Development of Machine Learning Regression Models for Predicting the Performance of Nanofibrous Scaffolds for Skin Tissue Engineering

Abstract

Tissue engineering is a branch of regenerative medicine that harnesses biomaterials and stem cells to utilize the body’s natural healing responses to regenerate tissue and organs. Skin components can be rebuilt by safeguarding their structure and function with the help of advanced scaffold manufacturing techniques. It is important to combine medical concerns with the vast explosion of artificial intelligence concepts to preserve human life and improve health. Currently, machine learning can make reliable contributions to critical decision-making in a wide range of applications. Regression machine learning models rely on correlations, associations, and other relationships between a dependent variable and a group of features. The main objective of this research was to study the effects of applying machine learning techniques on the performance of nanoscaffolds. A regression tree, a random forest, AdaBoost, and a gradient boosting algorithm were applied to the dataset and clustering data. By comparing our proposed models with the relevant studies to verify each machine learning model’s optimal performance, the AdaBoost technique was shown to have the highest accuracy (98.58%, 99.6%, 98.51%, and 98.85%), with a mean absolute percentage error of 1.41% and an R 2 value of 0.999, which indicates a strong correlation between the predicted and actual values for the whole dataset and all subgroups.