Deep Learning Assisted Automated Separation Platform of Single Cells and Microparticles Using Optoelectronic Tweezers

Abstract

A deep learning-assisted automated separation platform of single cells and microparticles using optoelectronic tweezers was proposed in this paper, which allows accurate manipulation and long-term dynamic observation of single cells without complex microfluidic structures. A single-cell detector based on YOLOv5 was developed to realize single-cell automation and high throughput recognition in optoelectronic tweezers chips. Then these recognized cells or particles were captured and operated by the light patterns generated by the optoelectronic tweezers system. We built a single-cell separation platform and realized the automatic queuing of disordered microspheres or assigning them to each nearest microchannel in a short time. This work can potentially facilitate the study of cell heterogeneity and biologics drug discovery.