MOCOnet: Robust Motion Correction of Cardiovascular Magnetic Resonance T1 Mapping Using Convolutional Neural Networks.
MOCOnet: Robust Motion Correction of Cardiovascular Magnetic Resonance T1 Mapping Using Convolutional Neural Networks.
复制标题
DOI：
10.3389/fcvm.2021.768245
复制
发表时间：
2021
影响因子：
3.6
通讯作者：
Piechnik SK
中科院分区：
文献类型：
--
作者：
Gonzales RA;Zhang Q;Papież BW;Werys K;Lukaschuk E;Popescu IA;Burrage MK;Shanmuganathan M;Ferreira VM;Piechnik SK
背景资料：定量心血管磁共振（CMR）T1标测已显示出在常规临床实践中用于高级组织表征的前景。然而，T1标测容易出现运动伪影，这会影响其鲁棒性和临床解释。目前的T1标测运动校正方法是模型驱动的，不能保证通用性，限制了其广泛使用。相比之下，新兴的数据驱动的深度学习方法在一般图像配准任务中表现出良好的性能。我们提出了MOCOnet，一种卷积神经网络解决方案，用于T1地图中的一般化运动伪影校正。研究方法：该网络架构采用U-Net来产生距离矢量场，并利用扭曲层以粗到细的方式对特征图应用变形。使用在1.5T下扫描的UK Biobank成像数据集，MOCOnet在1，536个具有运动伪影的中心室T1图（使用ShMOLLI方法获取）上进行训练，这些伪影由定制的变形程序生成，并在具有不同运动范围的200个样本的不同集合上进行测试。将MOCOnet与经过充分验证的基线多模态图像配准方法进行了比较。运动减少由3名人类专家进行视觉评估，运动评分范围从0%（严格无运动）到100%（非常严重的运动）。结果：MOCOnet实现了快速图像配准（每个T1图<1秒），并成功抑制了大范围的运动伪影。MOCOnet将运动评分从37.1±21.5显著降低至13.3±10.5（p < 0.001），而基线方法将其降低至15.8±15.6（p < 0.001）。MOCOnet在抑制运动伪影方面明显优于基线方法，并且更加一致（p = 0.007）。结论：与传统的图像配准方法相比，MOCOnet表现出明显更好的运动校正性能。以鲁棒性和时间有效的方式挽救受运动影响的数据可以实现更好的图像质量和可靠的图像以用于立即的临床解释。
Background: Quantitative cardiovascular magnetic resonance (CMR) T1 mapping has shown promise for advanced tissue characterisation in routine clinical practise. However, T1 mapping is prone to motion artefacts, which affects its robustness and clinical interpretation. Current methods for motion correction on T1 mapping are model-driven with no guarantee on generalisability, limiting its widespread use. In contrast, emerging data-driven deep learning approaches have shown good performance in general image registration tasks. We propose MOCOnet, a convolutional neural network solution, for generalisable motion artefact correction in T1 maps. Methods: The network architecture employs U-Net for producing distance vector fields and utilises warping layers to apply deformation to the feature maps in a coarse-to-fine manner. Using the UK Biobank imaging dataset scanned at 1.5T, MOCOnet was trained on 1,536 mid-ventricular T1 maps (acquired using the ShMOLLI method) with motion artefacts, generated by a customised deformation procedure, and tested on a different set of 200 samples with a diverse range of motion. MOCOnet was compared to a well-validated baseline multi-modal image registration method. Motion reduction was visually assessed by 3 human experts, with motion scores ranging from 0% (strictly no motion) to 100% (very severe motion). Results: MOCOnet achieved fast image registration (<1 second per T1 map) and successfully suppressed a wide range of motion artefacts. MOCOnet significantly reduced motion scores from 37.1±21.5 to 13.3±10.5 (p < 0.001), whereas the baseline method reduced it to 15.8±15.6 (p < 0.001). MOCOnet was significantly better than the baseline method in suppressing motion artefacts and more consistently (p = 0.007). Conclusion: MOCOnet demonstrated significantly better motion correction performance compared to a traditional image registration approach. Salvaging data affected by motion with robustness and in a time-efficient manner may enable better image quality and reliable images for immediate clinical interpretation.