Automated Methods to Determine Electrospun Fiber Alignment and Diameter Using the Radon Transform

Recent evidence demonstrates the ability to change cell function by altering the physical properties of electrospun scaffolds, but many studies still do not characterize electrospun fiber alignment and diameter. To aid in the reporting of these crucial properties, we demonstrate two methods of quantifying electrospun fiber alignment with one method capable of determining electrospun fiber diameter. The first method assesses fiber alignment in a scanning electron microscopy image using the Radon Transform (RT) to calculate the entropy of the fibers in the image. The RT entropy method was more sensitive than a commonly used Fast Fourier Transform (FFT) method because the RT method was able to assess smaller changes in alignment (+/- 2(circle)) than the FFT (+/- 4(circle), p<0.05). The second method used the RT to detect both fiber diameter and fiber alignment by recognition of fiber edges. The RT edge method was more capable of identifying electrospun fiber alignment and diameter than a manual method using ImageJ because the ImageJ results were statistically different from information contained in images with defined alignment and diameter (p<0.05) while the RT method showed no differences. However, the RT edge method of assessing fiber diameter was limited by the magnification of the image and was only capable of detecting fibers larger than four pixels in diameter. The RT edge method was more sensitive in differentiating between fiber scaffolds of different alignment than the entropy method since the RT edge method differentiated between all fiber alignment groups (p<0.05) while the RT entropy method was less capable at high degrees of misalignment (>+/- 8(circle)). The RT is introduced as a sensitive tool for assessing electrospun fiber alignment, but more importantly, we have demonstrated for the first time an automated method of determining electrospun fiber diameter.