Improving Tomographic Reconstruction from Limited Data Using Mixed-Scale Dense Convolutional Neural Networks

In many applications of tomography, the acquired data are limited in one or more ways due to unavoidable experimental constraints. In such cases, popular direct reconstruction algorithms tend to produce inaccurate images, and more accurate iterative algorithms often have prohibitively high computational costs. Using machine learning to improve the image quality of direct algorithms is a recently proposed alternative, for which promising results have been shown. However, previous attempts have focused on using encoder–decoder networks, which have several disadvantages when applied to large tomographic images, preventing wide application in practice. Here, we propose the use of the Mixed-Scale Dense convolutional neural network architecture, which was specifically designed to avoid these disadvantages, to improve tomographic reconstruction from limited data. Results are shown for various types of data limitations and object types, for both simulated data and large-scale real-world experimental data. The results are compared with popular tomographic reconstruction algorithms and machine learning algorithms, showing that Mixed-Scale Dense networks are able to significantly improve reconstruction quality even with severely limited data, and produce more accurate results than existing algorithms.

[1]  M. Stampanoni,et al.  Regridding reconstruction algorithm for real-time tomographic imaging , 2012, Journal of synchrotron radiation.

[2]  Geoffrey E. Hinton,et al.  Rectified Linear Units Improve Restricted Boltzmann Machines , 2010, ICML.

[3]  Eero P. Simoncelli,et al.  Image quality assessment: from error visibility to structural similarity , 2004, IEEE Transactions on Image Processing.

[4]  Nicolas Pinto,et al.  PyCUDA and PyOpenCL: A scripting-based approach to GPU run-time code generation , 2009, Parallel Comput..

[5]  Heikki Sipila Moving-object computer tomography for luggage inspection , 1993, Other Conferences.

[6]  Vladlen Koltun,et al.  Multi-Scale Context Aggregation by Dilated Convolutions , 2015, ICLR.

[7]  Hu Chen,et al.  Low-dose CT via convolutional neural network. , 2017, Biomedical optics express.

[8]  Jan Sijbers,et al.  The ASTRA Toolbox: A platform for advanced algorithm development in electron tomography. , 2015, Ultramicroscopy.

[9]  Xiaoou Tang,et al.  Image Super-Resolution Using Deep Convolutional Networks , 2014, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[10]  Daniël M Pelt,et al.  A mixed-scale dense convolutional neural network for image analysis , 2017, Proceedings of the National Academy of Sciences.

[11]  Tekin Bicer,et al.  Trace: a high-throughput tomographic reconstruction engine for large-scale datasets , 2017, Advanced Structural and Chemical Imaging.

[12]  Michael Elad,et al.  Spatially-Adaptive Reconstruction in Computed Tomography Using Neural Networks , 2013, IEEE Transactions on Medical Imaging.

[13]  Yuan Yu,et al.  TensorFlow: A system for large-scale machine learning , 2016, OSDI.

[14]  Avinash C. Kak,et al.  Principles of computerized tomographic imaging , 2001, Classics in applied mathematics.

[15]  B. Metscher MicroCT for comparative morphology: simple staining methods allow high-contrast 3D imaging of diverse non-mineralized animal tissues , 2009, BMC Physiology.

[16]  P. Midgley,et al.  Electron tomography and holography in materials science. , 2009, Nature materials.

[17]  Pedro J. Carvalho,et al.  Deep learning for plasma tomography using the bolometer system at JET , 2017, 1701.00322.

[18]  G. Chinn,et al.  An investigation of filter choice for filtered back-projection reconstruction in PET , 1997, 1997 IEEE Nuclear Science Symposium Conference Record.

[19]  Kees Joost Batenburg,et al.  Electron tomography based on highly limited data using a neural network reconstruction technique. , 2015, Ultramicroscopy.

[20]  Michael Unser,et al.  Deep Convolutional Neural Network for Inverse Problems in Imaging , 2016, IEEE Transactions on Image Processing.

[21]  Jan Sijbers,et al.  TomoBank: a tomographic data repository for computational x-ray science , 2018 .

[22]  Charudatta Phatak,et al.  A convolutional neural network approach to calibrating the rotation axis for X-ray computed tomography. , 2017, Journal of synchrotron radiation.

[23]  Francesco De Carlo,et al.  Density measurement of samples under high pressure using synchrotron microtomography and diamond anvil cell techniques , 2010, Journal of synchrotron radiation.

[24]  Marco Stampanoni,et al.  Dose optimization approach to fast X-ray microtomography of the lung alveoli , 2013, Journal of applied crystallography.

[25]  Marc Teboulle,et al.  Fast Gradient-Based Algorithms for Constrained Total Variation Image Denoising and Deblurring Problems , 2009, IEEE Transactions on Image Processing.

[26]  M. Vannier,et al.  Why do commercial CT scanners still employ traditional, filtered back-projection for image reconstruction? , 2009, Inverse problems.

[27]  Jian Sun,et al.  Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification , 2015, 2015 IEEE International Conference on Computer Vision (ICCV).

[28]  Thomas Brox,et al.  U-Net: Convolutional Networks for Biomedical Image Segmentation , 2015, MICCAI.

[29]  Francesco De Carlo,et al.  TomoPy: a framework for the analysis of synchrotron tomographic data , 2014, Optics & Photonics - Optical Engineering + Applications.

[30]  Marco Stampanoni,et al.  Synchrotron X-ray tomographic microscopy of fossil embryos , 2006, Nature.

[31]  M. Kachelriess,et al.  Improved total variation-based CT image reconstruction applied to clinical data , 2011, Physics in medicine and biology.

[32]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[33]  Nitish Srivastava,et al.  Dropout: a simple way to prevent neural networks from overfitting , 2014, J. Mach. Learn. Res..

[34]  Jonas Adler,et al.  Learned Primal-Dual Reconstruction , 2017, IEEE Transactions on Medical Imaging.

[35]  Eva L. Dyer,et al.  Low-dose x-ray tomography through a deep convolutional neural network , 2018, Scientific Reports.

[36]  Kees Joost Batenburg,et al.  Fast Tomographic Reconstruction From Limited Data Using Artificial Neural Networks , 2013, IEEE Transactions on Image Processing.