Rapid 3D nanoscale coherent imaging via physics-aware deep learning
Abstract
Phase retrieval, the problem of recovering lost phase information from measured intensity alone, is an inverse problem that is widely faced in various imaging modalities ranging from astronomy to nanoscale imaging. The current process of phase recovery is iterative in nature. As a result, the image formation is time consuming and computationally expensive, precluding real-time imaging. Here, we use 3D nanoscale X-ray imaging as a representative example to develop a deep learning model to address this phase retrieval problem. We introduce 3D-CDI-NN, a deep convolutional neural network and differential programing framework trained to predict 3D structure and strain, solely from input 3D X-ray coherent scattering data. Our networks are designed to be "physics-aware" in multiple aspects; in that the physics of the X-ray scattering process is explicitly enforced in the training of the network, and the training data are drawn from atomistic simulations that are representative of the physics of the material. We further refine the neural network prediction through a physics-based optimization procedure to enable maximum accuracy at lowest computational cost. 3D-CDI-NN can invert a 3D coherent diffraction pattern to real-space structure and strain hundreds of times faster than traditional iterative phase retrieval methods. Our integrated machine learning andmore »
- Authors:
-
- Argonne National Lab. (ANL), Argonne, IL (United States). Center for Nanoscale Materials; Univ. of Illinois, Chicago, IL (United States). Dept. of Mechanical and Industrial Engineering
- SLAC National Accelerator Lab., Menlo Park, CA (United States)
- Northwestern Univ., Evanston, IL (United States). Applied Physics
- Argonne National Lab. (ANL), Argonne, IL (United States). Materials Science Division
- Argonne National Lab. (ANL), Argonne, IL (United States). Advanced Photon Source (APS)
- Argonne National Lab. (ANL), Argonne, IL (United States). Center for Nanoscale Materials; Argonne National Lab. (ANL), Argonne, IL (United States). Advanced Photon Source (APS)
- Publication Date:
- Research Org.:
- SLAC National Accelerator Laboratory (SLAC), Menlo Park, CA (United States); Argonne National Laboratory (ANL), Argonne, IL (United States)
- Sponsoring Org.:
- USDOE Office of Science (SC), Basic Energy Sciences (BES)
- OSTI Identifier:
- 1785971
- Alternate Identifier(s):
- OSTI ID: 1783342; OSTI ID: 1788382
- Grant/Contract Number:
- AC02-76SF00515; 34532; AC02- 06CH11357; 2018-019-N0; Award Number 34532; AC02-06CH11357
- Resource Type:
- Accepted Manuscript
- Journal Name:
- Applied Physics Reviews
- Additional Journal Information:
- Journal Volume: 8; Journal Issue: 2; Journal ID: ISSN 1931-9401
- Publisher:
- American Institute of Physics (AIP)
- Country of Publication:
- United States
- Language:
- English
- Subject:
- 77 NANOSCIENCE AND NANOTECHNOLOGY
Citation Formats
Chan, Henry, Nashed, Youssef G., Kandel, Saugat, Hruszkewycz, Stephan O., Sankaranarayanan, Subramanian S., Harder, Ross J., and Cherukara, Mathew J. Rapid 3D nanoscale coherent imaging via physics-aware deep learning. United States: N. p., 2021.
Web. doi:10.1063/5.0031486.
Chan, Henry, Nashed, Youssef G., Kandel, Saugat, Hruszkewycz, Stephan O., Sankaranarayanan, Subramanian S., Harder, Ross J., & Cherukara, Mathew J. Rapid 3D nanoscale coherent imaging via physics-aware deep learning. United States. https://doi.org/10.1063/5.0031486
Chan, Henry, Nashed, Youssef G., Kandel, Saugat, Hruszkewycz, Stephan O., Sankaranarayanan, Subramanian S., Harder, Ross J., and Cherukara, Mathew J. Mon .
"Rapid 3D nanoscale coherent imaging via physics-aware deep learning". United States. https://doi.org/10.1063/5.0031486. https://www.osti.gov/servlets/purl/1785971.
