Deformable image registration is fundamental for many medical image analyses. A key obstacle for accurate image registration lies in image appearance variations such as the variations in texture, intensities, and noise. These variations are readily apparent in medical images, especially in brain images where registration is frequently used. Recently, deep learning-based registration methods (DLRs), using deep neural networks, have shown computational efficiency that is several orders of magnitude faster than traditional optimization-based registration methods (ORs). DLRs rely on a globally optimized network that is trained with a set of training samples to achieve faster registration. DLRs tend, however, to disregard the target-pair-specific optimization inherent in ORs and thus have degraded adaptability to variations in testing samples. This limitation is severe for registering medical images with large appearance variations, especially since few existing DLRs explicitly take into account appearance variations. In this study, we propose an Appearance Adjustment Network (AAN) to enhance the adaptability of DLRs to appearance variations. Our AAN, when integrated into a DLR, provides appearance transformations to reduce the appearance variations during registration. In addition, we propose an anatomy-constrained loss function through which our AAN generates anatomy-preserving transformations. Our AAN has been purposely designed to be readily inserted into a wide range of DLRs and can be trained cooperatively in an unsupervised and end-to-end manner. We evaluated our AAN with three state-of-the-art DLRs on three well-established public datasets of 3D brain magnetic resonance imaging (MRI). The results show that our AAN consistently improved existing DLRs and outperformed state-of-the-art ORs on registration accuracy, while adding a fractional computational load to existing DLRs.
翻译:对许多医学图像分析来说,变形图像登记是许多医学图像分析的基础。准确图像登记的一个关键障碍在于图像外观变异,如纹理、强度和噪音的变化。这些变异在医学图像中显而易见,特别是在经常使用注册的大脑图像中。最近,深层学习基注册方法(DLRs)使用深层神经网络显示的计算效率比传统优化注册方法(ORs)要快得多。DLRs依靠一个全球优化网络,该网络经过一系列培训,能够更快地实现注册。但是,DLRs倾向于无视ORs固有的特定目标面板优化,从而降低了测试样本中的变化适应性。这一限制对于以大外观变化登记医疗图像来说是十分严格的,特别是因为现有的DLRs(DLRs)明确考虑到外观变化。我们提议了一个“AANAAN调整网络”来提高DLRs的适应性适应性。我们AAN的变异性,我们AAN(AAN)在融入DRRRRRR时提供了外观的外观变形变异。此外,我们提出“AARLRRRRA-A-A-ARTM-A-A-ART-ADM-ARTM-LS-LS-LS-LS-LS-LS-LS-LS-LS-ALS-ALS-LS-A-A-A-S-LS-LS-LS-A-S-S-ART-ART-ART-ART-ART-ART-ART-ART-AD-S-S-AD-AD-A-A-AD-AD-A-A-ART-A-A-A-A-A-A-ART-A-A-A-A-A-A-A-A-A-A-A-A-A-A-A-A-A-AD-A-A-A-A-A-A-A-A-AD-R-R-R-A-A-A-A-A-A-ART-A-A-A-A-ART-A-A-A-A-A-A-ART-A-N-A