Fee Timothy, Downs Crawford, Eberhardt Alan, Zhou Yong, Berry Joel
Department of Biomedical Engineering, University of Alabama at Birmingham, Birmingham, Alabama.
Department of Ophthalmology, Center for Ocular Biomechanics and Biotransport, University of Alabama at Birmingham, Birmingham, Alabama.
J Biomed Mater Res A. 2016 Jul;104(7):1680-6. doi: 10.1002/jbm.a.35697. Epub 2016 Mar 17.
It is well documented that electrospun tissue engineering scaffolds can be fabricated with variable degrees of fiber alignment to produce scaffolds with anisotropic mechanical properties. Several attempts have been made to quantify the degree of fiber alignment within an electrospun scaffold using image-based methods. However, these methods are limited by the inability to produce a quantitative measure of alignment that can be used to make comparisons across publications. Therefore, we have developed a new approach to quantifying the alignment present within a scaffold from scanning electron microscopic (SEM) images. The alignment is determined by using the Sobel approximation of the image gradient to determine the distribution of gradient angles with an image. This data was fit to a Von Mises distribution to find the dispersion parameter κ, which was used as a quantitative measure of fiber alignment. We fabricated four groups of electrospun polycaprolactone (PCL) + Gelatin scaffolds with alignments ranging from κ = 1.9 (aligned) to κ = 0.25 (random) and tested our alignment quantification method on these scaffolds. It was found that our alignment quantification method could distinguish between scaffolds of different alignments more accurately than two other published methods. Additionally, the alignment parameter κ was found to be a good predictor the mechanical anisotropy of our electrospun scaffolds. The ability to quantify fiber alignment within and make direct comparisons of scaffold fiber alignment across publications can reduce ambiguity between published results where cells are cultured on "highly aligned" fibrous scaffolds. This could have important implications for characterizing mechanics and cellular behavior on aligned tissue engineering scaffolds. © 2016 Wiley Periodicals, Inc. J Biomed Mater Res Part A: 104A: 1680-1686, 2016.
有充分的文献记载，可通过控制不同程度的纤维排列来制备具有各向异性力学性能的电纺组织工程支架。人们已多次尝试使用基于图像的方法来量化电纺支架内的纤维排列程度。然而，这些方法的局限性在于无法生成可用于跨出版物进行比较的定量排列测量值。因此，我们开发了一种新方法，用于从扫描电子显微镜（SEM）图像中量化支架内的排列情况。通过使用图像梯度的Sobel近似来确定图像中梯度角的分布，从而确定排列情况。将此数据拟合到冯·米塞斯分布以找到离散参数κ，该参数用作纤维排列的定量测量值。我们制备了四组电纺聚己内酯（PCL）+明胶支架，其排列范围从κ = 1.9（排列整齐）到κ = 0.25（随机排列），并在这些支架上测试了我们的排列量化方法。结果发现，我们的排列量化方法比其他两种已发表的方法能更准确地区分不同排列的支架。此外，发现排列参数κ是我们电纺支架力学各向异性的良好预测指标。量化支架内纤维排列并直接比较不同出版物中支架纤维排列的能力，可以减少在“高度排列”的纤维支架上培养细胞的已发表结果之间的模糊性。这对于表征排列的组织工程支架上的力学和细胞行为可能具有重要意义。© 2016威利期刊公司。《生物医学材料研究杂志》A部分：第104A卷：1680 - 1686页，2016年。
