doga},
abstractNote = {Synchrotron-based X-ray tomography offers the potential of rapid large-scale reconstructions of the interiors of materials and biological tissue at fine resolution. However, for radiation sensitive samples, there remain fundamental trade-offs between damaging samples during longer acquisition times and reducing signals with shorter acquisition times. We present a deep convolutional neural network (CNN) method that increases the acquired X-ray tomographic signal by at least a factor of 10 during low-dose fast acquisition by improving the quality of recorded projections. Short exposure time projections enhanced with CNN show similar signal to noise ratios as compared with long exposure time projections and much lower noise and more structural information than low-dose fats acquisition without CNN. We optimized this approach using simulated samples and further validated on experimental nano-computed tomography data of radiation sensitive mouse brains acquired with a transmission X-ray microscopy. We demonstrate that automated algorithms can reliably trace brain structures in datasets collected with low dose-CNN. As a result, this method can be applied to other tomographic or scanning based X-ray imaging techniques and has great potential for studying faster dynamics in specimens.},
doi = {10.1038/s41598-018-19426-7},
journal = {Scientific Reports},
number = 1,
volume = 8,
place = {United States},
year = {2018},
month = {2}
}

Journal Article:
Free Publicly Available Full Text
Publisher's Version of Record

Citation Metrics:
Cited by: 9 works
Citation information provided by
Web of Science

Save / Share:

Works referenced in this record:

On the determination of functions from their integral values along certain manifolds
journal, December 1986


TomoPy: a framework for the analysis of synchrotron tomographic data
journal, August 2014

  • Gürsoy, Dogˇa; De Carlo, Francesco; Xiao, Xianghui
  • Journal of Synchrotron Radiation, Vol. 21, Issue 5
  • DOI: 10.1107/S1600577514013939

High-resolution, low-dose phase contrast X-ray tomography for 3D diagnosis of human breast cancers
journal, October 2012

  • Zhao, Y.; Brun, E.; Coan, P.
  • Proceedings of the National Academy of Sciences, Vol. 109, Issue 45
  • DOI: 10.1073/pnas.1204460109

Principles of CT: Radiation Dose and Image Quality
journal, December 2007


Image Super-Resolution Using Deep Convolutional Networks
journal, February 2016

  • Dong, Chao; Loy, Chen Change; He, Kaiming
  • IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 38, Issue 2
  • DOI: 10.1109/TPAMI.2015.2439281

Deep learning
journal, May 2015

  • LeCun, Yann; Bengio, Yoshua; Hinton, Geoffrey
  • Nature, Vol. 521, Issue 7553
  • DOI: 10.1038/nature14539

Incorporating image priors with deep convolutional neural networks for image super-resolution
journal, June 2016


High-contrast en bloc staining of neuronal tissue for field emission scanning electron microscopy
journal, January 2012

  • Tapia, Juan Carlos; Kasthuri, Narayanan; Hayworth, Kenneth J.
  • Nature Protocols, Vol. 7, Issue 2
  • DOI: 10.1038/nprot.2011.439

Algorithm-Enabled Low-Dose Micro-CT Imaging
journal, March 2011

  • Han, Xiao; Bian, Junguo; Eaker, Diane R.
  • IEEE Transactions on Medical Imaging, Vol. 30, Issue 3
  • DOI: 10.1109/tmi.2010.2089695

Practical, Unified, Motion and Missing Data Treatment in Degraded Video
journal, January 2004


The ID21 X-ray and infrared microscopy beamline at the ESRF: status and recent applications to artistic materials
journal, January 2017

  • Cotte, Marine; Pouyet, Emeline; Salomé, Murielle
  • Journal of Analytical Atomic Spectrometry, Vol. 32, Issue 3
  • DOI: 10.1039/c6ja00356g

Computerized transverse axial scanning (tomography): Part 1. Description of system
journal, December 1973


Fast Tomographic Reconstruction From Limited Data Using Artificial Neural Networks
journal, December 2013

  • Pelt, Daniel Maria; Batenburg, Kees Joost
  • IEEE Transactions on Image Processing, Vol. 22, Issue 12
  • DOI: 10.1109/TIP.2013.2283142

Noise reduction with low dose CT data based on a modified ROF model
journal, January 2012

  • Zhu, Yining; Zhao, Mengliu; Zhao, Yunsong
  • Optics Express, Vol. 20, Issue 16
  • DOI: 10.1364/oe.20.017987

A convolutional neural network approach to calibrating the rotation axis for X-ray computed tomography
journal, January 2017

  • Yang, Xiaogang; De Carlo, Francesco; Phatak, Charudatta
  • Journal of Synchrotron Radiation, Vol. 24, Issue 2
  • DOI: 10.1107/s1600577516020117

An assessment of the resolution limitation due to radiation-damage in X-ray diffraction microscopy
journal, March 2009

