Revolutionizing Epithelial Differentiability Analysis in Small Airway-on-a-Chip Models Using Label-Free Imaging and...

Area of Science:

• Biomedical Engineering
• Cell Biology
• Microfluidics

Background:

• Organ-on-a-chip (OOC) systems are vital for modeling human physiology and disease.
• Efficient analysis of OOC data is critical for drug development and toxicology.
• Current methods for assessing cell differentiation in OOC models can be time-consuming and lack automation.

Purpose of the Study:

• To develop a label-free imaging platform for analyzing human small airway epithelial cells (HSAECs) in small airway-on-a-chip systems.
• To utilize deep learning for predicting HSAEC differentiability.
• To enhance the efficiency and stability of establishing small airway-on-a-chip models.

Main Methods:

• A label-free morphology imaging platform was integrated with a small airway-on-a-chip system.
• Deep learning and image recognition techniques were employed to analyze HSAEC morphology.
• A customized MATLAB program was developed for automated analysis of ciliated cell beating.
• Automated fluorescent particle tracking was used to assess mucociliary clearance.

Main Results:

• The deep learning approach accurately predicted HSAEC differentiability after 4 weeks of incubation, using imaging data from day 3.
• Automated analysis of ciliary beating frequency and mucociliary clearance integrity was achieved.
• The platform demonstrated enhanced efficiency and stability in establishing small airway-on-a-chip models.

Conclusions:

• The integration of deep learning, label-free imaging, and advanced image analysis provides a powerful tool for OOC research.
• This approach offers significant advancements for drug testing, environmental toxicology, and physiological assessment of the respiratory system.
• The developed platform enables unprecedented insights into small airway epithelium function and facilitates targeted intervention development.