Overall Bone Structure as Assessed by Slice-by-Slice Profile
Overall Bone Structure as Assessed by Slice-by-Slice Profile
复制标题
DOI：
10.1007/s11692-019-09486-6
复制
发表时间：
2019-10-14
影响因子：
2.5
通讯作者：
Amson, Eli
中科院分区：
文献类型：
--
作者：
Amson, Eli
量化骨骼的内部结构是处理具有骨化骨骼的动物的表型进化的各种分析的核心。计算机断层扫描可以评估整个骨骼元素内骨组织的再分配。这种分析的两个重要参数是整体紧凑性（Cg）和总横截面积（Tt.Ar）。然而，没有开源的，时间效率高的方法可用于获取整个骨骼的这些参数。还需要一种方法来评估这些参数沿着所研究的骨的解剖轴之一的轮廓的变化。在这里，我提出了一个ImageJ宏和相关的R脚本，使用逐切片的方法自动获取Cg和Tt.Ar沿着感兴趣的骨骼元素的轴。不需要手动分割，只要感兴趣的骨骼是孤立的，并且每个切片上的元素最大，就可以在分析的扫描上存在多个骨骼。虽然一些偏见可能涉及的自动采集，半自动切片排除和校正程序可以用来有效地占it.As一个测试用例，亩CT数据收集超过70哺乳动物的中腰椎。两个评价的校正程序被证明表现同样好，与一个依赖于排除局部离群值的轻微优势。所提出的宏允许有效地构建与骨内部结构的量化有关的数据集。代码是现成的，进一步改进的方法和调整，以特定的需要可以很容易地进行。
Quantifying the inner structure of bones is central to various analyses dealing with the phenotypic evolution of animals with an ossified skeleton. Computed tomography allows to assess the repartition of bone tissue within an entire skeletal element. Two parameters of importance for such analyses are the global compactness (Cg) and total cross-sectional area (Tt.Ar). However, no open-source, time-efficient methods are available to acquire these parameters for whole bones. A methodology to assess the variation of these parameters along a profile following one of the studied bone's anatomical axes is also wanting. Here I present an ImageJ macro and associated R script to automatically acquire Cg and Tt.Ar along an axis of the skeletal element of interest using a slice-by-slice approach. No manual segmentation is required and several bones can be present on the analysed scan, as long as the bone of interest is isolated and the largest element on each slice. While some bias might be involved by the automatic acquisition, semi-automatic slice exclusion and correction procedures can be used to efficiently account for it. As a test case, mu CT-data was gathered for the mid-lumbar vertebra of over 70 mammals. The two evaluated correction procedures proved to perform equally well, with a slight advantage for the one relying on the exclusion of local outliers. The presented macro allows to efficiently build a dataset concerned with the quantification of bone inner structure. The code being readily available, further improvement of the methodology and adjustment to particular needs can be easily performed.