A deformability-based biochip for precise label-free stratification of metastatic subtypes using deep learning

This paper presents the development of a novel parallel constriction-based microfluidic flow cytometry device coupled with an integrated computational framework, termed ATMQcD, aimed at measuring cellular deformability as a biomarker for assessing cell physiological states. The ATMQcD framework streamlines the analysis process by incorporating automatic training set generation, multiple object tracking, segmentation, and quantification of cellular deformability. Validation with various cancer cell lines demonstrated a high classification accuracy of 92.4% for assessing invasiveness and effectively stratifying cells pre- and post-hypoxia treatment. The system also distinguished cancer cells from leukocytes with an accuracy of 89.5%. A power-law rheology model was developed to quantify stiffness, allowing for the evaluation of metastatic potential across different cancer types and mixed cell populations under realistic clinical conditions. This robust computational framework enhances the scalability of microfluidic assays, positioning it as a promising tool for high-throughput cellular analysis in clinical applications.

For more info, DOI: 10.1038/s41378-023-00577-1

a Schematic diagram of the microfluidic device. The device had one inlet, one outlet, and four groups of microconstrictions. Samples were introduced through the device inlet and deformed in the narrow microconstrictions. b Top view of laminar flow velocity simulation of the microfluidic chip under a 50 μL/min flow rate. c The fluid flow velocity across the 36 microconstrictions under a 50 μL/min flow rate. The blue lines in the left subplot demonstrate the measuring lines in the COMSOL simulation. A scaled subplot showing a measuring line crossing a microconstriction (upper right). A line plot (bottom-right) displays velocity across the 36 microconstrictions denoted in the left subplot. d Schematic view of the cDC platform’s experimental setup and the computational framework for automatic training set generation, multiple object tracking, segmentation, and cellular deformability quantification (ATMQcD). e Time-lapse imaging demonstrated the cell deformation and movement process while passing through a microconstriction