School of Information Engineering, Zhengzhou University, Zhengzhou 450001, Henan, People's Republic of China.
Collaborative Innovation Center for Internet Healthcare, Zhengzhou University, Zhengzhou 450052, Henan, People's Republic of China.
Phys Med Biol. 2022 Jul 4;67(14). doi: 10.1088/1361-6560/ac79fa.
Medical image registration aims to find the deformation field that can align two images in a spatial position. A medical image registration method based on U-Net architecture has been proposed currently. However, U-Net architecture has few training parameters, which leads to weak learning ability, and it ignores the adverse effects of image noise on the registration accuracy. The article aims at addressing the problem of weak network learning ability and the adverse effects of noisy images on registration.Here we propose a novel unsupervised 3D brain image registration framework, which introduces the residual unit and singular value decomposition (SVD) denoising layer on the U-Net architecture. Residual unit solves the problem of network degradation, that is, registration accuracy becomes saturated and then degrades rapidly with the increase in network depth. SVD denoising layer uses the estimated model order for SVD-based low-rank image reconstruction. we use Akaike information criterion to estimate the appropriate model order, which is used to remove noise components. We use the exponential linear unit (ELU) as the activation function, which is more robust to noise than other peers.The proposed method is evaluated on the publicly available brain MRI datasets: Mindboggle101 and LPBA40. Experimental results demonstrate our method outperforms several state-of-the-art methods for the metric of Dice Score. The mean number of folding voxels and registration time are comparable to state-of-the-art methods.This study shows that Deep Residual-SVD Network can improve registration accuracy. This study also demonstrate that the residual unit can enhance the learning ability of the network, the SVD denoising layer can denoise effectively, and the ELU is more robust to noise.
医学图像配准旨在找到可以在空间位置上对齐两幅图像的变形场。目前已经提出了一种基于 U-Net 架构的医学图像配准方法。然而，U-Net 架构的训练参数较少，导致学习能力较弱，并且忽略了图像噪声对配准精度的不利影响。本文旨在解决网络学习能力弱和噪声图像对配准不利的问题。在此，我们提出了一种新颖的无监督 3D 脑图像配准框架，该框架在 U-Net 架构上引入了残差单元和奇异值分解（SVD）去噪层。残差单元解决了网络退化的问题，即随着网络深度的增加，配准精度会饱和，然后迅速下降。SVD 去噪层使用基于 SVD 的低秩图像重建的估计模型阶数。我们使用赤池信息量准则估计合适的模型阶数，用于去除噪声分量。我们使用指数线性单元（ELU）作为激活函数，它比其他同行对噪声更鲁棒。该方法在公开的脑 MRI 数据集 Mindboggle101 和 LPBA40 上进行了评估。实验结果表明，在 Dice 得分的度量标准上，我们的方法优于几种最先进的方法。折叠体素的平均数量和注册时间与最先进的方法相当。这项研究表明，深度残差-SVD 网络可以提高配准精度。这项研究还表明，残差单元可以增强网络的学习能力，SVD 去噪层可以有效地去噪，ELU 对噪声更鲁棒。
