Need: The most common approach to characterizing electrospun fibers for tissue engineering applications has been to image them by scanning electron microscopy (SEM) and then use image analysis software line tools to measure fiber diameters manually. This approach is slow and may be biased by having the user select which fibers get measured.
Goal: Develop a validated, open-source tool for automated analysis of scanning electron micrographs for determining fiber diameter.
An algorithm for fiber diameter analysis called DiameterJ was written as a plugin for ImageJ software package.
Running on a typical desktop computer, DiameterJ can analyze a scanning electron micrograph of fibers in about 30 s, yielding hundreds to thousands of diameter measurements from each image.
Validation of fiber diameter measurements
DiameterJ was validated using digital synthetic images of white lines on a black background. DiameterJ had an error of approximately 1% (< 1 pixel) on synthetic images.
DiameterJ was also validated on scanning electron micrographs of highly uniform steel wires of known diameters. DiameterJ had an error of 1% to 2% on wire images.
A detailed training module for learning DiameterJ has been deployed on the web. Testing demonstrated that user accuracy and precision were improved upon completion of the training.
Hotaling NA, Bharti K, Kriel H, Simon Jr CG (2015) Dataset for the validation and use of DiameterJ, an open source nanofiber diameter measurement tool. Data in Brief 5, 13–22. https://doi.org/10.1016/j.dib.2015.07.012 (this article provides the set of test images that others could use to develop better algorithms)
Hotaling NA, Jeon J, Wade MB, Luong D, Palmer X-L, Bharti K, Simon Jr CG (2016) Training to improve precision and accuracy in the measurement of fiber morphology. PLOS One 11, e0167664. https://doi.org/10.1371/journal.pone.0167664