基于LUT的快速3D气道树骨架线提取

A fast LUT-based airway skeleton extraction algorithm for virtual bronchoscopy

摘要:

为提高肺部支气管骨架线的提取效率，提出并使用一种基于look-up-table (LUT) 的腐蚀细化算法.分析建立腐蚀模型，并根据该模型优化建立了LUT，以该LUT为依据通过索引查找对原始数据进行快速腐蚀细化，对得到的腐蚀结果进行剪枝处理以得到最终的骨架线.实验结果表明：提取过程中LUT的应用从根本上降低了腐蚀细化中判断的复杂度，将复杂的简单点判断问题转化为LUT中的查询问题，从而极大地优化了腐蚀细化中关键的腐蚀过程.相比传统方法，基于LUT的腐蚀细化算法显著提高了骨架线的提取速度，较传统细化法提速近22.95倍.

Abstract:

In order to improve the efficiency of lung airway skeleton extraction, this paper introduces a new look-up-table (LUT) based thinning algorithm. This new approach consists of three major steps: the analysis and creation of the thinning model and the establishment of LUT based on above thinning model result. Thinning process by index-searching uses the LUT. Branch cutting based on the result of thinning obtains the final result. The usage of LUT index-searching during the thinning step transforms simple point judgments into LUT index search and has significantly improved the performance of the whole algorithm. Experimental results demonstrate that the new skeleton algorithm is 22.95 times faster than the existing thinning algorithm.