@article{osti_1785971,
title = {Rapid 3D nanoscale coherent imaging via physics-aware deep learning},
author = {Chan, Henry and Nashed, Youssef G. and Kandel, Saugat and Hruszkewycz, Stephan O. and Sankaranarayanan, Subramanian S. and Harder, Ross J. and Cherukara, Mathew J.},
abstractNote = {Phase retrieval, the problem of recovering lost phase information from measured intensity alone, is an inverse problem that is widely faced in various imaging modalities ranging from astronomy to nanoscale imaging. The current process of phase recovery is iterative in nature. As a result, the image formation is time consuming and computationally expensive, precluding real-time imaging. Here, we use 3D nanoscale X-ray imaging as a representative example to develop a deep learning model to address this phase retrieval problem. We introduce 3D-CDI-NN, a deep convolutional neural network and differential programing framework trained to predict 3D structure and strain, solely from input 3D X-ray coherent scattering data. Our networks are designed to be "physics-aware" in multiple aspects; in that the physics of the X-ray scattering process is explicitly enforced in the training of the network, and the training data are drawn from atomistic simulations that are representative of the physics of the material. We further refine the neural network prediction through a physics-based optimization procedure to enable maximum accuracy at lowest computational cost. 3D-CDI-NN can invert a 3D coherent diffraction pattern to real-space structure and strain hundreds of times faster than traditional iterative phase retrieval methods. Our integrated machine learning and differential programing solution to the phase retrieval problem is broadly applicable across inverse problems in other application areas.},
doi = {10.1063/5.0031486},
journal = {Applied Physics Reviews},
number = 2,
volume = 8,
place = {United States},
year = {Mon May 17 00:00:00 EDT 2021},
month = {Mon May 17 00:00:00 EDT 2021}
}
Works referenced in this record:
Three-dimensional imaging of strain in a single ZnO nanorod
journal, December 2009
- Newton, Marcus C.; Leake, Steven J.; Harder, Ross
- Nature Materials, Vol. 9, Issue 2
Phase retrieval algorithm for JWST Flight and Testbed Telescope
conference, July 2006
- Dean, Bruce H.; Aronstein, David L.; Smith, J. Scott
- SPIE Astronomical Telescopes + Instrumentation, SPIE Proceedings
Topological defect dynamics in operando battery nanoparticles
journal, June 2015
- Ulvestad, A.; Singer, A.; Clark, J. N.
- Science, Vol. 348, Issue 6241
Three-dimensional X-ray diffraction imaging of dislocations in polycrystalline metals under tensile loading
journal, September 2018
- Cherukara, Mathew J.; Pokharel, Reeju; O’Leary, Timothy S.
- Nature Communications, Vol. 9, Issue 1
Association Between Surgical Skin Markings in Dermoscopic Images and Diagnostic Performance of a Deep Learning Convolutional Neural Network for Melanoma Recognition
journal, October 2019
- Winkler, Julia K.; Fink, Christine; Toberer, Ferdinand
- JAMA Dermatology, Vol. 155, Issue 10
Using automatic differentiation as a general framework for ptychographic reconstruction
journal, January 2019
- Kandel, Saugat; Maddali, S.; Allain, Marc
- Optics Express, Vol. 27, Issue 13
Beyond crystallography: Diffractive imaging using coherent x-ray light sources
journal, April 2015
- Miao, J.; Ishikawa, T.; Robinson, I. K.
- Science, Vol. 348, Issue 6234
AI-enabled high-resolution scanning coherent diffraction imaging
journal, July 2020
- Cherukara, Mathew J.; Zhou, Tao; Nashed, Youssef
- Applied Physics Letters, Vol. 117, Issue 4
Phase recovery and holographic image reconstruction using deep learning in neural networks
journal, October 2017
- Rivenson, Yair; Zhang, Yibo; Günaydın, Harun
- Light: Science & Applications, Vol. 7, Issue 2
Bragg coherent diffractive imaging of single-grain defect dynamics in polycrystalline films
journal, May 2017
- Yau, Allison; Cha, Wonsuk; Kanan, Matthew W.
