Computer Vision-Based Artificial Intelligence-Mediated Encoding-Decoding for Multiplexed Microfluidic Digital Immunoassay

作者：Weiqi Zhao, Yang Zhou, Y. X. Feng, Xiaohu Niu, Yongkun Zhao, Junpeng Zhao, Yongzhen Dong, Mingqian Tan, Yunlei Xianyu, Yiping Chen · 发表于：ACS Nano · 年份：2023 · DOI：10.1021/acsnano.3c02941 · 被引用次数：84 · 研究领域：Microfluidic and Bio-sensing Technologies、Biosensors and Analytical Detection、Electrowetting and Microfluidic Technologies

Digital immunoassays with multiplexed capacity, ultrahigh sensitivity, and broad affordability are urgently required in clinical diagnosis, food safety, and environmental monitoring. In this work, a multidimensional digital immunoassay has been developed through microparticle-based encoding and artificial intelligence-based decoding, enabling multiplexed detection with high sensitivity and convenient operation. The information encoded in the features of microspheres, including their size, number, and color, allows for the simultaneous identification and accurate quantification of multiple targets. Computer vision-based artificial intelligence can analyze the microscopy images for information decoding and output identification results visually. Moreover, the optical microscopy imaging can be well integrated with the microfluidic platform, allowing for encoding-decoding through the computer vision-based artificial intelligence. This microfluidic digital immunoassay can simultaneously analyze multiple inflammatory markers and antibiotics within 30 min with high sensitivity and a broad detection range from pg/mL to μg/mL, which holds great promise as an intelligent bioassay for next-generation multiplexed biosensing.