Segmentation of tomography datasets using 3D convolutional neural networks

James, Jim; Pruyne, Nathan; Stan, Tiberiu; Schwarting, Marcus; Yeom, Jiwon; Hong, Seungbum; Voorhees, Peter; Blaiszik, Ben; Foster, Ian

doi:10.1016/j.commatsci.2022.111847

Segmentation of tomography datasets using 3D convolutional neural networks

Journal Article · Thu Oct 20 00:00:00 EDT 2022 · Computational Materials Science

DOI:https://doi.org/10.1016/j.commatsci.2022.111847· OSTI ID:2433829

James, Jim ^[1]; Pruyne, Nathan ^[2]; Stan, Tiberiu ^[3]; Schwarting, Marcus ^[4]; Yeom, Jiwon ^[5]; Hong, Seungbum ^[5]; Voorhees, Peter ^[3]; Blaiszik, Ben ^[6]; Foster, Ian ^[6]

Argonne National Laboratory (ANL), Argonne, IL (United States)
Argonne National Laboratory (ANL), Argonne, IL (United States); Northwestern Univ., Evanston, IL (United States)
Northwestern Univ., Evanston, IL (United States)
Univ. of Chicago, IL (United States)
Korea Advanced Inst. Science and Technology (KAIST), Daejeon (Korea, Republic of)
Argonne National Laboratory (ANL), Argonne, IL (United States); Univ. of Chicago, IL (United States)

Dendritic microstructures are ubiquitous in nature and are the primary solidification morphologies in metallic materials. Techniques such as X-ray computed tomography (XCT) have provided new insights into dendritic phase transformation phenomena. However, manual identification of dendritic morphologies in microscopy data can be both labor intensive and potentially ambiguous. The analysis of 3D datasets is particularly challenging due to their large sizes (terabytes) and the presence of artifacts scattered within the imaged volumes. Here, in this study, we trained 3D convolutional neural networks (CNNs) to segment 3D datasets. Three CNN architectures were investigated, including a new version of FCDenseNet which we extended to 3D. We show that using hyperparameter optimization (HPO) and fine-tuning techniques, both 2D and 3D CNN architectures outperform the previous state of the art. The 3D U-Net architecture trained in this study produced the best segmentations according to quantitative metrics (intersection-over-union of 95.56% and a boundary displacement error of 0.58 pixels), while 3D FCDense produced the smoothest boundaries and best segmentations according to visual inspection. The trained 3D CNNs are able to segment entire 852 × 852 × 250 voxel 3D volumes in only ~60 s, thus hastening the progress towards a deeper understanding of phase transformation phenomena such as dendritic solidification.

View Accepted Manuscript (DOE)

Research Organization:: Argonne National Laboratory (ANL), Argonne, IL (United States)

Sponsoring Organization:: US Department of Commerce; USDOE Office of Science (SC), Basic Energy Sciences (BES). Scientific User Facilities (SUF)

Grant/Contract Number:: AC02-06CH11357

OSTI ID:: 2433829

Journal Information:: Computational Materials Science, Journal Name: Computational Materials Science Vol. 216; ISSN 0927-0256

Publisher:: ElsevierCopyright Statement

Country of Publication:: United States

Language:: English

References (27)

