Prediction of diameter in blended nanofibers of polycaprolactone-gelatin using ANN and RSM

Fabrication of nanofibers with a defined diameter is a primary purpose of the electrospinning process. The diameter of nanofiber is directly related to its individual features, such as mechanical property and porosity. The motivation to conduct the current study was to explore the diameter of hybrid nanofibers of polycaprolactone-gelatin (PCL-GT) as one of the most attractive scaffolds employed in various research fields, such as tissue engineering and industrial fields. We have developed two predictive models describing the electrospinning process of PCL-GT using response surface methodology (RSM) and artificial neural network (ANN). The effect of 4 variables on diameter was analyzed, including total polymer concentration, ratio of PCL to Gel, voltage, and tip-to-collector distance. The individual and interactive effects of the mentioned factors were analyzed using RSM. The total polymer concentration had the most significant individual effect on the diameter of PCL-Gel nanofiber, whereas the other three factors showed less strong individual effects, although, the interactive effects of these factors were more remarkable. It was demonstrated that both models, especially the ANN model, could accurately predict the diameter of PCL-GT nanofiber (regression coefficient > 0.92, mean absolute percentage error < 5.7). The represented predictive models could facilitate construction of electrospun nanofibers from PCL-Gel with wellcontrolled diameter required for any intended purpose.