Automated correlative segmentation of large Transmission X-ray Microscopy (TXM) tomograms using deep learning

doi:https://doi.org/10.1016/J.MATCHAR.2018.05.053

Title: Automated correlative segmentation of large Transmission X-ray Microscopy (TXM) tomograms using deep learning

Abstract

Highlights: • A unique automated segmentation approach for large 3D nanotomography datasets obtained by Transmission X-ray Microscopy (TXM) in an Al-Cu alloy. • Automated segmentation using a Convolutional Neural Network (CNN) architecture based on a deep learning approach was employed. • This extremely versatile technique is capable of emulating the manual segmentation process effectively. • Quantitative comparison between manual segmentation and CNN architecture established the accuracy of this automated technique. • Its ability to robustly process ultra-large volumes of data in relatively small time frames can exponentially accelerate tomographic data analysis. - Abstract: A unique correlative approach for automated segmentation of large 3D nanotomography datasets obtained using Transmission X-ray Microscopy (TXM) in an Al-Cu alloy has been introduced. Automated segmentation using a Convolutional Neural Network (CNN) architecture based on a deep learning approach was employed. This extremely versatile technique is capable of emulating the manual segmentation process effectively. Coupling this technique with post-scanning SEM imaging ensured precise estimation of 3D morphological parameters from nanotomography. The segmentation process as well as subsequent analysis was expedited by several orders of magnitude. Quantitative comparison between segmentation performed manually and using the CNN architecture established the accuracy of this automated technique. Its ability tomore »« less

Authors:

Shashank Kaira, C. ^[1]; Yang, Xiaogang; De Andrade, Vincent; De Carlo, Francesco ^[2]; Scullin, William ^[3]; Gursoy, Doga ^[2]; Chawla, Nikhilesh ^[1]

Center for 4D Materials Science, Arizona State University, Tempe, AZ 85287-6106 (United States)
Advanced Photon Source, Argonne National Laboratory, Building 401, 9700 S. Cass Avenue, Argonne, IL 60439 (United States)
Argonne Leadership Computing Facility, Argonne National Laboratory, Building 401, 9700 S. Cass Avenue, Argonne, IL 60439 (United States)

Publication Date:: 2018-08-15

OSTI Identifier:: 22805091

Resource Type:: Journal Article

Journal Name:: Materials Characterization

Additional Journal Information:: Journal Volume: 142; Other Information: Copyright (c) 2017 Elsevier Science B.V., Amsterdam, The Netherlands, All rights reserved.; Country of input: International Atomic Energy Agency (IAEA); Journal ID: ISSN 1044-5803

Country of Publication:: United States

Language:: English

Subject:: 36 MATERIALS SCIENCE; ALUMINIUM ALLOYS; COMPUTERIZED SIMULATION; COMPUTERIZED TOMOGRAPHY; DATA ANALYSIS; NEURAL NETWORKS; PRECIPITATION; SCANNING ELECTRON MICROSCOPY; X RADIATION

Citation Formats


                    Shashank Kaira, C., Yang, Xiaogang, De Andrade, Vincent, De Carlo, Francesco, Scullin, William, Gursoy, Doga, and Chawla, Nikhilesh. Automated correlative segmentation of large Transmission X-ray Microscopy (TXM) tomograms using deep learning.  United States: N. p., 2018. 
        Web.  doi:10.1016/J.MATCHAR.2018.05.053.

Copy to clipboard


                    Shashank Kaira, C., Yang, Xiaogang, De Andrade, Vincent, De Carlo, Francesco, Scullin, William, Gursoy, Doga, & Chawla, Nikhilesh. Automated correlative segmentation of large Transmission X-ray Microscopy (TXM) tomograms using deep learning.  United States.  https://doi.org/10.1016/J.MATCHAR.2018.05.053

Copy to clipboard


                    Shashank Kaira, C., Yang, Xiaogang, De Andrade, Vincent, De Carlo, Francesco, Scullin, William, Gursoy, Doga, and Chawla, Nikhilesh. Wed .  
        "Automated correlative segmentation of large Transmission X-ray Microscopy (TXM) tomograms using deep learning".  United States.  https://doi.org/10.1016/J.MATCHAR.2018.05.053.