A Parallel 3D Dendritic Growth Simulator Using the Phase-Field Method George, William L.; Warren, James A. Journal of Computational Physics, Vol. 177, Issue 2 https://doi.org/10.1006/jcph.2002.7005	journal	April 2002
U-Net: Convolutional Networks for Biomedical Image Segmentation Ronneberger, Olaf; Fischer, Philipp; Brox, Thomas Medical Image Computing and Computer-Assisted Intervention – MICCAI 2015: 18th International Conference, Munich, Germany, October 5-9, 2015, Proceedings, Part III https://doi.org/10.1007/978-3-319-24574-4_28	book	November 2015
3D U-Net: Learning Dense Volumetric Segmentation from Sparse Annotation Çiçek, Özgün; Abdulkadir, Ahmed; Lienkamp, Soeren S. Medical Image Computing and Computer-Assisted Intervention – MICCAI 2016 https://doi.org/10.1007/978-3-319-46723-8_49	conference	January 2016
Optimizing Intersection-Over-Union in Deep Neural Networks for Image Segmentation Rahman, Md Atiqur; Wang, Yang Advances in Visual Computing https://doi.org/10.1007/978-3-319-50835-1_22	book	January 2016
Dendrite growth directions in aluminum-zinc alloys Gonzales, F.; Rappaz, M. Metallurgical and Materials Transactions A, Vol. 37, Issue 9 https://doi.org/10.1007/BF02586112	journal	September 2006
The Materials Data Facility: Data Services to Advance Materials Science Research Blaiszik, B.; Chard, K.; Pruyne, J. JOM, Vol. 68, Issue 8 https://doi.org/10.1007/s11837-016-2001-3	journal	July 2016
Material structure-property linkages using three-dimensional convolutional neural networks Cecen, Ahmet; Dai, Hanjun; Yabansu, Yuksel C. Acta Materialia, Vol. 146 https://doi.org/10.1016/j.actamat.2017.11.053	journal	March 2018
Segmentation of experimental datasets via convolutional neural networks trained on phase field simulations Yeom, Jiwon; Stan, Tiberiu; Hong, Seungbum Acta Materialia, Vol. 214 https://doi.org/10.1016/j.actamat.2021.116990	journal	August 2021
DLHub: Simplifying publication, discovery, and use of machine learning models in science Li, Zhuozhao; Chard, Ryan; Ward, Logan Journal of Parallel and Distributed Computing, Vol. 147 https://doi.org/10.1016/j.jpdc.2020.08.006	journal	January 2021
Optimizing convolutional neural networks to perform semantic segmentation on large materials imaging datasets: X-ray tomography and serial sectioning Stan, Tiberiu; Thompson, Zachary T.; Voorhees, Peter W. Materials Characterization, Vol. 160 https://doi.org/10.1016/j.matchar.2020.110119	journal	February 2020
Segmentation of multispectral remote sensing images using active support vector machines Mitra, Pabitra; Uma Shankar, B.; Pal, Sankar K. Pattern Recognition Letters, Vol. 25, Issue 9 https://doi.org/10.1016/j.patrec.2004.03.004	journal	July 2004
Image Segmentation Using K -means Clustering Algorithm and Subtractive Clustering Algorithm Dhanachandra, Nameirakpam; Manglem, Khumanthem; Chanu, Yambem Jina Procedia Computer Science, Vol. 54 https://doi.org/10.1016/j.procs.2015.06.090	journal	January 2015
V-Net: Fully Convolutional Neural Networks for Volumetric Medical Image Segmentation Milletari, Fausto; Navab, Nassir; Ahmadi, Seyed-Ahmad 2016 Fourth International Conference on 3D Vision (3DV) https://doi.org/10.1109/3DV.2016.79	conference	October 2016
Gradient-based learning applied to document recognition Lecun, Y.; Bottou, L.; Bengio, Y. Proceedings of the IEEE, Vol. 86, Issue 11 https://doi.org/10.1109/5.726791	journal	January 1998
Fully convolutional networks for semantic segmentation Long, Jonathan; Shelhamer, Evan; Darrell, Trevor 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) https://doi.org/10.1109/CVPR.2015.7298965	conference	June 2015
Deep Residual Learning for Image Recognition He, Kaiming; Zhang, Xiangyu; Ren, Shaoqing 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) https://doi.org/10.1109/CVPR.2016.90	conference	June 2016
Residual Dense Network for Image Super-Resolution Zhang, Yulun; Tian, Yapeng; Kong, Yu 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) https://doi.org/10.1109/CVPR.2018.00262	conference	June 2018
4D Spatio-Temporal ConvNets: Minkowski Convolutional Neural Networks Choy, Christopher; Gwak, JunYoung; Savarese, Silvio 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) https://doi.org/10.1109/CVPR.2019.00319	conference	June 2019
Jaccard index compensation for object segmentation evaluation Shi, Ran; Ngan, King Ngi; Li, Songnan 2014 IEEE International Conference on Image Processing (ICIP) https://doi.org/10.1109/ICIP.2014.7025904	conference	October 2014
TIMBIR: A Method for Time-Space Reconstruction From Interlaced Views Aditya Mohan, K.; Venkatakrishnan, S. V.; Gibbs, John W. IEEE Transactions on Computational Imaging, Vol. 1, Issue 2 https://doi.org/10.1109/TCI.2015.2431913	journal	June 2015
Optical character recognition for cursive handwriting Arica, N.; Yarman-Vural, F. T. IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 24, Issue 6 https://doi.org/10.1109/TPAMI.2002.1008386	journal	June 2002
A Threshold Selection Method from Gray-Level Histograms Otsu, Nobuyuki IEEE Transactions on Systems, Man, and Cybernetics, Vol. 9, Issue 1 https://doi.org/10.1109/TSMC.1979.4310076	journal	January 1979
Building towards a universal neural network to segment large materials science imaging datasets Stan, Tiberiu; Thompson, Zachary; Voorhees, Peter Developments in X-Ray Tomography XII https://doi.org/10.1117/12.2525290	conference	September 2019
Dendritic growth Trivedi, R.; Kurz, W. International Materials Reviews, Vol. 39, Issue 2 https://doi.org/10.1179/imr.1994.39.2.49	journal	January 1994
A survey on Image Data Augmentation for Deep Learning Shorten, Connor; Khoshgoftaar, Taghi M. Journal of Big Data, Vol. 6, Issue 1 https://doi.org/10.1186/s40537-019-0197-0	journal	July 2019
Deep-learning tomography Araya-Polo, Mauricio; Jennings, Joseph; Adler, Amir The Leading Edge, Vol. 37, Issue 1 https://doi.org/10.1190/tle37010058.1	journal	January 2018
Transfer Learning with Deep Convolutional Neural Network for SAR Target Classification with Limited Labeled Data Huang, Zhongling; Pan, Zongxu; Lei, Bin Remote Sensing, Vol. 9, Issue 9 https://doi.org/10.3390/rs9090907	journal	August 2017

Similar Records

Artificial neural network approach for multiphase segmentation of battery electrode nano-CT images

Journal Article · Tue Feb 08 19:00:00 EST 2022 · npj Computational Materials · OSTI ID:1894650

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

Journal Article · Wed Aug 15 00:00:00 EDT 2018 · Materials Characterization · OSTI ID:22805091

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

Journal Article · Sun May 27 20:00:00 EDT 2018 · Materials Characterization · OSTI ID:1475563

Related Subjects

36 MATERIALS SCIENCE
3D image analysis
97 MATHEMATICS AND COMPUTING
Artificial neural networks
Dendritic formation
Solidification microstructure
X-ray computed tomography

Segmentation of tomography datasets using 3D convolutional neural networks

Citation Formats

References (27)

Similar Records

Related Subjects