Cell Viability Prediction Using Bioreactor Process Parameters
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current biomolecule manufacturing processes face challenges in efficiently determining cell viability during production, as existing methods are time-consuming, resource-intensive, and prone to product contamination, particularly due to the need for in-process sampling.
Innovation Solution
A computer-implemented method using a machine learning model, trained with manufacturing process parameters such as time elapsed, total base added, and bioreactor volume, to predict cell viability in real-time, allowing for continuous monitoring and adjustment without sampling, utilizing sensors and controllers to input data like pH, oxygen levels, and temperature.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If in-process sampling is performed to evaluate cell viability, then measurement accuracy is improved, but contamination risk and time consumption increase
Solution Approach 1:
The patent introduces an intermediary machine learning model that acts as a mediator between process parameters and cell viability assessment. Instead of directly sampling the cell culture, the model uses intermediate process parameters (pH, temperature, dissolved oxygen, agitation speed, sparge rate) as proxies to predict viability, thereby avoiding contamination while maintaining measurement accuracy.
Solution Approach 2:
The patent replaces the mechanical sampling process with a computational prediction system. The machine learning model substitutes the physical act of sampling and laboratory analysis with an in-silico prediction based on process data, eliminating the need for physical intervention that causes contamination and time delays.
2Measurement precision
If in-process sampling is performed to evaluate cell viability, then measurement accuracy is improved, but time consumption increases
Solution Approach 1:
The patent enables continuous monitoring of cell viability by continuously collecting process parameters and feeding them to the machine learning model. This replaces the discontinuous, batch-based sampling approach with a continuous prediction system that provides real-time viability assessment without interrupting the manufacturing process.
Solution Approach 2:
The machine learning model is trained in advance using historical process data and corresponding viability measurements. This preliminary training phase creates a ready-to-use prediction system that can immediately assess viability without requiring time-consuming laboratory analysis when actual measurements are needed.
3Reliability
If traditional sampling methods are used to monitor cell viability, then product quality control is improved, but resource consumption increases
Solution Approach 1:
The system uses the existing process control infrastructure and already-collected process parameters to perform viability assessment. Instead of requiring separate sampling equipment, laboratories, and personnel, the system leverages the bioreactor's own sensors and control data, making the monitoring process self-service and reducing external resource requirements.
Solution Approach 2:
The machine learning model serves multiple functions: it predicts cell viability, identifies optimal harvest times, and can potentially predict other quality attributes. This multi-functionality replaces multiple separate testing procedures with a single integrated prediction system, reducing overall resource consumption while maintaining product quality control.
Data Source
AI summary
A method, system, and non-transitory computer readable medium for predicting cell viability of a cell culture in a bioreactor during a biomolecule manufacturing process are disclosed. In various embodiments, at least three manufacturing process parameters related to the process for manufacturing molecules are input into a machine learning model that is trained to predict cell viabilities. The trained machine learning model may then analyze the at least three manufacturing process parameters to generate an indicator of cell viability of the cell culture.