Related Experiment Video
Updated: Dec 19, 2025
12:32
Combining Fluidic Devices with Microscopy and Flow Cytometry to Study Microbial Transport in Porous Media Across Spatial Scales
Published on: November 25, 2020
6.9K
Particulate impurities in cell-based medicinal products traced by flow imaging microscopy combined with deep learning
A D Grabarek1, E Senel2, T Menzen2
1Coriolis Pharma, Martinsried, Germany; Leiden Academic Centre for Drug Research, Leiden University, The Netherlands.
Cytotherapy
|June 9, 2020
Summary
Flow imaging microscopy combined with convolutional neural networks (CNNs) effectively detects and quantifies impurities in cell-based medicinal products (CBMPs). This advanced method significantly improves accuracy over traditional software, ensuring product safety and quality.
Area of Science:
- Biotechnology
- Analytical Chemistry
- Cell Therapy
Background:
- Cell-based medicinal products (CBMPs) are crucial for treating severe diseases.
- Characterizing CBMPs requires robust analytical methods, which are currently limited.
- Subvisible particulate impurities pose a risk to CBMP efficacy and safety.
Purpose of the Study:
- To develop a flow imaging microscopy (FIM) method for detecting and quantifying impurities in CBMPs.
- To utilize a convolutional neural network (CNN) for image analysis and classification of impurities.
- To assess the performance of FIM-CNN for characterizing cellular and non-cellular particulates.
Main Methods:
- Flow imaging microscopy (FIM) was employed for particle detection and imaging.
- A convolutional neural network (CNN) was developed for image classification of impurities.
- Jurkat cells and Dynabeads were used as model systems for cellular and non-cellular impurities.
Main Results:
- FIM-CNN accurately detected and quantified Dynabeads, cells, cell agglomerates, and debris.
- The CNN approach reduced misclassification rates by over 50-fold compared to standard FIM software.
- A detection limit of approximately 15,000 beads/mL was achieved in the presence of 500,000 cells/mL.
Conclusions:
- CNN-assisted FIM is a powerful tool for characterizing subvisible impurities in CBMPs.
- This method enhances the analytical capabilities for ensuring the quality of cell-based therapies.
- The approach is suitable for routine quality control of CBMPs, improving patient safety.
