Hybrid predictive modeling for control of cell culture

ChangeBridge: Patent Grants - AI & Computing (G06N)

Published March 31st, 2026

Detected March 31st, 2026

Summary

The USPTO granted Amgen Inc. Patent US12590283B2 covering a method of controlling cell culture processes using hybrid predictive modeling. The patent combines data-driven models (including neural networks) with first principle models in a model predictive controller to optimize physical inputs to cell culture processes. The patent includes 22 claims and names six inventors including Mohammad Rashedi and Matthew N. Demers.

View original document View source feed page

What changed

The USPTO issued Patent US12590283B2 to Amgen Inc. on March 31, 2026, covering a hybrid predictive modeling system for cell culture process control. The invention combines machine learning components (G06N 3/0464, 3/084, 3/09, 20/10) with first principle models within a model predictive controller framework to predict future cell culture attributes and optimize control values for physical inputs. The patent application was filed on February 24, 2023, as Application No. 18113967.

This patent grant represents an informational intellectual property milestone rather than a regulatory compliance requirement. Biopharmaceutical manufacturers and biotechnology companies engaged in cell culture-based production should review the granted claims (22 total) for potential licensing implications or freedom-to-operate concerns. The patent does not impose compliance deadlines or regulatory obligations on third parties; it grants Amgen exclusive rights to the claimed methods in the United States.

Source document (simplified)

← USPTO Patent Grants

Hybrid predictive modeling for control of cell culture

Grant US12590283B2 Kind: B2 Mar 31, 2026

Assignee

AMGEN INC.

Inventors

Mohammad Rashedi, Hamid Khodabandehlou, Seyedehmina Rafieishishavan, Matthew N. Demers, Aditya Tulsyan, Tony Y. Wang

Abstract

A method of controlling a cell culture process uses hybrid predictive modeling in a model predictive controller. The method includes, for multiple time intervals, obtaining current values of cell culture attributes associated, and generating a control value for a physical input to the cell culture process. Generating the control value includes predicting future values of the cell culture attributes based on the current values, by using one or more data-driven models to predict future values of a first one or more attributes of the cell culture attributes, and using one or more first principle models to predict future values of a second one or more attributes of the cell culture attributes. Generating the control value also includes determining the control value by optimizing an objective function subject to the predicted future values. The method also includes using the control value to control the physical input to the cell culture process.