Evaluation of FTIR and multivariate data analysis for the control of medical devices

Abstract

Verification of sterilization number of medical devices is one of the main objectives to control its quality and ensure the protection of patients. The objective of this study is to investigate the capability of an automatic tool based on the combination of mid-infrared spectroscopy with a mathematical and statistical algorithm to estimate the number of sterilization cycles of medical pipes. Data analysis is performed by three main methods of pattern recognition: principal component analysis (PCA), support vector learning regression (SVR), and partial least squares regression (PLSR). The result obtained shows high performance capabilities for the prediction of sterilization cycle number. These technologies could be successfully applied in hospitals to ensure the quality of medical devices.