基于自适应形变模型的胸部CT图像肺组织分割

肺组织分割是肺结节检测、肺功能定量分析、三维重建与可视化计算等胸部CT图像分析处理的基础。为此，首先提出了一种改进的自适应形变模型（T-Snake模型）。然后基于改进的T-Snake模型给出胸部CT图像肺组织分割方法。上述模型能够解决基本T-Snake模型的自交（Self-collisions）问题，有着更高的数值计算精度；并且，对T-Snake模型的节点更新算法也做了改进。使模型节点的更新速度加快。实验结果表明，方法准确性和可靠性较高，有着较好的应用前景。

Segmentation of Lung Parenchyma in Chest CT Image Based on T-Snakes

The segmentation of lung parenchyma is the foundation of further processing, such as lung nodule detection, the quantitative analysis of lung function, three-dimensional reconstruction and visualization analysis. Therefore, an improved topologically adaptable snake (T-Snake) was presented, and then a method was given to segment lung parenchyma in chest CT image based on the improved T-Snake. The improved T-Snake can avoid the Self-collisions of the T-Snake and has the advantage of higher numerical precision. The nodes updating algorithm was improved which could accelerate updating speed. The experiment shows that this method has good accuracy, robustness and future applications.