Geometry design of herringbone structures for cancer cell capture in a microfluidic device

Cancer cell detection with high capture efficiency is important for its extensive clinical applications. Herringbone structures in microfluidic devices have been widely adopted to increase the cell capture performance due to its chaotic effect. Given the fact of laminar flow in microfluidic devices, geometry-based optimization acting as a design strategy is effective and can help researchers reduce repetitive trial experiments. In this work, we presented a computational model to track the cell motion and used normalized capture efficiency to evaluate the tumor cell capture performance under various geometry settings. Cell adhesion probability was implemented in the model to consider the nature of ligand–receptor formation and breakage during cell–surface interactions. A facile approach was introduced to determine the two lumped coefficients of cell adhesion probability through two microfluidic experiments. A comprehensive geometric study was then performed by using this model, and results were explained from the fluid dynamics. Although most of the geometric guides agree with the general criterion concluded in the literature, we found herringbone structures with symmetric arms rather than a short arm–long arm ratio of 1/3 are optimal. This difference mainly comes from the fact that our model considers the particulate nature of cells while most studies in the literature optimize the geometry merely relying on mixing effects. Thus, our computational model implemented with cell adhesion probability can serve as a more accurate and reliable approach to optimize microfluidic devices for cancer cell capture.