Related Experiment Video
Updated: Jun 14, 2026
08:58
Double Emulsion Generation Using a Polydimethylsiloxane PDMS Co-axial Flow Focus Device
Published on: December 25, 2015
16.1K
Design automation of microfluidic single and double emulsion droplets with machine learning
Ali Lashkaripour1,2, David P McIntyre3,4, Suzanne G K Calhoun5
1Department of Bioengineering, Stanford University, Stanford, CA, USA. alilp@stanford.edu.
Nature Communications
|January 3, 2024
Summary
Machine learning models predict droplet microfluidics parameters, enabling stable droplet generation for life sciences applications. This accelerates the development of custom droplet-based platforms with high throughput and low cost.
Area of Science:
- Microfluidics
- Biotechnology
- Machine Learning
Background:
- Droplet microfluidics offers high-throughput screening of picoliter samples cost-effectively.
- Current methods require extensive empirical optimization, limiting accessibility to specialist laboratories.
- Standardization of droplet generation for diverse fluids and sizes remains a challenge.
Purpose of the Study:
- To develop predictive machine learning models for droplet microfluidics.
- To enable accurate prediction of device geometries and flow conditions for stable droplet generation.
- To create an accessible design automation tool for tailored droplet-based platforms.
Main Methods:
- Compiled a comprehensive dataset of droplet generation parameters.
- Trained machine learning models to predict optimal device designs and flow conditions.
- Validated model generalizability through blind predictions across various fluids and materials.
Main Results:
- Accurate prediction of device geometries and flow conditions for stable single and double emulsions.
- Successful generation of droplets from 15 to 250 μm at rates up to 12,000 Hz.
- Developed a user-friendly design automation tool achieving droplet diameters within 3 μm (<8%) of target.
Conclusions:
- Machine learning significantly simplifies and accelerates the optimization of droplet microfluidics.
- The developed tool enhances the accessibility and adoption of droplet microfluidics in life sciences.
- This approach facilitates the rapid development of customized droplet-based platforms for diverse applications.
