Raman-based estimation of glucose concentration for on-line monitoring of biological process

In biological process, it not simple to analyze concentrations of a mixture in bioreactor because pre-processing is required for analyzing bio-product. Also, on-line monitoring of biological processes is essential for maximizing productivity because reactions are very slow. The Raman spectroscopy is suitable for on-line monitoring of bioprocess with proper analysis algorithms. This work proposes a prediction technique based on Raman spectra. It first removes the background effect and further reduces noise using the Rolling-Circle Filter (RCF) and Savitzky-Golay smoothing filter, respectively. Then Partial Least Square (PLS) is used to reduce the dimension of spectra and predict the concentrations based on latent variables. The prediction performance of glucose concentration was improved by 30% by applying the proposed methods.