Sharpness-aware Low dose CT denoising using conditional generative adversarial network

Abstract

Low Dose Computed Tomography (LDCT) has offered tremendous benefits in radiation restricted applications, but the quantum noise as resulted by the insufficient number of photons could potentially harm the diagnostic performance. Current image-based denoising methods tend to produce a blur effect on the final reconstructed results especially in high noise levels. In this paper, a deep learning based approach was proposed to mitigate this problem. An adversarially trained network and a sharpness detection network were trained to guide the training process. Experiments on both simulated and real dataset shows that the results of the proposed method have very small resolution loss and achieves better performance relative to the-state-of-art methods both quantitatively and visually.

1. updated results, related works and discussion 2. fixed an error for the noise level calculation 3. redrawn two diagrams 4. added an experiment on unknown dose level CT scans

References in corpus (13)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
• Conditional Generative Adversarial Nets
• Low-Dose CT with a Residual Encoder-Decoder Convolutional Neural Network (RED-CNN)
• Low Dose CT Image Denoising Using a Generative Adversarial Network with Wasserstein Distance and Perceptual Loss
• Improved Training of Wasserstein GANs
• Image De-raining Using a Conditional Generative Adversarial Network
• Learning What and Where to Draw
• The Importance of Skip Connections in Biomedical Image Segmentation
• CT Image Denoising with Perceptive Deep Neural Networks
• Scribbler: Controlling Deep Image Synthesis with Sketch and Color
• Spatially-Varying Blur Detection Based on Multiscale Fused and Sorted Transform Coefficients of Gradient Magnitudes
• Low-Dose CT via Deep Neural Network

Cited by in corpus (27)

• Generative Adversarial Network in Medical Imaging: A Review
• Can Deep Learning Outperform Modern Commercial CT Image Reconstruction Methods?
• Cycle Consistent Adversarial Denoising Network for Multiphase Coronary CT Angiography
• CoreDiff: Contextual Error-Modulated Generalized Diffusion Model for Low-Dose CT Denoising and Generalization
• EDCNN: Edge enhancement-based Densely Connected Network with Compound Loss for Low-Dose CT Denoising
• Deep Bayesian Inversion
• A review of Generative Adversarial Networks for Electronic Health Records: applications, evaluation measures and data sources
• Parameter-Transferred Wasserstein Generative Adversarial Network (PT-WGAN) for Low-Dose PET Image Denoising
• Automatic catheter detection in pediatric X-ray images using a scale-recurrent network and synthetic data
• High-Resolution Mammogram Synthesis using Progressive Generative Adversarial Networks
• Conditional Generation of Medical Images via Disentangled Adversarial Inference
• deepPERFECT: Novel Deep Learning CT Synthesis Method for Expeditious Pancreatic Cancer Radiotherapy
• A hybrid quantum-classical conditional generative adversarial network algorithm for human-centered paradigm in cloud
• DU-GAN: Generative Adversarial Networks with Dual-Domain U-Net Based Discriminators for Low-Dose CT Denoising
• Physical Accuracy of Deep Neural Networks for 2D and 3D Multi-Mineral Segmentation of Rock micro-CT Images
• Medical Imaging Synthesis using Deep Learning and its Clinical Applications: A Review
• Medical Image Generation using Generative Adversarial Networks
• ML-LBM: Machine Learning Aided Flow Simulation in Porous Media
• Unsupervised/Semi-supervised Deep Learning for Low-dose CT Enhancement
• A SSIM Guided cGAN Architecture For Clinically Driven Generative Image Synthesis of Multiplexed Spatial Proteomics Channels
• Overview of image-to-image translation by use of deep neural networks: denoising, super-resolution, modality conversion, and reconstruction in medical imaging
• Learned Interferometric Imaging for the SPIDER Instrument
• HAGAN: Hybrid Augmented Generative Adversarial Network for Medical Image Synthesis
• Sinogram interpolation for sparse-view micro-CT with deep learning neural network
• Deep Learning for Low-Dose CT Denoising
• Three-dimensional visualization of X-ray micro-CT with large-scale datasets: Efficiency and accuracy for real-time interaction
• Few-Shot Meta-Denoising