Unsupervised Missing Cone Deep Learning in Optical Diffraction Tomography

Optical diffraction tomography (ODT) produces three dimensional distribution of refractive index (RI) by measuring scattering fields at various angles. Although the distribution of RI index is highly informative, due to the missing cone problem stemming from the limited-angle acquisition of holograms, reconstructions have very poor resolution along axial direction compared to the horizontal imaging plane. To solve this issue, here we present a novel unsupervised deep learning framework, which learns the probability distribution of missing projection views through optimal transport driven cycleGAN. Experimental results show that missing cone artifact in ODT can be significantly resolved by the proposed method.

[1]  Gabriel Popescu,et al.  Fourier phase microscopy for investigation of biological structures and dynamics. , 2004, Optics letters.

[2]  Jong Chul Ye,et al.  Comparative study of iterative reconstruction algorithms for missing cone problems in optical diffraction tomography. , 2015, Optics express.

[3]  M. Schmid.  Principles Of Optics Electromagnetic Theory Of Propagation Interference And Diffraction Of Light , 2016 .

[4]  Jeannot Trampert,et al.  Simultaneous iterative reconstruction technique: Physical interpretation based on the generalized least squares solution , 1990 .

[5]  Kaiming He,et al.  Group Normalization , 2018, ECCV.

[6]  Andrea Vedaldi,et al.  Deep Image Prior , 2017, International Journal of Computer Vision.

[7]  Shai Bagon,et al.  InGAN: Capturing and Remapping the "DNA" of a Natural Image , 2018 .

[8]  M. Radermacher.  Weighted Back-Projection Methods , 2007 .

[9]  YongKeun Park,et al.  Active illumination using a digital micromirror device for quantitative phase imaging. , 2015, Optics letters.

[10]  Michel Barlaud,et al.  Deterministic edge-preserving regularization in computed imaging , 1997, IEEE Trans. Image Process..

[11]  Tomer Michaeli,et al.  Deep-STORM: super-resolution single-molecule microscopy by deep learning , 2018, 1801.09631.

[12]  V. Lauer.  New approach to optical diffraction tomography yielding a vector equation of diffraction tomography and a novel tomographic microscope , 2002, Journal of microscopy.

[13]  Jong Chul Ye,et al.  CycleGAN With a Blur Kernel for Deconvolution Microscopy: Optimal Transport Geometry , 2019, IEEE Transactions on Computational Imaging.

[14]  Alexei A. Efros,et al.  Image-to-Image Translation with Conditional Adversarial Networks , 2016, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[15]  M. Glas,et al.  Principles of Computerized Tomographic Imaging , 2000 .

[16]  Michael Unser,et al.  CryoGAN: A New Reconstruction Paradigm for Single-Particle Cryo-EM via Deep Adversarial Learning , 2020, bioRxiv.

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

[18]  Yoram Bresler,et al.  Globally convergent edge-preserving regularized reconstruction: an application to limited-angle tomography , 1998, IEEE Trans. Image Process..

[19]  Jong Chul Ye,et al.  Efficient B-Mode Ultrasound Image Reconstruction From Sub-Sampled RF Data Using Deep Learning , 2017, IEEE Transactions on Medical Imaging.

[20]  Thomas Pock,et al.  Learning a variational network for reconstruction of accelerated MRI data , 2017, Magnetic resonance in medicine.

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

[22]  O. Haeberlé,et al.  Tomographic diffractive microscopy of transparent samples , 2008 .

[23]  Demetri Psaltis,et al.  Three-dimensional tomography of red blood cells using deep learning , 2020, Advanced Photonics.

[24]  R. Dasari,et al.  Diffraction phase microscopy for quantifying cell structure and dynamics. , 2006, Optics letters.

[25]  Quanzheng Li,et al.  Iterative PET Image Reconstruction Using Convolutional Neural Network Representation , 2017, IEEE Transactions on Medical Imaging.

[26]  George Nehmetallah,et al.  3D Optical Diffraction Tomography Using Deep Learning , 2018 .

[27]  E. Wolf.  Three-dimensional structure determination of semi-transparent objects from holographic data , 1969 .

[28]  Reiner Hegerl,et al.  Approximation of missing-cone data in 3D electron microscopy , 1989 .

[29]  YongKeun Park,et al.  Time-multiplexed structured illumination using a DMD for optical diffraction tomography. , 2017, Optics letters.

[30]  YoungJu Jo,et al.  Quantitative Phase Imaging Techniques for the Study of Cell Pathophysiology: From Principles to Applications , 2013, Sensors.

[31]  Michael Unser,et al.  Learning approach to optical tomography , 2015, 1502.01914.

[32]  Zhuo Wang,et al.  Optical measurement of cycle-dependent cell growth , 2011, Proceedings of the National Academy of Sciences.

[33]  Gabriel Popescu,et al.  Optical imaging of cell mass and growth dynamics. , 2008, American journal of physiology. Cell physiology.

[34]  Jong Chul Ye,et al.  ${k}$ -Space Deep Learning for Accelerated MRI , 2020, IEEE Transactions on Medical Imaging.

[35]  Jong Chul Ye,et al.  DeepRegularizer: Rapid Resolution Enhancement of Tomographic Imaging Using Deep Learning , 2020, IEEE Transactions on Medical Imaging.

[36]  Hyun-seok Min,et al.  Quantitative Phase Imaging and Artificial Intelligence: A Review , 2018, IEEE Journal of Selected Topics in Quantum Electronics.

[37]  Jong Chul Ye,et al.  A deep convolutional neural network using directional wavelets for low‐dose X‐ray CT reconstruction , 2016, Medical physics.

[38]  Yibo Zhang,et al.  Deep Learning Microscopy , 2017, ArXiv.

[39]  Samuel Matej,et al.  3D-FRP: direct Fourier reconstruction with Fourier reprojection for fully 3-D PET , 2000 .

[40]  A. Papoulis.  A new algorithm in spectral analysis and band-limited extrapolation. , 1975 .

[41]  C. Fang-Yen,et al.  Optical diffraction tomography for high resolution live cell imaging. , 2009, Optics express.

[42]  Victor Perez-Mendez,et al.  Limited-Angle Three-Dimensional Reconstructions Using Fourier Transform Iterations And Radon Transform Iterations , 1981 .

[43]  R. Gerchberg.  Super-resolution through Error Energy Reduction , 1974 .

[44]  Steve B. Jiang,et al.  Low-dose CT reconstruction via edge-preserving total variation regularization. , 2010, Physics in medicine and biology.

[45]  Kevin C Zhou,et al.  Diffraction tomography with a deep image prior. , 2019, Optics express.

[46]  J.K. Aggarwal,et al.  The missing cone problem and low-pass distortion in optical serial sectioning microscopy , 1988, ICASSP-88., International Conference on Acoustics, Speech, and Signal Processing.

[47]  Alexandros G. Dimakis,et al.  AmbientGAN: Generative models from lossy measurements , 2018, ICLR.

[48]  Shuai Li,et al.  Lensless computational imaging through deep learning , 2017, ArXiv.

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