IEEE Trans Med Imaging. 2021 Oct;40(10):2939-2953. doi: 10.1109/TMI.2021.3052972. Epub 2021 Sep 30.
Anisotropic multi-slice Cardiac Magnetic Resonance (CMR) Images are conventionally acquired in patient-specific short-axis (SAX) orientation. In specific cardiovascular diseases that affect right ventricular (RV) morphology, acquisitions in standard axial (AX) orientation are preferred by some investigators, due to potential superiority in RV volume measurement for treatment planning. Unfortunately, due to the rare occurrence of these diseases, data in this domain is scarce. Recent research in deep learning-based methods mainly focused on SAX CMR images and they had proven to be very successful. In this work, we show that there is a considerable domain shift between AX and SAX images, and therefore, direct application of existing models yield sub-optimal results on AX samples. We propose a novel unsupervised domain adaptation approach, which uses task-related probabilities in an attention mechanism. Beyond that, cycle consistency is imposed on the learned patient-individual 3D rigid transformation to improve stability when automatically re-sampling the AX images to SAX orientations. The network was trained on 122 registered 3D AX-SAX CMR volume pairs from a multi-centric patient cohort. A mean 3D Dice of 0.86 ± 0.06 for the left ventricle, 0.65 ± 0.08 for the myocardium, and 0.77 ± 0.10 for the right ventricle could be achieved. This is an improvement of 25% in Dice for RV in comparison to direct application on axial slices. To conclude, our pre-trained task module has neither seen CMR images nor labels from the target domain, but is able to segment them after the domain gap is reduced. Code: https://github.com/Cardio-AI/3d-mri-domain-adaptation.
传统上，各向异性多切片心脏磁共振（CMR）图像是在患者特定的短轴（SAX）方向采集的。在某些特定的心血管疾病影响右心室（RV）形态的情况下，一些研究人员更喜欢在标准轴向（AX）方向采集图像，因为这在 RV 容积测量方面可能具有优势，有利于治疗计划。不幸的是，由于这些疾病的罕见发生，该领域的数据很少。基于深度学习的方法的最新研究主要集中在 SAX CMR 图像上，并且已经证明它们非常成功。在这项工作中，我们表明，AX 和 SAX 图像之间存在相当大的域转移，因此，直接应用现有模型在 AX 样本上的效果不理想。我们提出了一种新的无监督域自适应方法，该方法在注意力机制中使用与任务相关的概率。除此之外，还对学习到的患者个体 3D 刚性变换施加循环一致性，以在自动将 AX 图像重新采样到 SAX 方向时提高稳定性。该网络是在一个多中心患者队列的 122 个已注册的 3D AX-SAX CMR 体积对上进行训练的。左心室的平均 3D Dice 为 0.86 ± 0.06，心肌为 0.65 ± 0.08，右心室为 0.77 ± 0.10。与直接在轴向切片上应用相比，RV 的 Dice 提高了 25%。总之，我们的预训练任务模块既没有看到过 CMR 图像，也没有看到过目标域的标签，但在缩小域差距后，它能够对其进行分割。代码：https://github.com/Cardio-AI/3d-mri-domain-adaptation。
