School of Automation, Central South University, Changsha 410083, China.
Hunan Xiangjiang Artificial Intelligence Academy, Changsha 410083, China.
Sensors (Basel). 2021 Sep 18;21(18):6254. doi: 10.3390/s21186254.
Medical image registration is an essential technique to achieve spatial consistency geometric positions of different medical images obtained from single- or multi-sensor, such as computed tomography (CT), magnetic resonance (MR), and ultrasound (US) images. In this paper, an improved unsupervised learning-based framework is proposed for multi-organ registration on 3D abdominal CT images. First, the explored coarse-to-fine recursive cascaded network (RCN) modules are embedded into a basic U-net framework to achieve more accurate multi-organ registration results from 3D abdominal CT images. Then, a topology-preserving loss is added in the total loss function to avoid a distortion of the predicted transformation field. Four public databases are selected to validate the registration performances of the proposed method. The experimental results show that the proposed method is superior to some existing traditional and deep learning-based methods and is promising to meet the real-time and high-precision clinical registration requirements of 3D abdominal CT images.
医学图像配准是一种将不同医学图像（如计算机断层扫描（CT）、磁共振（MR）和超声（US）图像）从单传感器或多传感器中获得的空间一致的几何位置的关键技术。本文提出了一种基于改进的无监督学习的框架，用于 3D 腹部 CT 图像上的多器官配准。首先，将探索的粗到精递归级联网络（RCN）模块嵌入到基本 U-net 框架中，以实现从 3D 腹部 CT 图像中更准确的多器官配准结果。然后，在总损失函数中添加保持拓扑结构的损失，以避免预测变换场的失真。选择了四个公共数据库来验证所提出方法的配准性能。实验结果表明，所提出的方法优于一些现有的传统和基于深度学习的方法，有望满足 3D 腹部 CT 图像实时和高精度的临床配准要求。
