3D Convolutional Encoder-Decoder Network for Low-Dose CT via Transfer Learning from a 2D Trained Network

Abstract

Low-dose computed tomography (CT) has attracted a major attention in the medical imaging field, since CT-associated x-ray radiation carries health risks for patients. The reduction of CT radiation dose, however, compromises the signal-to-noise ratio, and may compromise the image quality and the diagnostic performance. Recently, deep-learning-based algorithms have achieved promising results in low-dose CT denoising, especially convolutional neural network (CNN) and generative adversarial network (GAN). This article introduces a Contracting Path-based Convolutional Encoder-decoder (CPCE) network in 2D and 3D configurations within the GAN framework for low-dose CT denoising. A novel feature of our approach is that an initial 3D CPCE denoising model can be directly obtained by extending a trained 2D CNN and then fine-tuned to incorporate 3D spatial information from adjacent slices. Based on the transfer learning from 2D to 3D, the 3D network converges faster and achieves a better denoising performance than that trained from scratch. By comparing the CPCE with recently published methods based on the simulated Mayo dataset and the real MGH dataset, we demonstrate that the 3D CPCE denoising model has a better performance, suppressing image noise and preserving subtle structures.

arXiv commentTo be published in the IEEE TMI

References in corpus (2)

• Deep Convolutional Neural Network for Inverse Problems in Imaging
• Low-Dose CT with a Residual Encoder-Decoder Convolutional Neural Network (RED-CNN)

Cited by in corpus (37)

• Generative Adversarial Network in Medical Imaging: A Review
• CT Super-resolution GAN Constrained by the Identical, Residual, and Cycle Learning Ensemble(GAN-CIRCLE)
• Can Deep Learning Outperform Modern Commercial CT Image Reconstruction Methods?
• DeepPET: A deep encoder-decoder network for directly solving the PET reconstruction inverse problem
• Multi-Contrast Super-Resolution MRI Through a Progressive Network
• Boundary-weighted Domain Adaptive Neural Network for Prostate MR Image Segmentation
• CoreDiff: Contextual Error-Modulated Generalized Diffusion Model for Low-Dose CT Denoising and Generalization
• TomoGAN: Low-Dose Synchrotron X-Ray Tomography with Generative Adversarial Networks
• EDCNN: Edge enhancement-based Densely Connected Network with Compound Loss for Low-Dose CT Denoising
• MRI Super-Resolution with Ensemble Learning and Complementary Priors
• Hierarchical Amortized Training for Memory-efficient High Resolution 3D GAN
• A Deep Learning Framework for Single-Sided Sound Speed Inversion in Medical Ultrasound
• The LoDoPaB-CT Dataset: A Benchmark Dataset for Low-Dose CT Reconstruction Methods
• Noise Conscious Training of Non Local Neural Network powered by Self Attentive Spectral Normalized Markovian Patch GAN for Low Dose CT Denoising
• Shape and Margin-Aware Lung Nodule Classification in Low-dose CT Images via Soft Activation Mapping
• Parameter-Transferred Wasserstein Generative Adversarial Network (PT-WGAN) for Low-Dose PET Image Denoising
• LIT-Former: Linking In-plane and Through-plane Transformers for Simultaneous CT Image Denoising and Deblurring
• Deep Efficient End-to-end Reconstruction (DEER) Network for Few-view Breast CT Image Reconstruction
• MANAS: Multi-Scale and Multi-Level Neural Architecture Search for Low-Dose CT Denoising
• Meta Ordinal Regression Forest for Medical Image Classification with Ordinal Labels
• IQAGPT: Image Quality Assessment with Vision-language and ChatGPT Models
• Systematic Review on Learning-based Spectral CT
• Noise2Context: Context-assisted Learning 3D Thin-layer Low Dose CT Without Clean Data
• Deep Encoder-decoder Adversarial Reconstruction (DEAR) Network for 3D CT from Few-view Data
• ASCON: Anatomy-aware Supervised Contrastive Learning Framework for Low-dose CT Denoising
• Generative Models Improve Radiomics Reproducibility in Low Dose CTs: A Simulation Study
• DU-GAN: Generative Adversarial Networks with Dual-Domain U-Net Based Discriminators for Low-Dose CT Denoising
• 2D and 3D Segmentation of uncertain local collagen fiber orientations in SHG microscopy
• Limited Parameter Denoising for Low-dose X-ray Computed Tomography Using Deep Reinforcement Learning
• Simultaneous reconstruction of the initial pressure and sound speed in photoacoustic tomography using a deep-learning approach
• Segmentation-free PVC for Cardiac SPECT using a Densely-connected Multi-dimensional Dynamic Network
• Impact of loss functions on the performance of a deep neural network designed to restore low-dose digital mammography
• Dual Network Architecture for Few-view CT -- Trained on ImageNet Data and Transferred for Medical Imaging
• Stochastic Modeling of Inhomogeneities in the Aortic Wall and Uncertainty Quantification using a Bayesian Encoder-Decoder Surrogate
• HAGAN: Hybrid Augmented Generative Adversarial Network for Medical Image Synthesis
• Cross-domain Denoising for Low-dose Multi-frame Spiral Computed Tomography
• Deep-learning-based Breast CT for Radiation Dose Reduction