Abstract:Background Magnetic resonance imaging (MRI) is a crucial technique for both scientific research and clinical diagnosis. However, noise generated during MR data acquisition degrades image quality, particularly in hyperpolarized (HP) gas MRI. While deep learning (DL) methods have shown promise for MR image denoising, most of them fail to adequately utilize the long‐range information which is important to improve denoising performance. Furthermore, the sample size of paired noisy and noise‐free MR images also limits denoising performance. Purpose To develop an effective DL method that enhances denoising performance and reduces the requirement of paired MR images by utilizing the long‐range information and pretraining. Methods In this work, a hybrid Transformer‐convolutional neural network (CNN) network (HTC‐net) and a self‐supervised pretraining strategy are proposed, which effectively enhance the denoising performance. In HTC‐net, a CNN branch is exploited to extract the local features. Then a Transformer‐CNN branch with two parallel encoders is designed to capture the long‐range information. Within this branch, a residual fusion block (RFB) with a residual feature processing module and a feature fusion module is proposed to aggregate features at different resolutions extracted by two parallel encoders. After that, HTC‐net exploits the comprehensive features from the CNN branch and the Transformer‐CNN branch to accurately predict noise‐free MR images through a reconstruction module. To further enhance the performance on limited MRI datasets, a self‐supervised pretraining strategy is proposed. This strategy employs self‐supervised denoising to equip the HTC‐net with denoising capabilities during pretraining, and then the pre‐trained parameters are transferred to facilitate subsequent supervised training. Results Experimental results on the pulmonary HP 129Xe MRI dataset (1059 images) and IXI dataset (5000 images) all demonstrate the proposed method outperforms the state‐of‐the‐art methods, exhibiting superior preservation of edges and structures. Quantitatively, on the pulmonary HP 129Xe MRI dataset, the proposed method outperforms the state‐of‐the‐art methods by 0.254–0.597 dB in PSNR and 0.007–0.013 in SSIM. On the IXI dataset, the proposed method outperforms the state‐of‐the‐art methods by 0.3–0.927 dB in PSNR and 0.003–0.016 in SSIM. Conclusions The proposed method can effectively enhance the quality of MR images, which helps improve the diagnosis accuracy in clinical.

Denoising of 3D Magnetic Resonance Images with Multi-Channel Residual Learning of Convolutional Neural Network

Denoising of volumetric magnetic resonance imaging using multi-channel three-dimensional convolutional neural network with applications on fast spin echo acquisitions

CNN and Multi-Feature Extraction Based Denoising of CT Images

Denoising of 3D Brain MR Images with Parallel Residual Learning of Convolutional Neural Network Using Global and Local Feature Extraction

Denoising Magnetic Resonance Images Using Collaborative Non-Local Means

Denoising of MR Images with Rician Noise Using a Wider Neural Network and Noise Range Division

Global and local feature extraction based on convolutional neural network residual learning for MR image denoising

Magnetic resonance image denoising for Rician noise using a novel hybrid transformer‐CNN network (HTC‐net) and self‐supervised pretraining

COLLABORATIVE NON-LOCAL MEANS DENOISING OF MAGNETIC RESONANCE IMAGES.

Denoising of 3D MR images using a voxel-wise hybrid residual MLP-CNN model to improve small lesion diagnostic confidence

Denoising of 3D Magnetic Resonance Images Using a Residual Encoder-Decoder Wasserstein Generative Adversarial Network.

MFCA-MICNN: a convolutional neural network with multiscale fast channel attention and multibranch irregular convolution for noise removal in dMRI

MRI Denoising Using Progressively Distribution-Based Neural Network

MRI denoising with a non‐blind deep complex‐valued convolutional neural network

Denoising of MR and CT images using cascaded multi-supervision convolutional neural networks with progressive training

MR-self Noise2Noise: self-supervised deep learning-based image quality improvement of submillimeter resolution 3D MR images

Msdra-Net: A Multi-Scale Attention-Guided Network For Magnetic Resonance Image Restoration

Denoising 3-D Magnitude Magnetic Resonance Images Based on Weighted Nuclear Norm Minimization.

An Improved Attentive Residue Multi-Dilated Network for Thermal Noise Removal in Magnetic Resonance Images

DESN: an Unsupervised MR Image Denoising Network with Deep Image Prior

Improved Magnetic Resonance Myelin Water Imaging Using Multi-Channel Denoising Convolutional Neural Networks (Mcdncnn).