Fully Dense UNet for 2D Sparse Photoacoustic Tomography Artifact Removal

Abstract

Photoacoustic imaging is an emerging imaging modality that is based upon the photoacoustic effect. In photoacoustic tomography (PAT), the induced acoustic pressure waves are measured by an array of detectors and used to reconstruct an image of the initial pressure distribution. A common challenge faced in PAT is that the measured acoustic waves can only be sparsely sampled. Reconstructing sparsely sampled data using standard methods results in severe artifacts that obscure information within the image. We propose a modified convolutional neural network (CNN) architecture termed Fully Dense UNet (FD-UNet) for removing artifacts from 2D PAT images reconstructed from sparse data and compare the proposed CNN with the standard UNet in terms of reconstructed image quality.

References in corpus (13)

• Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
• TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems
• Fully Convolutional Networks for Semantic Segmentation
• Deep Convolutional Neural Network for Inverse Problems in Imaging
• Training Very Deep Networks
• Model based learning for accelerated, limited-view 3D photoacoustic tomography
• Investigation of iterative image reconstruction in three-dimensional optoacoustic tomography
• Deep Residual Learning for Compressed Sensing CT Reconstruction via Persistent Homology Analysis
• Accelerated High-Resolution Photoacoustic Tomography via Compressed Sensing
• A Reconstruction Algorithm for Photoacoustic Imaging based on the Nonuniform FFT
• Photoacoustic image reconstruction via deep learning
• A simple Fourier transform-based reconstruction formula for photoacoustic computed tomography with a circular or spherical measurement geometry
• Efficient regularization with wavelet sparsity constraints in PAT

Cited by in corpus (40)

• Medical Image Segmentation Using Deep Learning: A Survey
• RTNet: Relation Transformer Network for Diabetic Retinopathy Multi-lesion Segmentation
• SUNet: Swin Transformer UNet for Image Denoising
• MDU-Net: Multi-scale Densely Connected U-Net for biomedical image segmentation
• Deep Learning in Photoacoustic Tomography: Current approaches and future directions
• Limited View and Sparse Photoacoustic Tomography for Neuroimaging with Deep Learning
• Deep image prior for undersampling high-speed photoacoustic microscopy
• A Two-step-training Deep Learning Framework for Real-time Computational Imaging without Physics Priors
• Application of belief functions to medical image segmentation: A review
• Compensating for visibility artefacts in photoacoustic imaging with a deep learning approach providing prediction uncertainties
• CovidCTNet: An Open-Source Deep Learning Approach to Identify Covid-19 Using CT Image
• The Butterfly Effect in Primary Visual Cortex
• Deep Ice Layer Tracking and Thickness Estimation using Fully Convolutional Networks
• Deep learning-based photoacoustic imaging of vascular network through thick porous media
• Fourier Neural Operator Networks: A Fast and General Solver for the Photoacoustic Wave Equation
• Internal-Illumination Photoacoustic Tomography Enhanced by a Graded-scattering Fiber Diffuser
• EAR-U-Net: EfficientNet and attention-based residual U-Net for automatic liver segmentation in CT
• Deep learning for photoacoustic imaging: a survey
• Small but Mighty: Enhancing 3D Point Clouds Semantic Segmentation with U-Next Framework
• Stain Style Transfer using Transitive Adversarial Networks
• Dense Dilated UNet: Deep Learning for 3D Photoacoustic Tomography Image Reconstruction
• A CNN-based tool for automatic tongue contour tracking in ultrasound images
• Leveraging Labelled Data Knowledge: A Cooperative Rectification Learning Network for Semi-supervised 3D Medical Image Segmentation
• Utilizing variational autoencoders in the Bayesian inverse problem of photoacoustic tomography
• Advances in Photoacoustic Imaging Reconstruction and Quantitative Analysis for Biomedical Applications
• Cross-modality Attention Adapter: A Glioma Segmentation Fine-tuning Method for SAM Using Multimodal Brain MR Images
• Preserving Dense Features for Ki67 Nuclei Detection
• FIBA: Frequency-Injection based Backdoor Attack in Medical Image Analysis
• Combining band-frequency separation and deep neural networks for optoacoustic imaging
• A Lightweight Structure Aimed to Utilize Spatial Correlation for Sparse-View CT Reconstruction
• Ground-based image deconvolution with Swin Transformer UNet
• Deep Multicameral Decoding for Localizing Unoccluded Object Instances from a Single RGB Image
• DualSwinUnet++: An Enhanced Swin-Unet Architecture With Dual Decoders For PTMC Segmentation
• CAggNet: Crossing Aggregation Network for Medical Image Segmentation
• Invariant Representations in Deep Learning for Optoacoustic Imaging
• A Jointed Feature Fusion Framework for Photoacoustic Reconstruction
• PDFNet: Pointwise Dense Flow Network for Urban-Scene Segmentation
• A new smart-cropping pipeline for prostate segmentation using deep learning networks
• Compressed Sensing for Photoacoustic Computed Tomography Using an Untrained Neural Network
• A lateral semicircular canal segmentation based geometric calibration for human temporal bone CT Image