Modeling tensile strength and suture retention of polycaprolactone electrospun nanofibrous scaffolds by artificial neural networks

Electrospun polycaprolactone (PCL) scaffolds are broadly used in tissue engineering applications due to their superior biomechanical properties and compatibility with the cell matrix. The properties of PCL scaffolds depend on electrospinning parameters. The relationships between electrospinning process parameters and scaffold properties are complicated and nonlinear. In this study, we used the artificial neural networks (ANN) technique to estimate the tensile strength and suture retention of PCL scaffolds as a function of electrospinning parameters (polymer concentration, solution feed rate, applied voltage, and nozzle to collector distance). A standalone ANN software was developed, and the predicted properties were a good agreement with the experimental data. The present model has excellent learning precision for both training and testing data sets. The precise predictions revealed that the model could estimate the relationships between electrospinning parameters and properties of PCL scaffolds adequately.

Automated summary

Electrospun polycaprolactone (PCL) scaffolds are widely used in tissue engineering applications for their superior biomechanical properties and cell matrix compatibility, with properties depending on electrospinning parameters. This study successfully estimated the properties of PCL scaffolds using artificial neural networks (ANN) technique with high learning precision. The precise predictions demonstrated the model's ability to adequately estimate the relationships between electrospinning parameters and properties of PCL scaffolds.