Prediction of the isotherms of human IgG adsorption on Ni(II)-IDA-PEVA membrane using artificial neural networks

Abstract

The use of artificial neural networks (ANNs) to predict the adsorption isotherms of human immunoglobulin G on immobilized Ni(II) affinity hollow fiber membranes was studied. Neural networks were trained using the Levenberg-Marquardt algorithm combined with Bayesian regularization technique and experimental data from different temperatures. The resulting neural network demonstrated to be able to interpolate the behavior of the maximum adsorption capacity and equilibrium concentration in the temperature range (4, 37°C) with correlation coefficients higher than 0.96. Results demonstrated to be very similar to those achieved with traditionally Langmuir model adjustment. The advantage of interpolation ability of ANNs was also showed.

Establishment of Fuzzy Langmuir Adsorption Model and Prediction of Chromatographic Behavior

“…The ANN model is constructed by mapping the relationship between input and output data without having any prior knowledge of their mechanism. It is widely used in modeling, [22,23] prediction, [24][25][26] and other processes.…”