Decoding crystallography from high-resolution electron imaging and diffraction datasets with deep learning

J. A. Aguiar; M. L. Gong; R. R. Unocic; T. Tasdizen; B. D. Miller

doi:10.1126/sciadv.aaw1949

Decoding crystallography from high-resolution electron imaging and diffraction datasets with deep learning

J. A. Aguiar, M. L. Gong, R. R. Unocic, T. Tasdizen, B. D. Miller

Materials MicroAnalysis

Research output: Contribution to journal › Article › peer-review

90 Scopus citations

Abstract

While machine learning has been making enormous strides in many technical areas, it is still massively underused in transmission electron microscopy. To address this, a convolutional neural network model was developed for reliable classification of crystal structures from small numbers of electron images and diffraction patterns with no preferred orientation. Diffraction data containing 571,340 individual crystals divided among seven families, 32 genera, and 230 space groups were used to train the network. Despite the highly imbalanced dataset, the network narrows down the space groups to the top two with over 70% confidence in the worst case and up to 95% in the common cases. As examples, we benchmarked against alloys to two-dimensional materials to cross-validate our deep-learning model against high-resolution transmission electron images and diffraction patterns. We present this result both as a research tool and deep-learning application for diffraction analysis.

Original language	English
Article number	eaaw1949
Journal	Science Advances
Volume	5
Issue number	10
DOIs	https://doi.org/10.1126/sciadv.aaw1949
State	Published - Oct 30 2019

Funding

We thank I. Harvey for the many useful discussions and contributions to this work. M. Patel and H. Yoon are acknowledged for providing samples and data for analysis. A portion of this research was conducted at the Center for Nanophase Materials Sciences, which is a DOE Office of Science User Facility (R.R.U.). We also acknowledge D. Masiel, B. Reed, K. Jungjohann, S. Misra, J. Gu, and R. Mariani for the helpful discussions. This work was supported through the INL Laboratory Directed Research and Development (LDRD) Program under DOE Idaho Operations Office Contract DE-AC07-05ID145142.

Funders	Funder number
DOE Idaho Operations Office	DE-AC07-05ID145142
DOE Office of Science user facility
Laboratory Directed Research and Development

Access to Document

10.1126/sciadv.aaw1949

Cite this

@article{8a40419bc8944614a09246079b95a2a9,

title = "Decoding crystallography from high-resolution electron imaging and diffraction datasets with deep learning",

abstract = "While machine learning has been making enormous strides in many technical areas, it is still massively underused in transmission electron microscopy. To address this, a convolutional neural network model was developed for reliable classification of crystal structures from small numbers of electron images and diffraction patterns with no preferred orientation. Diffraction data containing 571,340 individual crystals divided among seven families, 32 genera, and 230 space groups were used to train the network. Despite the highly imbalanced dataset, the network narrows down the space groups to the top two with over 70% confidence in the worst case and up to 95% in the common cases. As examples, we benchmarked against alloys to two-dimensional materials to cross-validate our deep-learning model against high-resolution transmission electron images and diffraction patterns. We present this result both as a research tool and deep-learning application for diffraction analysis.",

author = "Aguiar, {J. A.} and Gong, {M. L.} and Unocic, {R. R.} and T. Tasdizen and Miller, {B. D.}",

note = "Publisher Copyright: Copyright {\textcopyright} 2019 The Authors,",

year = "2019",

month = oct,

day = "30",

doi = "10.1126/sciadv.aaw1949",

language = "English",

volume = "5",

journal = "Science Advances",

issn = "2375-2548",

publisher = "American Association for the Advancement of Science",

number = "10",

}

TY - JOUR

T1 - Decoding crystallography from high-resolution electron imaging and diffraction datasets with deep learning

AU - Aguiar, J. A.

AU - Gong, M. L.

AU - Unocic, R. R.

AU - Tasdizen, T.

AU - Miller, B. D.

PY - 2019/10/30

Y1 - 2019/10/30

N2 - While machine learning has been making enormous strides in many technical areas, it is still massively underused in transmission electron microscopy. To address this, a convolutional neural network model was developed for reliable classification of crystal structures from small numbers of electron images and diffraction patterns with no preferred orientation. Diffraction data containing 571,340 individual crystals divided among seven families, 32 genera, and 230 space groups were used to train the network. Despite the highly imbalanced dataset, the network narrows down the space groups to the top two with over 70% confidence in the worst case and up to 95% in the common cases. As examples, we benchmarked against alloys to two-dimensional materials to cross-validate our deep-learning model against high-resolution transmission electron images and diffraction patterns. We present this result both as a research tool and deep-learning application for diffraction analysis.

AB - While machine learning has been making enormous strides in many technical areas, it is still massively underused in transmission electron microscopy. To address this, a convolutional neural network model was developed for reliable classification of crystal structures from small numbers of electron images and diffraction patterns with no preferred orientation. Diffraction data containing 571,340 individual crystals divided among seven families, 32 genera, and 230 space groups were used to train the network. Despite the highly imbalanced dataset, the network narrows down the space groups to the top two with over 70% confidence in the worst case and up to 95% in the common cases. As examples, we benchmarked against alloys to two-dimensional materials to cross-validate our deep-learning model against high-resolution transmission electron images and diffraction patterns. We present this result both as a research tool and deep-learning application for diffraction analysis.

UR - http://www.scopus.com/inward/record.url?scp=85074600573&partnerID=8YFLogxK

U2 - 10.1126/sciadv.aaw1949

DO - 10.1126/sciadv.aaw1949

M3 - Article

C2 - 31976364

AN - SCOPUS:85074600573

SN - 2375-2548

VL - 5

JO - Science Advances

JF - Science Advances

IS - 10

M1 - eaaw1949

ER -

Decoding crystallography from high-resolution electron imaging and diffraction datasets with deep learning

Abstract

Funding

Access to Document

Other files and links

Fingerprint

Cite this