Self-Score: Self-Supervised Learning on Score-Based Models for MRI
Reconstruction
- URL: http://arxiv.org/abs/2209.00835v1
- Date: Fri, 2 Sep 2022 06:21:42 GMT
- Title: Self-Score: Self-Supervised Learning on Score-Based Models for MRI
Reconstruction
- Authors: Zhuo-Xu Cui, Chentao Cao, Shaonan Liu, Qingyong Zhu, Jing Cheng,
Haifeng Wang, Yanjie Zhu, Dong Liang
- Abstract summary: This paper proposes a fully-sampled-data-free score-based diffusion model for MRI reconstruction.
It learns the fully sampled MR image prior in a self-supervised manner on undersampled data.
Experiments on the public dataset show that the proposed method outperforms existing self-supervised MRI reconstruction methods.
- Score: 18.264778497591603
- License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
- Abstract: Recently, score-based diffusion models have shown satisfactory performance in
MRI reconstruction. Most of these methods require a large amount of fully
sampled MRI data as a training set, which, sometimes, is difficult to acquire
in practice. This paper proposes a fully-sampled-data-free score-based
diffusion model for MRI reconstruction, which learns the fully sampled MR image
prior in a self-supervised manner on undersampled data. Specifically, we first
infer the fully sampled MR image distribution from the undersampled data by
Bayesian deep learning, then perturb the data distribution and approximate
their probability density gradient by training a score function. Leveraging the
learned score function as a prior, we can reconstruct the MR image by
performing conditioned Langevin Markov chain Monte Carlo (MCMC) sampling.
Experiments on the public dataset show that the proposed method outperforms
existing self-supervised MRI reconstruction methods and achieves comparable
performances with the conventional (fully sampled data trained) score-based
diffusion methods.
Related papers
- Highly Accelerated MRI via Implicit Neural Representation Guided Posterior Sampling of Diffusion Models [2.5412006057370893]
Implicit neural representation (INR) has emerged as a powerful paradigm for solving inverse problems.
Our proposed framework can be a generalizable framework to solve inverse problems in other medical imaging tasks.
arXiv Detail & Related papers (2024-07-03T01:37:56Z) - Optimizing Sampling Patterns for Compressed Sensing MRI with Diffusion
Generative Models [75.52575380824051]
We present a learning method to optimize sub-sampling patterns for compressed sensing multi-coil MRI.
We use a single-step reconstruction based on the posterior mean estimate given by the diffusion model and the MRI measurement process.
Our method requires as few as five training images to learn effective sampling patterns.
arXiv Detail & Related papers (2023-06-05T22:09:06Z) - Dual-Domain Self-Supervised Learning for Accelerated Non-Cartesian MRI
Reconstruction [14.754843942604472]
We present a fully self-supervised approach for accelerated non-Cartesian MRI reconstruction.
In training, the undersampled data are split into disjoint k-space domain partitions.
For the image-level self-supervision, we enforce appearance consistency obtained from the original undersampled data.
arXiv Detail & Related papers (2023-02-18T06:11:49Z) - Iterative Data Refinement for Self-Supervised MR Image Reconstruction [18.02961646651716]
We propose a data refinement framework for self-supervised MR image reconstruction.
We first analyze the reason of the performance gap between self-supervised and supervised methods.
Then, we design an effective self-supervised training data refinement method to reduce this data bias.
arXiv Detail & Related papers (2022-11-24T06:57:16Z) - Towards performant and reliable undersampled MR reconstruction via
diffusion model sampling [67.73698021297022]
DiffuseRecon is a novel diffusion model-based MR reconstruction method.
It guides the generation process based on the observed signals.
It does not require additional training on specific acceleration factors.
arXiv Detail & Related papers (2022-03-08T02:25:38Z) - MRI Reconstruction via Data Driven Markov Chain with Joint Uncertainty
Estimation [3.5751623095926806]
We introduce a framework that enables efficient sampling from learned probability distributions for MRI reconstruction.
The data-driven Markov chains are constructed from the generative model learned from a given image database.
The performance of the method is evaluated on an open dataset using 10-fold accelerated acquisition.
arXiv Detail & Related papers (2022-02-03T09:13:49Z) - Reference-based Magnetic Resonance Image Reconstruction Using Texture
Transforme [86.6394254676369]
We propose a novel Texture Transformer Module (TTM) for accelerated MRI reconstruction.
We formulate the under-sampled data and reference data as queries and keys in a transformer.
The proposed TTM can be stacked on prior MRI reconstruction approaches to further improve their performance.
arXiv Detail & Related papers (2021-11-18T03:06:25Z) - Score-based diffusion models for accelerated MRI [35.3148116010546]
We introduce a way to sample data from a conditional distribution given the measurements, such that the model can be readily used for solving inverse problems in imaging.
Our model requires magnitude images only for training, and yet is able to reconstruct complex-valued data, and even extends to parallel imaging.
arXiv Detail & Related papers (2021-10-08T08:42:03Z) - Robust Compressed Sensing MRI with Deep Generative Priors [84.69062247243953]
We present the first successful application of the CSGM framework on clinical MRI data.
We train a generative prior on brain scans from the fastMRI dataset, and show that posterior sampling via Langevin dynamics achieves high quality reconstructions.
arXiv Detail & Related papers (2021-08-03T08:52:06Z) - Learning Energy-Based Models by Diffusion Recovery Likelihood [61.069760183331745]
We present a diffusion recovery likelihood method to tractably learn and sample from a sequence of energy-based models.
After training, synthesized images can be generated by the sampling process that initializes from Gaussian white noise distribution.
On unconditional CIFAR-10 our method achieves FID 9.58 and inception score 8.30, superior to the majority of GANs.
arXiv Detail & Related papers (2020-12-15T07:09:02Z) - Improved Slice-wise Tumour Detection in Brain MRIs by Computing
Dissimilarities between Latent Representations [68.8204255655161]
Anomaly detection for Magnetic Resonance Images (MRIs) can be solved with unsupervised methods.
We have proposed a slice-wise semi-supervised method for tumour detection based on the computation of a dissimilarity function in the latent space of a Variational AutoEncoder.
We show that by training the models on higher resolution images and by improving the quality of the reconstructions, we obtain results which are comparable with different baselines.
arXiv Detail & Related papers (2020-07-24T14:02:09Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.