YOLOv8-Based Visual Localization and Fixed-Point Immobilization of Cells Driven by Surface Acoustic Waves

Read full text Summarise with VIRAT
Authors: Jiguang Han, Jun Wei
Cited by: 0
References: 7

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Automated cell manipulation is a key technology in biomedical research. Realizing real-time detection and precise positioning of targets is an essential prerequisite for applications such as single-cell analysis and drug screening. Most existing methods rely on complex vision systems or multi-degree-of-freedom manipulators, which suffer from issues including low integration and limited automation. This paper proposes an automated system that combines deep learning-based object detection with Surface Acoustic Wave (SAW) technology driven by Interdigital Transducers (IDTs) for the real-time recognition and fixed-point immobilization control of zebrafish embryos. The system obtains the positional coordinates of embryos in the microchannel in real time using the YOLOv8n model. When an embryo enters the target area, SAW is automatically triggered to drive the embryo to the designated immobilization point and maintain its stability. Experimental results demonstrate that the mean average precision (mAP@0.5) of the YOLOv8n model for embryo detection reaches 0.994, with a single-frame inference time of 78 ms. In 15 repeated experiments, the system achieves a success rate of 93.3% in driving embryos to the target position. This method provides a compact, non-contact, and high-precision integrated solution for detection and control in automated cell manipulation.

Fields of Study & Subjects

Microfluidic and Bio-sensing Technologies Cell Image Analysis Techniques Zebrafish Biomedical Research Applications

Publication details

Journal / source
Frontiers in Computing and Intelligent Systems
Volume
15
Issue
3
Pages
100–103
Publication date
28 March 2026
Language
EN