Automatic label-free image-based system for cell viability monitoring on-a-chip

Monitoring and assessing cell viability is essential in biochemical systems analysis and medical diagnostics. Currently, several methodologies are available, broadly classified into label-based and label-free viability assays. This work presents a novel, label-free cell viability characterization technique that leverages on-a-chip hydrodynamical stimulations. As a proof-of-concept, the collective behavior of healthy and unhealthy eukaryotic yeast cells, suspended in a PBS (phosphate-buffered saline) solution and subjected to an oscillatory input flow within a microchannel, was studied and compared to that of inert silica beads. To enable long-term real-time measurements, a Digital Particle Image Velocimetry (DPIV)-based algorithm was extended in this work, incorporating a Time-Slot implementation for continuous monitoring. This is crucial for drug investigations and in-vitro diagnostics, enabling deeper insights into dynamic biological processes and therapeutic responses. This method analyzes the hydrodynamic response of micro-particles, automatically extracting parameters in both the time and frequency domains within predefined time windows, named Time-Slot, overcoming the limitations of traditional offline analysis. These parameters were successfully used in this study to accurately classify the different nature of micro-particles. This work demonstrates, for the first time, the feasibility of using label-free, low-frequency, oscillating stimulations to assess cell viability on-a-chip. The parameters extracted from the hydrodynamic cell response provide an automated, non-invasive solution for classifying cell viability, while avoiding data overload and allowing for extended monitoring times. The developed system, simple to implement and versatile, offers a powerful tool for long-term, on-a-chip viability monitoring and introduces a new paradigm for label-free cell viability assessment in microfluidic devices.