Copy to clipboard


                    
@article{osti_22805091,

  title        = {Automated correlative segmentation of large Transmission X-ray Microscopy (TXM) tomograms using deep learning},

  author       = {Shashank Kaira, C. and Yang, Xiaogang and De Andrade, Vincent and De Carlo, Francesco and Scullin, William and Gursoy, Doga and Chawla, Nikhilesh},

  abstractNote = {Highlights: • A unique automated segmentation approach for large 3D nanotomography datasets obtained by Transmission X-ray Microscopy (TXM) in an Al-Cu alloy. • Automated segmentation using a Convolutional Neural Network (CNN) architecture based on a deep learning approach was employed. • This extremely versatile technique is capable of emulating the manual segmentation process effectively. • Quantitative comparison between manual segmentation and CNN architecture established the accuracy of this automated technique. • Its ability to robustly process ultra-large volumes of data in relatively small time frames can exponentially accelerate tomographic data analysis. - Abstract: A unique correlative approach for automated segmentation of large 3D nanotomography datasets obtained using Transmission X-ray Microscopy (TXM) in an Al-Cu alloy has been introduced. Automated segmentation using a Convolutional Neural Network (CNN) architecture based on a deep learning approach was employed. This extremely versatile technique is capable of emulating the manual segmentation process effectively. Coupling this technique with post-scanning SEM imaging ensured precise estimation of 3D morphological parameters from nanotomography. The segmentation process as well as subsequent analysis was expedited by several orders of magnitude. Quantitative comparison between segmentation performed manually and using the CNN architecture established the accuracy of this automated technique. Its ability to robustly process ultra-large volumes of data in relatively small time frames can exponentially accelerate tomographic data analysis, possibly opening up novel avenues for performing 4D characterization experiments with finer time steps.},

  doi          = {10.1016/J.MATCHAR.2018.05.053},

  url          = {https://www.osti.gov/biblio/22805091},
  journal      = {Materials Characterization},

  issn         = {1044-5803},

  number       = ,

  volume       = 142,

  place        = {United States},

  year         = {2018},

  month        = {8}

}

Copy to clipboard

Journal Article:

https://doi.org/10.1016/J.MATCHAR.2018.05.053

Other availability

Find in Google Scholar

Search WorldCat to find libraries that may hold this journal

Save / Share:

Export Metadata

Save to My Library

Works referencing / citing this record:

Semantic segmentation of synchrotron tomography of multiphase Al-Si alloys using a convolutional neural network with a pixel-wise weighted loss function
journal, December 2019

Strohmann, Tobias; Bugelnig, Katrin; Breitbarth, Eric
Scientific Reports, Vol. 9, Issue 1
https://doi.org/10.1038/s41598-019-56008-7

Automatic projection image registration for nanoscale X-ray tomographic reconstruction
journal, October 2018

Yu, Haiyan; Xia, Sihao; Wei, Chenxi
Journal of Synchrotron Radiation, Vol. 25, Issue 6
https://doi.org/10.1107/s1600577518013929

Tomographic reconstruction with a generative adversarial network
journal, February 2020

Yang, Xiaogang; Kahnt, Maik; Brückner, Dennis
Journal of Synchrotron Radiation, Vol. 27, Issue 2
https://doi.org/10.1107/s1600577520000831

SWIRL: High-performance many-core CPU code generation for deep neural networks
journal, May 2019

Venkat, Anand; Rusira, Tharindu; Barik, Raj
The International Journal of High Performance Computing Applications, Vol. 33, Issue 6
https://doi.org/10.1177/1094342019866247

Similar records in OSTI.GOV collections:

Automated correlative segmentation of large Transmission X-ray Microscopy (TXM) tomograms using deep learning

Journal Article Shashank Kaira, C. ; Yang, Xiaogang ; De Andrade, Vincent ; ... - Materials Characterization

