Application of intelligent neural network method for prediction of mechanical behavior of wire-rope scaffold in tissue engineering

Abstract

In tissue engineering, proper mechanical property is one of the key subjects for modeling tendon and ligament scaffolds. In this research, attempts were made to determine the factors of structural parameters of wire-rope silk scaffolds by using the intelligent neural network method to find optimum mechanical behavior for tendon and ligament regeneration. Experimental design of wire-rope scaffolds was made according to Taguchi orthogonal method. The input structural parameters were the number of filament and twist in each layer in four levels wire-rope and output was determined by an index that was defined according to mechanical properties of anterior cruciate ligament for young age group. Finally, a back-propagation neural network with a high accuracy was designed to predict the mechanical properties of wire-rope tendon and ligament scaffold and according to sensitivity analysis, the number of filaments and twist in outer layers is less important than other input parameters. Finally, a multiple linear regression, a most widely used statistical method, developed from the data.

Keywords:

• mechanical behavior
• wire-rope scaffold
• tissue engineering
• modeling
• intelligent neural network
• tendon and ligament