Page 114 - IJB-5-1

Electrohydrodynamic printing process monitoring by microscopic image identification

Table 3. The accuracy of predicted classes
Classification category Fine Broken Discharge Dry Huge Tiny Multi Meniscus
Accuracy 93.7 95.1 90.6 95.1 95.7 93.8 95.3 96.5

4000, and the training time increases significantly References
with the size of training samples. The corresponding
testing accuracy is 88.9%, 91.3%, 92.3%, and 94.7%, 1. Zhang B, He J, Li X, et al., 2016, Micro/nanoscale
respectively. The CNN models with larger size of electrohydrodynamic printing: From 2D to 3D. Nanoscale,
training data have some improvement on the testing 8(34): 15376–15388. https://doi.org/10.1039/c6nr04106j.
accuracy. With the training data size of 4000 samples, 2. Park J U, Hardy M, Kang S J, et al., 2007, High–resolution
the accuracy of the trained CNN model under each electrohydrodynamic jet printing. Nat Mater, 6(10): 782–789.
category is shown in Table 3. https://doi.org/10.1038/nmat1974.
To have better classification in EHDP real-time 3. Sun J, Vijayavenkataraman S, Liu H, 2017, An overview of
monitoring, continuous 15 frames of the microscope
images are used for classification and the cone mode scaffold design and fabrication technology for engineered
is finalized by the maximum vote. This analysis would knee meniscus. Materials, 10(1): 29. https://doi.org/10.3390/
facilitate the development of EHDP process control ma10010029.
algorithms at the next stage, which could regulate the 4. Garg K, Bowlin G L, 2011, Electrospinning jets and
applied voltage and the printing SS from run to run, for nanofibrous structures. Microfluidics, 5(1): 13403. https://
more reliable and viable manufacturing. doi.org/10.1063/1.3567097.

5. Conclusion 5. Reneker D H, Yarin A L, 2008, Electrospinning jets and
polymer nanofibers. Polymer, 49(10): 2387–2425. https://doi.
EHDP has recently drawn great attention for its capability org/10.1016/j.polymer.2008.02.002.
to print micron to nano scaffolds using biopolymer 6. Cai Y, Gevelber M, 2013, The effect of relative humidity
solutions. This great potential motivates researchers to and evaporation rate on electrospinning: Fiber diameter and
devote intensive efforts on the development of EHDP
monitoring and control system. In the paper, we collect measurement for control implications. J Mater Sci, 48(22):
digital microscopic images to monitor EHDP scaffold 7812–7826. https://doi.org/10.1007/s10853–013–7544–x.
fabrication process. The captured microscopic images 7. Yarin A L, Koombhongse S, Reneker D H, 2001, Bending
are used to evaluate the influence of the key process instability in electrospinning of nanofibers. J Appl Phys,
parameters on the cone shape and jet by features such 89(5): 3018–3026.
as centroid and jet diameter. A CNN identification 8. Wang, X., 2016, Rheology behaviors of stable
model is developed to classify EHDP cone modes into electrohydrodynamic direct–write jet. AIP Adv, 6(10):
eight categories with good classification accuracy and 105103.
low computational time. While, besides the information
derived from the microscopic images, material properties 9. Huang Y, Duan Y, Ding Y, et al., 2014, Versatile, kinetically
such as conductivity or solvent evaporation rate should controlled, high precision electrohydrodynamic writing of
also be taken into account to identify the EHDP process. micro/nanofibers. Sci Rep, 4: 5949. https://doi.org/10.1038/
With the aid of advanced control techniques, we may srep05949.
regulate the EHDP key process parameters to achieve a 10. Park J, Park J W, Nasrabadi A M, et al., 2016, Methodology
more viable manufacturing process in near future. These to set up nozzle–substrate gap for high resolution
efforts would eventually translate EHDP fabrication from electrohydrodynamic jet printing. Appl Phys Lett, 109(13):
a research stage to the commercially viable manufacturing 134104. https://doi.org/10.1063/1.4963846.
process.
11. Gonzalez R C, Woods R E, Eddins S L, 2004, Digital Image
Acknowledgment Processing using MATLAB. Vol. 624. Upper Saddle River,
This project is financially Supported by Key Program New Jersey: Pearson–Prentice–Hall.
Special Fund in Xi’an Jiaotong-Liverpool University 12. Reneker D H and Yarin, AL, 2008, Electrospinning jets and
(XJTLU) under Grant KSF-A-09. This work is also polymer nanofibers. Polymer, 49(10), 2387–2425.
partially supported by Suzhou S&T Project-Key Industrial 13. Duan Y, Ding Y, Xu Z, et al., 2017, Helix electrohydrodynamic
Technology Innovation under Grant SYG201842. printing of highly aligned serpentine micro/nanofibers.

8 International Journal of Bioprinting (2019)–Volume 5, Issue 1