Skip to main content
2025 PharmSci 360 Main banner
Icon Legend
This session is not in your schedule.
This session is in your schedule. Click again to remove it.

Manufacturing & Analytical Characterization
Keynote

Keynote: Unsupervised AI-based Image Analysis of Cell Therapy Products: A New Non-Invasive Characterization Approach for QC

Wednesday, November 12, 2025
1:30 PM - 2:30 PM CT
Location:302 ABC

Manufacturing and Analytical Characterization

This talk will highlight our research in developing neural network-based image analysis techniques for characterizing CAR-T Cell Based Medicinal Products (CBMPs) and other biologics. Problems encountered when using traditional supervised neural network analyses in these applications will also be discussed. In addition, recent advances and open-problems associated with applying unsupervised machine learning (ML) methods to analyze cell therapy products will be presented.

Learning Objectives:

  • Upon completion, participants will be exposed to the basic elements of traditional neural network supervised classification.
  • Upon completion, participants will understand fundamental problems inherent in traditional neural network supervised classification.
  • Upon completion, participants will understand what we refer to as a neural network “fingerprint” and how it overcomes some inherent problems in supervised classification.
  • Upon completion, participants will be exposed to new ideas for using unsupervised machine learning methods for analyzing cell therapy products.