Spectroscopic models for real-time monitoring of cell culture processes using spatiotemporal just-in-time Gaussian processes

Spectroscopic methods play an instrumental role in the implementation of the U.S. Food and Drug Administration outlined process analytical technology for biopharmaceutical manufacturing. Industrial spectroscopic calibration models are typically developed in an offline setting using traditional methods, such as partial least squares and principal component regression. Apart from the limiting performances of these conventional models under time-varying operating conditions, these methods require access to extensive historical data, which are seldom available in biopharmaceutical manufacturing. In this article, we propose a novel spatiotemporal just-in-time learning (ST-JITL) based spectroscopic model calibration platform for automatic training and maintenance of calibration models using routine batch data. The proposed ST-JITL framework uses Gaussian processes (GPs) for local model calibration. A GP model not only exhibits superior performance over traditional methods but also provides credibility intervals around the model predictions. The efficacy of the ST-JITL based model calibration platform is demonstrated in predicting the critical performance parameters of an industrial cell culture process.