SDnDTI: Self-supervised deep learning-based denoising for diffusion tensor MRI

@article{Tian2021SDnDTISD,
  title={SDnDTI: Self-supervised deep learning-based denoising for diffusion tensor MRI},
  author={Qiyuan Tian and Ziyu Li and Qiuyun Fan and J. Polimeni and Berkin Bilgiç and David H. Salat and Susie Yi Huang},
  journal={NeuroImage},
  year={2021},
  volume={253}
}

Figures and Tables from this paper

Supervised Denoising of Diffusion-Weighted Magnetic Resonance Images Using a Convolutional Neural Network and Transfer Learning

Visual comparisons made in the acquired brain images indicate that the denoised single-repetition images are less noisy than multi-rePETition averaged images.

Multiple B-Value Model-Based Residual Network (MORN) for Accelerated High-Resolution Diffusion-Weighted Imaging

The proposed MORN model outperformed conventional PI reconstruction and two state-of-the-art deep learning methods in terms of PSNR, Peak Signal-to-Noise Ratio, SSIM and apparent diffusion coefficient maps and achieved consistent fractional anisotrophy and mean diffusivity reconstructed from multiple diffusion directions.

An unsupervised convolutional neural network method for estimation of intravoxel incoherent motion parameters

The preliminary results suggest that it is feasible to estimate IVIM parameters using CNN, and the DNN- and CNN-based methods yielded less biased IVIM parameter estimates.

Sparse Feature Aware Noise Removal Technique for Brain Multiple Sclerosis Lesions using Magnetic Resonance Imaging

  • S. DAditya C R
  • Computer Science
    International Journal of Advanced Computer Science and Applications
  • 2022
Experimental results demonstrated that the model SFANR outperforms all other state-of-art noise removal techniques in terms of Peak-SignalNoise-Ratio (PSNR), Structural Similarity Index Metric (SSIM) with less running time.

Deep Learning for Image Enhancement and Correction in Magnetic Resonance Imaging-State-of-the-Art and Challenges.

This review paper provides a comprehensive overview of deep learning-based methods for post-processing MR images to enhance image quality and correct image artefacts and discusses the current limitations of the application of artificial intelligence in MRI.

A personalized deep learning denoising strategy for low-count PET images

This study demonstrated that in deep learning-based low dose PET denoising, noise levels in the training input images have a substantial impact on the model performance.

Low-rank Tensor Assisted K-space Generative Model for Parallel Imaging Reconstruction

A new idea is presented, low-rank tensor assisted k-space generative model (LR-KGM), for parallel imaging reconstruction, which transforms original prior information into high-dimensional prior information for learning.

High B-value diffusion tensor imaging for early detection of hippocampal microstructural alteration in a mouse model of multiple sclerosis

The findings stress the needs for both high b-values and sufficient NDIR to achieve a GM DTI with more biologically meaningful correlations, though DTI-metrics should be interpreted with caution in these settings.

Water/fat separation for self‐navigated diffusion‐weighted multishot echo‐planar imaging

The purpose of this study was to develop a self‐navigation strategy to improve scan efficiency and image quality of water/fat‐separated, diffusion‐weighted multishot echo‐planar imaging (ms‐EPI).

Recommendations and guidelines from the ISMRM Diffusion Study Group for preclinical diffusion MRI: Part 1 -- In vivo small-animal imaging

The value that small animal imaging adds to the field of dMRI is described, followed by general considerations and foundational knowledge that must be considered when designing experiments, and guidelines for in vivo acquisition protocols are given.

References

SHOWING 1-10 OF 130 REFERENCES

Fast learning of fiber orientation distribution function for MR tractography using convolutional neural network.

A method to reconstruct the fODF from downsampled diffusion-weighted images (DWIs) by leveraging the strong inference ability of the deep convolutional neural network (CNN) and exhibits promising potential in acquisition acceleration for the reconstruction of fODFs with good accuracy.

Patch2Self: Denoising Diffusion MRI with Self-Supervised Learning

A self-supervised learning method for denoising DWI data, Patch2Self, which uses the entire volume to learn a full-rank locally linear denoiser for that volume, and demonstrates the effectiveness via quantitative and qualitative improvements in microstructure modeling, tracking and model estimation relative to other unsupervised methods on real and simulated data.

Fast and Robust Diffusion Kurtosis Parametric Mapping Using a Three-Dimensional Convolutional Neural Network

This result suggests that it is possible to achieve kurtosis mapping in most clinical scanners within one minute, which could significantly extend the clinical utility of the DKI.

Deep Learning Based Noise Reduction for Brain MR Imaging: Tests on Phantoms and Healthy Volunteers

  • M. KidohKensuke Shinoda Y. Yamashita
  • Medicine
    Magnetic resonance in medical sciences : MRMS : an official journal of Japan Society of Magnetic Resonance in Medicine
  • 2019
dDLR reduces image noise while preserving image quality on brain MR images, and has equivalent or better image quality than NAQ5, and superior quality to that of NAQ2 (P < 0.05), for all criteria except artifact.

q-Space Deep Learning: Twelve-Fold Shorter and Model-Free Diffusion MRI Scans

It is demonstrated how deep learning, a group of algorithms based on recent advances in the field of artificial neural networks, can be applied to reduce diffusion MRI data processing to a single optimized step and how classical data processing can be streamlined by means of deep learning.

Denoising of 3D magnetic resonance images with multi-channel residual learning of convolutional neural network

The proposed MCDnCNN model has been demonstrated to robustly denoise three dimensional MR images with Rician noise to show the most robust denoising performance in all three datasets.

Highly accelerated, model‐free diffusion tensor MRI reconstruction using neural networks

DiffNet improved DTI reconstruction accuracy, precision, and tumor delineation performance in GBM while permitting reconstruction from only three diffusion-encoding directions.

Improved diffusion imaging through SNR‐enhancing joint reconstruction

A new approach to reduce noise while largely maintaining resolution in diffusion weighted images, using a statistical reconstruction method that takes advantage of the high level of structural correlation observed in typical datasets.

MR diffusion kurtosis imaging for neural tissue characterization

It is demonstrated that DKI offers a more comprehensive approach than DTI in describing the complex water diffusion process in vivo by estimating both diffusivity and kurtosis, which may provide improved sensitivity and specificity in MR diffusion characterization of neural tissues.
...