基于层级循环神经网络的术中X线图像腰椎自动识别

Automatic Lumbar Vertebrae Recognition in Intraoperative X-Ray Images Based on Hierarchical Recurrent Neural Network

针对图像引导微创脊柱手术中移动C型臂X线成像特点,通过学习人体腰椎的曲率特征实现腰椎识别,提出一种基于层级循环神经网络的X线图像腰椎自动识别方法.首先为解决X线图像中腰椎纹理混叠的问题,提取腰椎三维模型与二维X线图像中共有的曲率特征作为模型的输入;其次为模拟术中移动C型臂多角度成像的特点,采用双向循环神经网络学习腰椎曲率特征,刻画腰椎曲率特征在不同成像角度下的关联性;最后为解决病理情况下腰椎部分信息缺失的问题,提出一种层级循环神经网络模型,通过逐层融合的网络架构对人体腰椎间天然的上下文关系进行建模,提高模型在病理情况下的腰椎识别率.在开源数据集和术中移动C型臂X线图像上的实验结果表明,文中方法在正常情况和病理情况下的腰椎识别率均优于其他4种方法,且由于使用了数据量较少的二维曲率特征,该方法在训练和测试阶段的计算效率更高,更适合于术中图像引导的应用.

According to the characteristic of mobile C-arm X-ray imaging in image-guided minimally invasive spine surgery,an automatic lumbar vertebrae recognition method is proposed,which based on hierarchical recurrent neural network.Its purpose is to identify lumbar vertebrae automatically by learning the curvature features.First,in order to solve the problem of lumbar vertebrae texture overlapping in X-ray images,the curvature features of 3D lumbar vertebrae model,which are common to the 2D X-ray images,are taken as the input of the model.Second,in order to simulate the multi-view imaging of intraoperative C-arm,the bidirectional recurrent neural network is exploited to learn the correlation of lumbar curvature features at different imaging angles.Finally,in order to solve the problem of partial occlusion of the lumbar vertebrae in the pathological condition,a hierarchical recurrent neural network model is introduced.The natural context between human lumbar vertebrae is modeled by the layer-by-layer fusion architecture to improve the recognition rate in the pathological condition.The results of the verification on open source datasets and intraoperative mobile C-arm X-ray images show that the lumbar vertebrae recognition rate of the proposed method is superior to the other four methods in both normal and pathological conditions.Furthermore,due to the utilization of two-dimensional curvature features,the proposed method is more efficient in the training and testing phases,and more suitable for applications in intraoperative image-guided navigation.