IEEE Trans Med Imaging. 2022 Oct;41(10):2856-2866. doi: 10.1109/TMI.2022.3174513. Epub 2022 Sep 30.
Cephalometric analysis relies on accurate detection of craniomaxillofacial (CMF) landmarks from cone-beam computed tomography (CBCT) images. However, due to the complexity of CMF bony structures, it is difficult to localize landmarks efficiently and accurately. In this paper, we propose a deep learning framework to tackle this challenge by jointly digitalizing 105 CMF landmarks on CBCT images. By explicitly learning the local geometrical relationships between the landmarks, our approach extends Mask R-CNN for end-to-end prediction of landmark locations. Specifically, we first apply a detection network on a down-sampled 3D image to leverage global contextual information to predict the approximate locations of the landmarks. We subsequently leverage local information provided by higher-resolution image patches to refine the landmark locations. On patients with varying non-syndromic jaw deformities, our method achieves an average detection accuracy of 1.38± 0.95mm, outperforming a related state-of-the-art method.
头影测量分析依赖于从锥形束 CT（CBCT）图像中准确检测颅面（CMF）标志点。然而，由于 CMF 骨骼结构的复杂性，很难有效地、准确地定位标志点。在本文中，我们提出了一个深度学习框架，通过联合数字化 CBCT 图像上的 105 个 CMF 标志点来解决这个挑战。通过明确学习标志点之间的局部几何关系，我们的方法扩展了 Mask R-CNN，用于地标位置的端到端预测。具体来说，我们首先在降采样的 3D 图像上应用检测网络，利用全局上下文信息来预测标志点的大致位置。然后，我们利用来自更高分辨率图像补丁的局部信息来细化地标位置。在患有不同的非综合征性颌骨畸形的患者中，我们的方法实现了平均 1.38±0.95mm 的检测精度，优于相关的最先进方法。
