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.
翻译:最近,基于分数的传播模型在MRI重建中表现出令人满意的表现,这些方法大多要求大量完全抽样的MRI数据作为培训组,有时在实践中很难获得。本文建议为MRI重建采用完全抽样的无数据分数的传播模型,在之前以自我监督的方式,对抽样不足的数据进行充分抽样的MR图像学习。具体地说,我们首先从Bayesian深度学习的未充分抽样的数据中推断出完全抽样的MR图像分布,然后通过培训一个分数函数来破坏数据分布并估计其概率密度梯度。把学过分的分函数作为以前的一种利用,我们可以通过按条件进行Langevin Markov连锁Monte Carlo(MC)取样来重建MR的图像。对公共数据集的实验表明,拟议的方法比现有的自我监督的MRI重建方法要好,并取得与常规的(完全抽样的数据)分数的传播方法可比的业绩。