[论文解读] MR image reconstruction using the learned data distribution as prior.
本文提出了一种新颖的MR图像重建方法,该方法利用变分自编码器(VAE)学习MR图像块的概率分布,并将其作为最大后验概率(MAP)框架中的非线性先验。该方法在重建精度和病灶保留方面优于总变差(TV)、字典学习和ADMM-Net方法,在2倍和3倍 undersampling 比例下分别实现了2.77%和4.29%的RMSE。
MR image reconstruction from undersampled data exploits priors to compensate for missing k-space data. This has previously been achieved by using regularization methods, such as TV and wavelets, or data adaptive methods, such as dictionary learning. We propose to explicitly learn the probability distribution of MR image patches and to constrain patches to have a high probability according to this distribution in reconstruction, effectively employing it as the prior. We use variational autoencoders (VAE) to learn the distribution of MR image patches. This high dimensional distribution is modelled by a latent parameter model of lower dimensions in a non-linear fashion. We develop a reconstruction algorithm that uses the learned prior in a Maximum-A-Posteriori estimation formulation. We evaluate the proposed method with T1 weighted images and compare it to existing alternatives. We also apply our method on images with white matter lesions. Visual evaluation of the samples drawn from the learned model showed that the VAE algorithm was able to approximate the distribution of MR image patches. Furthermore, the reconstruction algorithm using the approximate distribution produced qualitatively better results. The proposed technique achieved RMSE, CNR and CN values of 2.77%, 0.43, 0.11 and 4.29%, 0.43, 0.11 for undersampling ratios of 2 and 3, respectively. It outperformed other evaluated methods in terms of used metrics. In the experiments on images with white matter lesions, the method faithfully reconstructed the lesions. We introduced a novel method for MR reconstruction, which takes a new perspective on regularization by learning priors. Results suggest the method compares favorably against TV and dictionary based methods as well as the neural-network based ADMM-Net in terms of the RMSE, CNR and CN and perceptual image quality and can reconstruct lesions as well.
研究动机与目标
- 解决从k空间欠采样数据重建高质量MR图像的挑战。
- 通过学习数据驱动的非线性先验,改进传统正则化方法(如总变差和小波变换)。
- 开发一种利用图像块学习分布作为概率先验的重建框架。
- 在健康T1加权图像和存在白质病灶的病理扫描上评估该方法。
- 与最先进方法相比,展示在定量指标和感知图像质量方面的优越性能。
提出的方法
- 在完全采样的MR图像中提取的图像块上训练变分自编码器(VAE),以学习底层数据分布。
- VAE通过使用非线性深度神经网络映射,将图像块的高维分布建模于低维潜在空间中。
- 利用学习到的VAE定义最大后验概率(MAP)估计框架中的先验分布,用于图像重建。
- 重建算法在保持与欠采样k空间数据一致的同时,强制满足学习到的图像块分布的高似然性。
- 该方法将VAE先验整合到迭代优化方案中,平衡数据保真度与先验一致性。
- 该方法实现了端到端重建,无需显式正则化项(如TV或稀疏性约束)。
实验结果
研究问题
- RQ1深度生成模型(如VAE)是否能有效捕捉MR图像块的复杂高维分布?
- RQ2与经典正则化方法相比,使用学习到的基于VAE的先验是否能提升MR图像重建质量?
- RQ3与现有最先进技术相比,该方法在存在白质病灶的病理图像上的表现如何?
- RQ4在高undersampling比例下,VAE先验在多大程度上提升了重建精度?
- RQ5与传统先验相比,基于VAE的先验是否能更好地保留细微解剖结构和病灶?
主要发现
- 通过生成图像块的可视化采样,证实VAE成功近似了MR图像块的分布。
- 在undersampling比例为2和3时,该方法分别实现了2.77%和4.29%的RMSE,优于对比方法。
- 在两种undersampling比例下,该方法将对比噪声比(CNR)提升至0.43,相关系数(CN)提升至0.11。
- 在存在白质病灶的图像中,该方法忠实地重建了病灶结构,保留了关键病理细节。
- 在感知评估中,该方法的重建质量更优,伪影更少,结构保真度更高,优于TV、字典学习和ADMM-Net。
- 结果表明,通过VAE学习数据驱动的非线性先验,相比手工设计或线性先验,在MR图像重建中表现更优。
更好的研究,从现在开始
从论文设计到论文写作,大幅缩短您的研究时间。
无需绑定信用卡
本解读由 AI 生成,并经人工编辑审核。