Related Experiment Video
Updated: May 5, 2026
09:06
Preparation of Neuronal Co-cultures with Single Cell Precision
Published on: May 20, 2014
13.3K
CeyeHao: AI-driven microfluidic flow programming with hierarchically assembled obstacles and receptive
Zhenyu Yang1,2, Zhongning Jiang3, Haisong Lin4,5
1Advanced Biomedical Instrumentation Centre, Hong Kong Science Park, Shatin, New Territories, Hong Kong, China.
Science Advances
|July 30, 2025
Summary
CeyeHao, an AI-driven method, enables advanced microfluidic flow control for intricate microstructure fabrication. This artificial intelligence approach significantly enhances design possibilities and precision in microfluidic applications.
Area of Science:
- Microfluidics and Nanotechnology
- Artificial Intelligence in Engineering
- Materials Science and Engineering
Background:
- Microfluidic fabrication is crucial for anisotropic microstructures but limited by flow manipulation.
- Producing complex microstructures is hindered by current flow control methods.
Purpose of the Study:
- To introduce CeyeHao, an AI-driven methodology for advanced microfluidic flow programming.
- To enable unprecedented flow manipulations for intricate microstructure design.
Main Methods:
- Developed CeyeHao, featuring hierarchically assembled obstacles for enhanced flow control.
- Integrated CEyeNet, an AI model, for predicting flow profiles and reducing computation.
- Enabled both human-guided and automatic microchannel design modes.
Main Results:
- CeyeHao offers over double the flow transformation modes and configurability.
- CEyeNet achieved up to 97% accuracy and reduced computation time by 2700-fold.
- Demonstrated creation of complex geometries, artistic patterns, and precise stream topology manipulation.
Conclusions:
- CeyeHao significantly advances microfluidic design capabilities.
- The methodology facilitates complex microstructure fabrication and precise reaction control.
- AI-driven flow programming opens new avenues in microfluidic applications.
