Optimal microchannel design using genetic algorithms

This paper presents a novel method of optimizing particle-suspended microfluidic channels using genetic algorithms (GAs). The GAs can be used to generate an optimal microchannel design by varying its geometrical parameters. A heuristic simulation can be useful for simulating the emergent behaviors of particles resulting from their interaction with a virtual microchannel environment. At the same time, fitness evaluation enables us to direct evolutions towards an optimized microchannel design. Specifically, this technique can be used to demonstrate its feasibility by optimizing one commercialized product for clinical applications such as the microchannel-type imaging flow cytometry of human erythrocytes. The resulting channel design can also be fabricated and then compared to its counterpart. This result implies that this approach can be potentially beneficial for developing a complex microchannel design in a controlled manner.