- Science, Vol. 356, Issue 6339
Ultrafast Three-Dimensional X-ray Imaging of Deformation Modes in ZnO Nanocrystals
journal, January 2017
- Cherukara, Mathew J.; Sasikumar, Kiran; Cha, Wonsuk
- Nano Letters, Vol. 17, Issue 2
Fast Parallel Algorithms for Short-Range Molecular Dynamics
journal, March 1995
- Plimpton, Steve
- Journal of Computational Physics, Vol. 117, Issue 1
Image reconstruction by domain-transform manifold learning
journal, March 2018
- Zhu, Bo; Liu, Jeremiah Z.; Cauley, Stephen F.
- Nature, Vol. 555, Issue 7697
Phase retrieval algorithms: a comparison
journal, January 1982
- Fienup, J. R.
- Applied Optics, Vol. 21, Issue 15
Three-Dimensional Study of the Vector Potential of Magnetic Structures
journal, June 2010
- Phatak, Charudatta; Petford-Long, Amanda K.; De Graef, Marc
- Physical Review Letters, Vol. 104, Issue 25
3D lattice distortions and defect structures in ion-implanted nano-crystals
journal, April 2017
- Hofmann, Felix; Tarleton, Edmund; Harder, Ross J.
- Scientific Reports, Vol. 7, Issue 1
Complex imaging of phase domains by deep neural networks
journal, January 2021
- Wu, Longlong; Juhas, Pavol; Yoo, Shinjae
- IUCrJ, Vol. 8, Issue 1
Sparsity-based single-shot subwavelength coherent diffractive imaging
journal, April 2012
- Szameit, A.; Shechtman, Y.; Osherovich, E.
- Nature Materials, Vol. 11, Issue 5
Identifying Defects with Guided Algorithms in Bragg Coherent Diffractive Imaging
journal, August 2017
- Ulvestad, A.; Nashed, Y.; Beutier, G.
- Scientific Reports, Vol. 7, Issue 1
High-resolution three-dimensional structural microscopy by single-angle Bragg ptychography
journal, November 2016
- Hruszkewycz, S. O.; Allain, M.; Holt, M. V.
- Nature Materials, Vol. 16, Issue 2
In Situ 3D Imaging of Catalysis Induced Strain in Gold Nanoparticles
journal, July 2016
- Ulvestad, Andrew; Sasikumar, Kiran; Kim, Jong Woo
- The Journal of Physical Chemistry Letters, Vol. 7, Issue 15
Active site localization of methane oxidation on Pt nanocrystals
journal, August 2018
- Kim, Dongjin; Chung, Myungwoo; Carnis, Jerome
- Nature Communications, Vol. 9, Issue 1
Ultrafast Three-Dimensional Imaging of Lattice Dynamics in Individual Gold Nanocrystals
journal, May 2013
- Clark, J. N.; Beitra, L.; Xiong, G.
- Science, Vol. 341, Issue 6141
On resizing images in the dct domain
conference, January 2004
- Salazar, C. L.; Tran, T. D.
- 2004 International Conference on Image Processing, 2004. ICIP '04.
Real-time coherent diffraction inversion using deep generative networks
journal, November 2018
- Cherukara, Mathew J.; Nashed, Youssef S. G.; Harder, Ross J.
- Scientific Reports, Vol. 8, Issue 1
Coherent lensless X-ray imaging
journal, November 2010
- Chapman, Henry N.; Nugent, Keith A.
- Nature Photonics, Vol. 4, Issue 12
Ultrafast Three-Dimensional Integrated Imaging of Strain in Core/Shell Semiconductor/Metal Nanostructures
journal, November 2017
- Cherukara, Mathew J.; Sasikumar, Kiran; DiChiara, Anthony
- Nano Letters, Vol. 17, Issue 12
Training Deep Neural Networks for the Inverse Design of Nanophotonic Structures
journal, January 2018
- Liu, Dianjing; Tan, Yixuan; Khoram, Erfan
- ACS Photonics, Vol. 5, Issue 4
X-ray image reconstruction from a diffraction pattern alone
journal, October 2003
- Marchesini, S.; He, H.; Chapman, H. N.
- Physical Review B, Vol. 68, Issue 14