Wang Zhiwei, Lv Jinxin, Yang Yunqiao, Lin Yi, Li Qiang, Li Xin, Yang Xin
Britton Chance Center for Biomedical Photonics, Wuhan National Laboratory for Optoelectronics, Huazhong University of Science and Technology, Wuhan 430074, China.
MoE Key Laboratory for Biomedical Photonics, Collaborative Innovation Center for Biomedical Engineering, Huazhong University of Science and Technology, Wuhan 430074, China.
Fundam Res. 2022 Nov 10;4(6):1657-1665. doi: 10.1016/j.fmre.2022.10.014. eCollection 2024 Nov.
Vertebral landmark localization is a crucial step in various spine-related clinical applications, which requires detecting the corner points of 17 vertebrae. However, the neighboring landmarks often disturb each other because of the homogeneous appearance of vertebrae, making vertebral landmark localization extremely difficult. In this paper, we propose a multi-stage cascaded convolutional neural network (CNN) to split a single task into two sequential steps: center point localization to roughly locate 17 center points of vertebrae, and corner point localization to determine four corner points for each vertebra without any disturbance. The landmarks in each step were located gradually from a set of initialized points by regressing offsets using cascaded CNNs. To resist the mutual attraction of the vertebrae, principal component analysis was employed to preserve the shape constraint in offset regression. We evaluated our method on the AASCE dataset, comprising 609 tight spinal anteroposterior X-ray images, and each image contained 17 vertebrae composed of the thoracic and lumbar spine for spinal shape characterization. The experimental results demonstrated the superior performance of vertebral landmark localization over other state-of-the-art methods, with the relative error decreasing from to .
椎体标志点定位是各种脊柱相关临床应用中的关键步骤，这需要检测17个椎体的角点。然而，由于椎体外观相似，相邻标志点常常相互干扰，使得椎体标志点定位极其困难。在本文中，我们提出了一种多阶段级联卷积神经网络（CNN），将单个任务分为两个连续步骤：中心点定位以大致定位17个椎体的中心点，以及角点定位以在无任何干扰的情况下确定每个椎体的四个角点。通过使用级联CNN回归偏移量，在每个步骤中从一组初始化点逐步定位标志点。为了抵抗椎体之间的相互吸引力，采用主成分分析在偏移量回归中保留形状约束。我们在AASCE数据集上评估了我们的方法，该数据集包含609张紧密的脊柱前后位X射线图像，每张图像包含由胸椎和腰椎组成的17个椎体用于脊柱形状表征。实验结果表明我们的椎体标志点定位方法相对于其他现有方法具有卓越的性能，相对误差从 降至 。
