Diffraction tomography with a deep image prior

Abstract

We present a tomographic imaging technique, termed Deep Prior Diffraction Tomography (DP-DT), to reconstruct the 3D refractive index (RI) of thick biological samples at high resolution from a sequence of low-resolution images collected under angularly varying illumination. DP-DT processes the multi-angle data using a phase retrieval algorithm that is extended by a deep image prior (DIP), which reparameterizes the 3D sample reconstruction with an untrained, deep generative 3D convolutional neural network (CNN). We show that DP-DT effectively addresses the missing cone problem, which otherwise degrades the resolution and quality of standard 3D reconstruction algorithms. As DP-DT does not require pre-captured data or pre-training, it is not biased towards any particular dataset. Hence, it is a general technique that can be applied to a wide variety of 3D samples, including scenarios in which large datasets for supervised training would be infeasible or expensive. We applied DP-DT to obtain 3D RI maps of bead phantoms and complex biological specimens, both in simulation and experiment, and show that DP-DT produces higher-quality results than standard regularization techniques. We further demonstrate the generality of DP-DT, using two different scattering models, the first Born and multi-slice models. Our results point to the potential benefits of DP-DT for other 3D imaging modalities, including X-ray computed tomography, magnetic resonance imaging, and electron microscopy.

References in corpus (6)

• Wide-field, high-resolution Fourier ptychographic microscopy
• Probing 10 μK stability and residual drifts in the cross-polarized dual-mode stabilization of single-crystal ultrahigh-Q optical resonators
• High-resolution 3D refractive index microscopy of multiple-scattering samples from intensity images
• Diffraction tomography with a deep image prior
• Image Prediction for Limited-angle Tomography via Deep Learning with Convolutional Neural Network
• Single-cell diffraction tomography with optofluidic rotation about a tilted axis

Cited by in corpus (21)

• On the use of deep learning for phase recovery
• Diffraction tomography with a deep image prior
• Deep image prior for undersampling high-speed photoacoustic microscopy
• Untrained networks for compressive lensless photography
• Robust Phase Unwrapping via Deep Image Prior for Quantitative Phase Imaging
• PtyLab.m/py/jl: a cross-platform, open-source inverse modeling toolbox for conventional and Fourier ptychography
• Coordinate-based neural representations for computational adaptive optics in widefield microscopy
• High-fidelity intensity diffraction tomography with a non-paraxial multiple-scattering model
• Unified k-space theory of optical coherence tomography
• An Unsupervised Reconstruction Method For Low-Dose CT Using Deep Generative Regularization Prior
• Blind Image Deblurring with Unknown Kernel Size and Substantial Noise
• Tensorial tomographic Fourier Ptychography with applications to muscle tissue imaging
• Accelerated deep self-supervised ptycho-laminography for three-dimensional nanoscale imaging of integrated circuits
• Differentiable model-based adaptive optics for two-photon microscopy
• A Dataset-free Deep learning Method for Low-Dose CT Image Reconstruction
• High-speed multiview imaging approaching 4pi steradians using conic section mirrors: theoretical and practical considerations
• Physical model simulator-trained neural network for computational 3D phase imaging of multiple-scattering samples
• Differentiable model-based adaptive optics with transmitted and reflected light
• Phase retrieval via Zernike phase contrast microscopy with an untrained neural network
• Tailored 3D microphantoms: an essential tool for quantitative phase tomography analysis of organoids
• Missing Cone Artifacts Removal in ODT using Unsupervised Deep Learning in Projection Domain