3D printed microfluidics for cell analysis using digital holographic microscopy

Microfluidic devices have significantly advanced cell analysis by enabling precise control over sample handling and imaging. Despite their potential, conventional fabrication methods like soft lithography are time-consuming and require specialized facilities, limiting scalability and accessibility. To overcome these challenges, we developed a cost-effective, silicone extrusion-based 3D printing system, featuring a streamlined design scheme and precise control system to enable rapid prototyping of microfluidic devices with sub-200 µm resolution. Using this approach, we fabricated a microfluidic device with an inertial focusing structure to streamline cells for consistent imaging using digital holographic microscopy, a label-free technique that captures the optical properties of cells. The performance of the 3D printed device was evaluated through yeast viability analysis using live and dead yeast samples. The platform demonstrated effective cell focusing at a flow rate of 0.5 mL/min, yielding high-quality holographic images suitable for cell classification using deep learning methods. A deep learning framework was applied to distinguish live from dead yeast cells, achieving a true positive rate of 96.5% for live cells and 86% for dead cells. By integrating innovative 3D printing with advanced imaging techniques, this approach addresses key challenges in microfluidic fabrication, providing a scalable solution for applications in clinical diagnostics, environmental monitoring, and industrial processes. Future work will focus on further refining printing resolution and expanding training datasets to enhance the accuracy of deep learning models across diverse cell types.