A unique correlative approach for automated segmentation of large 3D nanotomography datasets obtained using Transmission X-ray Microscopy (TXM) in an Al-Cu alloy has been introduced. Automated segmentation using a Convolutional Neural Network (CNN) architecture based on a deep learning approach was employed. This extremely versatile technique is capable of emulating the manual segmentation process effectively. Coupling this technique with post-scanning SEM imaging ensured precise estimation of 3D morphological parameters from nanotomography. The segmentation process as well as subsequent analysis was expedited by several orders of magnitude. Quantitative comparison between segmentation performed manually and using the CNN architecture established the accuracymore »« less
Cited by 4
https://doi.org/10.1016/j.matchar.2018.05.053

Full Text Available
MABAL: a Novel Deep-Learning Architecture for Machine-Assisted Bone Age Labeling

Journal Article Mutasa, Simukayi ; Chang, Peter D. ; Ruzal-Shapiro, Carrie ; ... - Journal of Digital Imaging (Online)

Bone age assessment (BAA) is a commonly performed diagnostic study in pediatric radiology to assess skeletal maturity. The most commonly utilized method for assessment of BAA is the Greulich and Pyle method (Pediatr Radiol 46.9:1269–1274, 2016; Arch Dis Child 81.2:172–173, 1999) atlas. The evaluation of BAA can be a tedious and time-consuming process for the radiologist. As such, several computer-assisted detection/diagnosis (CAD) methods have been proposed for automation of BAA. Classical CAD tools have traditionally relied on hard-coded algorithmic features for BAA which suffer from a variety of drawbacks. Recently, the advent and proliferation of convolutional neural networks (CNNs) hasmore »« less
https://doi.org/10.1007/S10278-018-0053-3
A Two-Tier Convolutional Neural Network for Combined Detection and Segmentation in Biological Imagery

Conference Ziabari, Amir ; Shirinifard, Abbas ; Eicholtz, Matthew ; ...

Deep learning techniques have been useful for modern microscopy imaging techniques to further study and analyze biological structures and organs. Convolutional neural networks (CNN) have improved 2D object detection, localization, and segmentation. For imagery containing biological structures with depth, it is especially desirable to perform these tasks in 3D. Traditionally, performing these tasks simultaneously in 3D has proven to be computationally expensive. Currently available methodologies thus largely work to segment 3D objects from 2D images (without context from captured 3D volumes). In this work, we present a novel approach to perform fast and accurate localization, detection, and segmentation of volumesmore »« less
https://doi.org/10.1109/GlobalSIP45357.2019.8969303

Full Text Available
Low-dose x-ray tomography through a deep convolutional neural network

Journal Article Yang, Xiaogang ; De Andrade, Vincent ; Scullin, William ; ... - Scientific Reports

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 muchmore »« less
Cited by 9
https://doi.org/10.1038/s41598-018-19426-7

Full Text Available
Computationally Efficient Learning of Quality Controlled Word Embeddings for Natural Language Processing

Conference Alawad, Mohammed M. ; Tourassi, Georgia

Deep learning (DL) has been used for many natural language processing (NLP) tasks due to its superior performance as compared to traditional machine learning approaches. In DL models for NLP, words are represented using word embeddings, which capture both semantic and syntactic information in text. However, 90-95% of the DL trainable parameters are associated with the word embeddings, resulting in a large storage or memory footprint. Therefore, reducing the number of word embedding parameters is critical, especially with the increase of vocabulary size. In this work, we propose a novel approximate word embeddings approach for convolutional neural networks (CNNs) usedmore »« less
Full Text Available

Similar Records

Title: Automated correlative segmentation of large Transmission X-ray Microscopy (TXM) tomograms using deep learning

Abstract

Citation Formats

Semantic segmentation of synchrotron tomography of multiphase Al-Si alloys using a convolutional neural network with a pixel-wise weighted loss function journal, December 2019

Automatic projection image registration for nanoscale X-ray tomographic reconstruction journal, October 2018

Tomographic reconstruction with a generative adversarial network journal, February 2020

SWIRL: High-performance many-core CPU code generation for deep neural networks journal, May 2019

Semantic segmentation of synchrotron tomography of multiphase Al-Si alloys using a convolutional neural network with a pixel-wise weighted loss function
journal, December 2019

Automatic projection image registration for nanoscale X-ray tomographic reconstruction
journal, October 2018

Tomographic reconstruction with a generative adversarial network
journal, February 2020

SWIRL: High-performance many-core CPU code generation for deep neural networks
journal, May 2019