Chen Yan, Kecskemeti Steve R, Holmes James H, Corum Curtis A, Yaghoobi Nima, Magnotta Vincent A, Jacob Mathews
Electrical and Computer Engineering, University of Virginia, Charlottesville, Virginia, USA.
Waisman Center, University of Wisconsin-Madison, Madison, Wisconsin, USA.
Magn Reson Med. 2025 Sep;94(3):1191-1201. doi: 10.1002/mrm.30549. Epub 2025 Jun 2.
To develop a self-supervised and memory-efficient deep learning image reconstruction method for 4D non-Cartesian MRI with high resolution and a large parametric dimension.
The deep factor model (DFM) represents a parametric series of 3D multicontrast images using a neural network conditioned by the inversion time using efficient zero-filled reconstructions as input estimates. The model parameters are learned in a single-shot learning (SSL) fashion from the k-space data of each acquisition. A compatible transfer learning (TL) approach using previously acquired data is also developed to reduce reconstruction time. The DFM is compared to subspace methods with different regularization strategies in a series of phantom and in vivo experiments using the MPnRAGE acquisition for multicontrast imaging and quantitative estimation.
DFM-SSL improved the image quality and reduced bias and variance in quantitative estimates in both phantom and in vivo studies, outperforming all other tested methods. DFM-TL reduced the inference time while maintaining a performance comparable to DFM-SSL and outperforming subspace methods with multiple regularization techniques.
The proposed DFM offers a superior representation of the multicontrast images compared to subspace models, especially in the highly accelerated MPnRAGE setting. The self-supervised training is ideal for methods with both high resolution and a large parametric dimension, where training neural networks can become computationally demanding without a dedicated high-end GPU array.
开发一种用于高分辨率和大参数维度的4D非笛卡尔MRI的自监督且内存高效的深度学习图像重建方法。
深度因子模型（DFM）使用以反转时间为条件的神经网络，将一系列3D多对比度图像表示为参数形式，以高效的零填充重建作为输入估计。模型参数以单样本学习（SSL）方式从每次采集的k空间数据中学习。还开发了一种使用先前采集数据的兼容迁移学习（TL）方法，以减少重建时间。在一系列体模和体内实验中，使用MPnRAGE采集进行多对比度成像和定量估计，将DFM与具有不同正则化策略的子空间方法进行比较。
在体模和体内研究中，DFM-SSL均提高了图像质量，减少了定量估计中的偏差和方差，优于所有其他测试方法。DFM-TL减少了推理时间，同时保持了与DFM-SSL相当的性能，并且优于具有多种正则化技术的子空间方法。
与子空间模型相比，所提出的DFM对多对比度图像具有更好的表示能力，特别是在高度加速的MPnRAGE设置中。自监督训练对于具有高分辨率和大参数维度的方法非常理想，在没有专用高端GPU阵列的情况下，训练神经网络可能在计算上要求很高。
