Abstract:Photoacoustic imaging potentially allows for the real-time visualization of functional human tissue parameters such as oxygenation but is subject to a challenging underlying quantification problem. While in silico studies have revealed the great potential of deep learning (DL) methodology in solving this problem, the inherent lack of an efficient gold standard method for model training and validation remains a grand challenge. This work investigates whether DL can be leveraged to accurately and efficiently simulate photon propagation in biological tissue, enabling photoacoustic image synthesis. Our approach is based on estimating the initial pressure distribution of the photoacoustic waves from the underlying optical properties using a back-propagatable neural network trained on synthetic data. In proof-of-concept studies, we validated the performance of two complementary neural network architectures, namely a conventional U-Net-like model and a Fourier Neural Operator (FNO) network. Our in silico validation on multispectral human forearm images shows that DL methods can speed up image generation by a factor of 100 when compared to Monte Carlo simulations with 5×108 photons. While the FNO is slightly more accurate than the U-Net, when compared to Monte Carlo simulations performed with a reduced number of photons (5×106), both neural network architectures achieve equivalent accuracy. In contrast to Monte Carlo simulations, the proposed DL models can be used as inherently differentiable surrogate models in the photoacoustic image synthesis pipeline, allowing for back-propagation of the synthesis error and gradient-based optimization over the entire pipeline. Due to their efficiency, they have the potential to enable large-scale training data generation that can expedite the clinical application of photoacoustic imaging.

Towards accurate quantitative photoacoustic imaging: learning vascular blood oxygen saturation in 3D

O-Net: A Convolutional Neural Network for Quantitative Photoacoustic Image Segmentation and Oximetry

Quantitative Endoscopic Photoacoustic Tomography Using a Convolutional Neural Network.

Deep learning-based photoacoustic imaging of vascular network through thick porous media

A Sparse Deep Learning Approach for Automatic Segmentation of Human Vasculature in Multispectral Optoacoustic Tomography

Deep Learning Enhanced Volumetric Photoacoustic Imaging of Vasculature in Human (Adv. Sci. 29/2023)

Method for measuring blood flow and depth of blood vessel based on laser speckle contrast imaging using 3D convolutional neural network: A preliminary study

EDA-Net: Dense Aggregation of Deep and Shallow Information Achieves Quantitative Photoacoustic Blood Oxygenation Imaging Deep in Human Breast

Numerical Simulation and Analysis of Accurate Blood Oxygenation Measurement by Using Optical Resolution Photoacoustic Microscopy

Efficient Photoacoustic Image Synthesis with Deep Learning

Physics-Informed Learning for Time-Resolved Angiographic Contrast Agent Concentration Reconstruction

Distribution-informed and wavelength-flexible data-driven photoacoustic oximetry

End-to-end Deep Neural Network for Optica L Inversion in Quantitative Photoacoustic Imaging

Enhancing synthetic training data for quantitative photoacoustic tomography with generative deep learning

Simulation-informed learning for time-resolved angiographic contrast agent concentration reconstruction

Deep-Learning Image Reconstruction for Real-Time Photoacoustic System

QOCT-Net: A Physics-Informed Neural Network for Intravascular Optical Coherence Tomography Attenuation Imaging

Enhancing MR vascular Fingerprinting through realistic microvascular geometries

Machine-learning-based mapping of blood oxygen saturation from dual-wavelength optoacoustic measurements

A Deep Learning Framework for Spatiotemporal Ultrasound Localization Microscopy

Deep Learning in Photoacoustic Tomography: Current approaches and future directions