Model based learning for accelerated, limited-view 3D photoacoustic tomography

Abstract

Recent advances in deep learning for tomographic reconstructions have shown great potential to create accurate and high quality images with a considerable speed-up. In this work we present a deep neural network that is specifically designed to provide high resolution 3D images from restricted photoacoustic measurements. The network is designed to represent an iterative scheme and incorporates gradient information of the data fit to compensate for limited view artefacts. Due to the high complexity of the photoacoustic forward operator, we separate training and computation of the gradient information. A suitable prior for the desired image structures is learned as part of the training. The resulting network is trained and tested on a set of segmented vessels from lung CT scans and then applied to in-vivo photoacoustic measurement data.

References in corpus (7)

• Learned Primal-dual Reconstruction
• Solving ill-posed inverse problems using iterative deep neural networks
• Recurrent Inference Machines for Solving Inverse Problems
• Mathematics of Deep Learning
• Deep Learning-Guided Image Reconstruction from Incomplete Data
• A Framework for Directional and Higher-Order Reconstruction in Photoacoustic Tomography
• LEARN: Learned Experts' Assessment-based Reconstruction Network for Sparse-data CT

Cited by in corpus (55)

• Fully Dense UNet for 2D Sparse Photoacoustic Tomography Artifact Removal
• Deep D-bar: Real time Electrical Impedance Tomography Imaging with Deep Neural Networks
• Deep learning for biomedical photoacoustic imaging: A review
• Computed Tomography Reconstruction Using Deep Image Prior and Learned Reconstruction Methods
• Deep Learning in Photoacoustic Tomography: Current approaches and future directions
• TomoGAN: Low-Dose Synchrotron X-Ray Tomography with Generative Adversarial Networks
• Learning The Invisible: A Hybrid Deep Learning-Shearlet Framework for Limited Angle Computed Tomography
• A Deep Learning Framework for Single-Sided Sound Speed Inversion in Medical Ultrasound
• Limited View and Sparse Photoacoustic Tomography for Neuroimaging with Deep Learning
• Regularization by architecture: A deep prior approach for inverse problems
• Deep learning in biomedical optics
• Deep Bayesian Inversion
• Deep image prior for undersampling high-speed photoacoustic microscopy
• Numerical Solution of Inverse Problems by Weak Adversarial Networks
• Computationally Efficient Deep Neural Network for Computed Tomography Image Reconstruction
• A Partially Learned Algorithm for Joint Photoacoustic Reconstruction and Segmentation
• Context encoding enables machine learning-based quantitative photoacoustics
• Structural engineering from an inverse problems perspective
• Compensating for visibility artefacts in photoacoustic imaging with a deep learning approach providing prediction uncertainties
• Task adapted reconstruction for inverse problems
• Neural Networks-based Regularization for Large-Scale Medical Image Reconstruction
• Deep learning-based photoacoustic imaging of vascular network through thick porous media
• SGD-Net: Efficient Model-Based Deep Learning with Theoretical Guarantees
• An Educated Warm Start For Deep Image Prior-Based Micro CT Reconstruction
• Fourier Neural Operator Networks: A Fast and General Solver for the Photoacoustic Wave Equation
• Real-time photoacoustic projection imaging using deep learning
• Deep Learning Adapted Acceleration for Limited-view Photoacoustic Computed Tomography
• CoIL: Coordinate-based Internal Learning for Imaging Inverse Problems
• Reconstructing undersampled photoacoustic microscopy images using deep learning
• Learning-based sound speed estimation and aberration correction in linear-array photoacoustic imaging
• Deep learning versus ${\ell }^1$-minimization for compressed sensing photoacoustic tomography
• Approximate k-space models and Deep Learning for fast photoacoustic reconstruction
• Dense Dilated UNet: Deep Learning for 3D Photoacoustic Tomography Image Reconstruction
• Y-Net: A Hybrid Deep Learning Reconstruction Framework for Photoacoustic Imaging in vivo
• Fully three-dimensional sound speed-corrected multi-wavelength photoacoustic breast tomography
• Utilizing variational autoencoders in the Bayesian inverse problem of photoacoustic tomography
• Advances in Photoacoustic Imaging Reconstruction and Quantitative Analysis for Biomedical Applications
• A Hybrid Reconstruction Approach for Absorption Coefficient by Fluorescence Photoacoustic Tomography
• Invertible residual networks in the context of regularization theory for linear inverse problems
• NETT Regularization for Compressed Sensing Photoacoustic Tomography
• Towards accurate quantitative photoacoustic imaging: learning vascular blood oxygen saturation in 3D
• Quantifying Sources of Uncertainty in Deep Learning-Based Image Reconstruction
• Deep network series for large-scale high-dynamic range imaging
• Quantifying Model Uncertainty in Inverse Problems via Bayesian Deep Gradient Descent
• On the randomised stability constant for inverse problems
• Inverse Aerodynamic Design of Gas Turbine Blades using Probabilistic Machine Learning
• Learned backprojection for sparse and limited view photoacoustic tomography
• Broadband model-based inversion enables optoacoustic microscopy beyond the acoustic diffraction limit
• A Projectional Ansatz to Reconstruction
• A Jointed Feature Fusion Framework for Photoacoustic Reconstruction
• Swap-Net: A Memory-Efficient 2.5D Network for Sparse-View 3D Cone Beam CT Reconstruction
• Importance of Signal and Image Processing in Photoacoustic Imaging
• A Machine-learning Based Initialization for Joint Statistical Iterative Dual-energy CT with Application to Proton Therapy
• Deep Learning Enabled Real-Time Photoacoustic Tomography System via Single Data Acquisition Channel
• AHP-Net: adaptive-hyper-parameter deep learning based image reconstruction method for multilevel low-dose CT