ABSTRACT

lyrical procedures, & control strategies to ensure that patients receive safe and reliable goods. Raman spectroscopy is a versatile assessment technique used in biopharmaceutical production to anticipate important parameters in the culture of cell processes in actual time. Raman spectroscopy depends on chemometric concepts that must be properly calibrated to ensure accuracy. The current calibration approach is difficult to apply since it requires running numerous properly crafted experiments to generate appropriate calibration groups. Furthermore, the current technique results in the calibration technique are only reliable in the operating conditions in which they were calibrated. In clinical production, when items have a short overview of production, this poses a distinct issue. To calibrate Raman models, a new machine learning approach based upon Quick Learning Technique is proposed in this study. Unlike older methods, Quick Learning Technique-based generic Raman models could be employed with confidence for a variety of modalities, cell lines, culture medium, & operating circumstances. Several validation tests including actual time predictions of crucial cell culture performance metrics illustrate the reliability of Quick Learning Technique-based generic systems. The suggested Quick Learning Technique methodology represents a paradigm leap in the calibration of industrial Raman models. Keywords: Quick Learning Technique; Chemometric Models; Raman models
Publication date: 01/11/2021

https://ijbpas.com/pdf/2021/November/MS_IJBPAS_2021_NOV_SPCL1044.pdf

Download PDF
https://doi.org/10.31032/IJBPAS/2021/10.11.1044