论文阅读-使用CNN从双平面X光片走向自动化3D脊柱重建以进行统计脊柱模型拟合

Toward Automated 3D Spine Reconstruction from Biplanar Radiographs Using CNN for Statistical Spine Model Fitting----使用CNN从双平面X光片走向自动化3D脊柱重建以进行统计脊柱模型拟合

来源： IEEE Transactions on Medical Imaging（TMI ）

时间：2019 .11

作者：B. Aubert， Carlos Vazquez等 （加拿大蒙特利尔CHUM研究中心）

⚡️ 初步阅读笔记

📑作者为什么研究这个课题？

To date, 3D spine reconstruction from biplanar radiographs involves intensive user supervision and semi-automated methods that are time-consuming and not effective in clinical routine.

现有的半监督实时的方法耗时且不能应用于临床

📑文献解决的问题？

This paper proposes a new, fast, and automated 3D spine reconstruction method through which a realistic statistical shape model of the spine is fifitted to images using convolutional neural networks (CNN). The CNNs automatically detect the anatomical landmarks controlling the spine model deformation through a hier archical and gradual iterative process.

本文提出了一种新的，快速且自动化的3D脊柱重建方法，通过该方法，可以使用卷积神经网络（CNN）将逼真的脊柱统计形状模型拟合到图像。 CNN通过分层和逐步的迭代过程自动检测控制脊柱模型变形的解剖标志。

(用的是校准的双平面放射图–PA和LAT)

The main challenge of using a stereoradiographic imaging system to reconstruct bones in 3D is to accurately estimate 3D models from partial information (only two projections). Moreover, the visibility of important anatomical landmarks can vary considerably across patients in radiographs due to the overlap of structures in projection (soft tissues, organs, air cavities), noise and poor contrast [17], [18]. Hence, reconstructing a spine in 3D from two radiographs remains a challenging problem.

使用立体射线照相成像系统以3D形式重建骨骼的主要挑战是从局部信息（仅两个投影