Machine learning enhanced droplet microfluidics

Abstract

Machine learning has recently been introduced in the context of droplet microfluidics to simplify the process of droplet formation, which is usually controlled by a variety of parameters. However, the studies introduced so far have mainly focused on droplet size control using water and mineral oil in microfluidic devices fabricated using soft lithography or rapid prototyping. This approach negated the applicability of machine learning results to other types of fluids more relevant to biomedical applications, while also preventing users that do not have access to microfluidic fabrication facilities to take advantage of previous findings. There are a number of different algorithms that could be used as part of a data driven approach, and no clear comparison has been previously offered among multiple machine learning architectures with respect to the predictions of flow rate values and generation rate. We here employed machine learning to predict the experimental parameters required for droplet generation in three commercialized microfluidic flow-focusing devices using phosphate buffer saline and biocompatible fluorinated oil as dispersed and continuous liquid phases, respectively. We compared three different machine learning architectures and established the one leading to more accurate predictions. We also compared the predictions with a new set of experiments performed at a different day to account for experimental variability. Finally, we provided a proof of concept related to algae encapsulation and designed a simple app that can be used to generate accurate predictions for a given droplet size and generation rate across the three commercial devices.

A New Approach for On-Chip Production of Biological Microgels Using Photochemical Cross-Linking

“…We observed that the droplet length L normalized by the width of the droplet generation area w was inversely proportional to the capillary number of the continued phase, defined as Ca c = μ c U c /γ, where μ c is the shear viscosity, γ is the interfacial tension, and U c is the velocity of the continuous phase calculated as U c = Q c / hw , where h is the height of the microfluidic device (Figure b). When increasing the value of Ca c , which practically meant an increase of Q c in each microfluidic device, the droplet size decreased, in agreement with previous observations. , Similarly, we observed that an increase in the flow rate ratio Q d / Q c resulted in the formation of elongated droplets (Figure c). This behavior is because larger Q d values require larger values of Q c to keep the droplet size constant .…”

“…When increasing the value of Ca c , which practically meant an increase of Q c in each microfluidic device, the droplet size decreased, in agreement with previous observations. 27 , 28 Similarly, we observed that an increase in the flow rate ratio Q d / Q c resulted in the formation of elongated droplets ( Figure 2 c). This behavior is because larger Q d values require larger values of Q c to keep the droplet size constant.…”

Verification of the Inverse Scale Effect Hypothesis on Viscosity and Diffusion by Azo-Amino Acid Schiff Base Copper Complexes

“…The droplet (formation) mechanism and also the internal reaction environment are complex and controlled by numerous conditional parameters such as surface tension, viscosity of two immiscible liquids, surface wetting properties, inertia, elasticity, and surfactant addition. However, in recent years, machine learning (ML) and artificial intelligence (AI) have begun to be used for parameter optimization as a data-driven study [20].…”