Hybrid deep modeling of a CHO-K1 fed-batch process: combining first-principles with deep neural networks

José Pinto, João R. C. Ramos, Rafael S. Costa, Sergio Rossell, Patrick Dumas, Rui Oliveira

Research output: Contribution to journalReview articlepeer-review

4 Citations (Scopus)
35 Downloads (Pure)

Abstract

Hybrid modeling combining First-Principles with machine learning is becoming a pivotal methodology for Biopharma 4.0 enactment. Chinese Hamster Ovary (CHO) cells, being the workhorse for industrial glycoproteins production, have been the object of several hybrid modeling studies. Most previous studies pursued a shallow hybrid modeling approach based on threelayered Feedforward Neural Networks (FFNNs) combined with macroscopic material balance equations. Only recently, the hybrid modeling field is incorporating deep learning into its framework with significant gains in descriptive and predictive power.
Original languageEnglish
Article number1237963
Number of pages16
JournalFrontiers in Bioengineering and Biotechnology
Volume11
DOIs
Publication statusPublished - 2023

Keywords

  • hybrid modeling
  • deep neural networks
  • first-principles
  • ADAM
  • stochastic regularization
  • CHO-K1 cells
  • biopharma 4.0

Fingerprint

Dive into the research topics of 'Hybrid deep modeling of a CHO-K1 fed-batch process: combining first-principles with deep neural networks'. Together they form a unique fingerprint.

Cite this