Learn an Index Operator by CNN for Solving Diffusive Optical Tomography: A Deep Direct Sampling Method

In this work, we investigate the diffusive optical tomography (DOT) problem in the case that limited boundary measurements are available. Motivated by the direct sampling method (DSM) proposed in [13], we develop a deep direct sampling method (DDSM) to recover the inhomogeneous inclusions buried in a homogeneous background. In this method, we design a convolutional neural network (CNN) to approximate the index functional that mimics the underling mathematical structure. The benefits of the proposed DDSM include fast and easy implementation, capability of incorporating multiple measurements to attain high-quality reconstruction, and advanced robustness against the noise. Numerical experiments show that the reconstruction accuracy is improved without degrading the efficiency, demonstrating its potential for solving the real-world DOT problems.

[1]  B. Pogue,et al.  Three-dimensional optical tomography: resolution in small-object imaging. , 2003, Applied optics.

[2]  Lexing Ying,et al.  Solving Optical Tomography with Deep Learning , 2019, Annals of Mathematical Sciences and Applications.

[3]  Jaejun Yoo,et al.  Deep Learning Diffuse Optical Tomography , 2017, IEEE Transactions on Medical Imaging.

[4]  L. Fajardo,et al.  Differentiation of cysts from solid tumors in the breast with diffuse optical tomography. , 2004, Academic radiology.

[5]  Charles A. Bouman,et al.  Nonlinear multigrid algorithms for Bayesian optical diffusion tomography , 2001, IEEE Trans. Image Process..

[6]  Aldo Di Costanzo Mata,et al.  Multimodal imaging combining time-domain near-infrared optical tomography and continuous-wave fluorescence molecular tomography. , 2020, Optics Express.

[7]  S. Siltanen,et al.  D-BAR METHOD AND EXCEPTIONAL POINTS AT POSITIVE ENERGY : A COMPUTATIONAL STUDY , 2014 .

[8]  Thomas J. Downar,et al.  Modified distorted Born iterative method with an approximate Fréchet derivative for optical diffusion tomography , 1999 .

[9]  Yuqi Yao,et al.  Born type iterative method for imaging of heterogeneous scattering media and its application to simulated breast tissue , 1997, Photonics West - Biomedical Optics.

[10]  Jong Chul Ye,et al.  Compressive Diffuse Optical Tomography: Noniterative Exact Reconstruction Using Joint Sparsity , 2011, IEEE Transactions on Medical Imaging.

[11]  Kazufumi Ito,et al.  Direct Sampling Method for Diffusive Optical Tomography , 2014, SIAM J. Sci. Comput..

[12]  Stefan A. Carp,et al.  Deep learning enabled high-speed image reconstruction for breast diffuse optical tomography , 2021 .

[13]  Bangti Jin,et al.  A direct sampling method for inverse electromagnetic medium scattering , 2012, 1212.5085.

[14]  Iasonas Kokkinos,et al.  DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFs , 2016, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[15]  Jong Chul Ye,et al.  DR AF T Deep Learning Can Reverse Photon Migration for Diffuse Optical Tomography , 2017 .

[16]  V. Ntziachristos,et al.  Three-dimensional diffuse optical tomography in the parallel plane transmission geometry: evaluation of a hybrid frequency domain/continuous wave clinical system for breast imaging. , 2003, Medical physics.

[17]  Yat Tin Chow,et al.  A Direct Sampling Method for the Inversion of the Radon Transform , 2020, ArXiv.

[18]  Seungseok Oh,et al.  Multigrid inversion algorithms with applications to optical diffusion tomography , 2002, Conference Record of the Thirty-Sixth Asilomar Conference on Signals, Systems and Computers, 2002..

[19]  Margaret Cheney,et al.  The Linear Sampling Method and the MUSIC Algorithm , 2001 .

[20]  Erkki Somersalo,et al.  Estimation of optical absorption in anisotropic background , 2002 .

[21]  Simon R. Arridge,et al.  Three-dimensional reconstruction of shape and piecewise constant region values for optical tomography using spherical harmonic parametrization and a boundary element method , 2006 .

[22]  Guillaume Bal.  Reconstructions in impedance and optical tomography with singular interfaces , 2005 .

[23]  Josselin Garnier,et al.  Reconstruction of a Piecewise Smooth Absorption Coefficient by an Acousto-Optic Process , 2012, 1206.1412.

[24]  Andreas Kirsch,et al.  Characterization of the shape of a scattering obstacle using the spectral data of the far field operator , 1998 .

[25]  H. Ammari,et al.  Reconstruction and stability in acousto-optic imaging for absorption maps with bounded variation ☆ , 2014, 1405.2679.

[26]  Bangti Jin,et al.  A direct sampling method to an inverse medium scattering problem , 2012 .

[27]  Kazufumi Ito,et al.  A Time-Dependent Direct Sampling Method for Recovering Moving Potentials in a Heat Equation , 2018, SIAM J. Sci. Comput..

[28]  A. Hielscher,et al.  Optical tomography as a PDE-constrained optimization problem , 2005 .

[29]  Lin Zhang,et al.  Brief review on learning-based methods for optical tomography , 2019, Journal of Innovative Optical Health Sciences.

[30]  Matthew B. Applegate,et al.  Recent advances in high speed diffuse optical imaging in biomedicine , 2020 .

[31]  A. Kleefeld,et al.  The factorization method for the acoustic transmission problem , 2013 .

[32]  Ruchi Guo,et al.  Construct Deep Neural Networks Based on Direct Sampling Methods for Solving Electrical Impedance Tomography , 2020, SIAM J. Sci. Comput..

[33]  T. R. Lucas,et al.  A fast and accurate imaging algorithm in optical/diffusion tomography , 1997 .

[34]  Ville Kolehmainen,et al.  3D shape based reconstruction of experimental data in Diffuse Optical Tomography. , 2009, Optics express.

[35]  Sergey Ioffe,et al.  Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift , 2015, ICML.

[36]  Michael V. Klibanov,et al.  New imaging algorithm in diffusion tomography , 1997, Photonics West - Biomedical Optics.

[37]  Jong Chul Ye,et al.  Joint sparsity-driven non-iterative simultaneous reconstruction of absorption and scattering in diffuse optical tomography. , 2013, Optics express.

[38]  Ghassan Hamarneh,et al.  Deep Learning Based Image Reconstruction for Diffuse Optical Tomography , 2018, MLMIR@MICCAI.

[39]  M. Langlois,et al.  Society of Photo-Optical Instrumentation Engineers , 2005 .

[40]  Kazufumi Ito,et al.  A direct sampling method for electrical impedance tomography , 2014 .

[41]  R. Choe.  Diffuse optical tomography and spectroscopy of breast cancer and fetal brain , 2005 .

[42]  Samuli Siltanen,et al.  The D-Bar Method for Diffuse Optical Tomography: A Computational Study , 2016, Exp. Math..