Search
Close this search box.

Machine Learning Approaches to Characterization and Root-Cause Analysis of Particulates Formed During Protein Formulation Fill-Finish Operations

In this project, machine learning approaches using convolutional neural networks will be used to extract useful information from collections of images of particles measured after biopharmaceutical fill-finish operations.
Categories
Proteins/ Antibodies
Process control
Data
Project status
100% Completed

Solution

Performance Period: 10/1/2020 to 12/31/2021

This project is in the NIIMBL Project Call 3.1 “Big Data Analytics & Automation” topic area. In response to regulatory requirements, a variety of imaging techniques are routinely used to measure particle levels and particle size distributions within therapeutic protein formulations in both drug substance and drug product. As a result, vast amounts of data are recorded. Currently, much of the information contained within these “big data” sets is not utilized.

In this project, machine learning approaches using convolutional neural networks will be used to extract useful information from collections of images of particles measured after biopharmaceutical fill-finish operations. Algorithms will be developed and tested to a) facilitate root-cause analysis for protein aggregation, b) enable process monitoring for fill-finish process-induced aggregation; and c) allow more rapid automated discrimination between innocuous particles such as air bubbles and particles of regulatory concern in automated visual inspection for fill-finish operations.

Impacts

Machine learning approaches using convolutional neural networks will be used to extract useful information from collections of images of particles measured after biopharmaceutical fill-finish operations

Updates, Related Publications, and Deliverables

  • Login to the NIIMBL member portal for access to project updates, related publications, and deliverables. 

Not yet a member? Learn more about which level of NIIMBL membership is right for you and your organization.

Project Lead

Regents of the University of Colorado (Boulder)

Regents of the University of Colorado (Boulder)

Participating Organizations

Merck Sharp & Dohme LLC

Merck Sharp & Dohme LLC

SentrySciences LLC

SentrySciences LLC