  • Howells, M. R.; Beetz, T.; Chapman, H. N.
  • Journal of Electron Spectroscopy and Related Phenomena, Vol. 170, Issue 1-3
  • DOI: 10.1016/j.elspec.2008.10.008

Scientific data exchange: a schema for HDF5-based storage of raw and analyzed data
journal, October 2014

  • De Carlo, Francesco; Gürsoy, Dogˇa; Marone, Federica
  • Journal of Synchrotron Radiation, Vol. 21, Issue 6
  • DOI: 10.1107/S160057751401604X

Localization and Classification of Paddy Field Pests using a Saliency Map and Deep Convolutional Neural Network
journal, February 2016

  • Liu, Ziyi; Gao, Junfeng; Yang, Guoguo
  • Scientific Reports, Vol. 6, Issue 1
  • DOI: 10.1038/srep20410

Applications of nonlocal means algorithm in low-dose X-ray CT image processing and reconstruction: A review
journal, March 2017

  • Zhang, Hao; Zeng, Dong; Zhang, Hua
  • Medical Physics, Vol. 44, Issue 3
  • DOI: 10.1002/mp.12097

Image Quality Assessment: From Error Visibility to Structural Similarity
journal, April 2004

  • Wang, Z.; Bovik, A. C.; Sheikh, H. R.
  • IEEE Transactions on Image Processing, Vol. 13, Issue 4
  • DOI: 10.1109/TIP.2003.819861

Developments in synchrotron x-ray computed microtomography at the National Synchrotron Light Source
conference, September 1999

  • Dowd, Betsy A.; Campbell, Graham H.; Marr, Robert B.
  • SPIE's International Symposium on Optical Science, Engineering, and Instrumentation, SPIE Proceedings
  • DOI: 10.1117/12.363725

Exposure strategies for polymethyl methacrylate from in situ x-ray absorption near edge structure spectroscopy
journal, July 1995

  • Zhang, X.
  • Journal of Vacuum Science & Technology B: Microelectronics and Nanometer Structures, Vol. 13, Issue 4
  • DOI: 10.1116/1.588175

A new transmission x-ray microscope for in-situ nano-tomography at the APS (Conference Presentation)
conference, November 2016

  • De Andrade, Vincent; Deriy, Alex; Wojcik, Michael
  • Developments in X-Ray Tomography X
  • DOI: 10.1117/12.2239449

Deep convolutional architecture for natural image denoising
conference, October 2015

  • Wang, Xuejiao; Tao, Qiuyan; Wang, Lianghao
  • 2015 International Conference on Wireless Communications & Signal Processing (WCSP)
  • DOI: 10.1109/WCSP.2015.7341021

aLow-dose CT via convolutional neural network
journal, January 2017

  • Chen, Hu; Zhang, Yi; Zhang, Weihua
  • Biomedical Optics Express, Vol. 8, Issue 2
  • DOI: 10.1364/BOE.8.000679

Deconvolutional networks
conference, June 2010

  • Zeiler, Matthew D.; Krishnan, Dilip; Taylor, Graham W.
  • 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition
  • DOI: 10.1109/CVPR.2010.5539957

Improving deep neural networks for LVCSR using rectified linear units and dropout
conference, May 2013

  • Dahl, George E.; Sainath, Tara N.; Hinton, Geoffrey E.
  • ICASSP 2013 - 2013 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
  • DOI: 10.1109/ICASSP.2013.6639346

Nanoscale 3D imaging at the Advanced Photon Source
journal, May 2016


User-guided 3D active contour segmentation of anatomical structures: Significantly improved efficiency and reliability
journal, July 2006


The ID21 Scanning X-ray Microscope at ESRF
journal, March 2013


The Bionanoprobe: hard X-ray fluorescence nanoprobe with cryogenic capabilities
journal, December 2013


Denoising Convolutional Neural Network
conference, August 2015

  • Xu, Qingyang; Zhang, Chengjin; Zhang, Li
  • 2015 IEEE International Conference on Information and Automation (ICIA)
  • DOI: 10.1109/ICInfA.2015.7279466

A Fast Spectral Method for Active 3D Shape Reconstruction
journal, January 2004


Biomimetic antimicrobial cloak by graphene-oxide agar hydrogel
journal, December 2016

  • Papi, Massimiliano; Palmieri, Valentina; Bugli, Francesca
  • Scientific Reports, Vol. 6, Issue 1
  • DOI: 10.1038/s41598-016-0010-7

    Works referencing / citing this record:

    Prospects of deep learning for medical imaging
    journal, June 2018

    • Kim, Jonghoon; Hong, Jisu; Park, Hyunjin
    • Precision and Future Medicine, Vol. 2, Issue 2
    • DOI: 10.23838/pfm.2018.00030