Research And Application Of Imaging Flow Cytometer Based On Microfluidic Chip

In recent years,microfluidic chips have been increasingly used in the research of cells and micro-scale organisms.Flow cytometry is one of the essential devices for research in this field.Traditional flow cytometers are large and complicated to operate,while microfluidic chip-based flow cytometers have the advantages of less reagent consumption,automated operation,and smaller footprint.With the development of machine vision technology in recent years and the increasing demand of researchers for the accuracy of flow cytometry detection,imaging flow cytometers based on image processing have attracted much attention.This paper takes Caenorhabditis elegans as the main research object to build an imaging flow cytometer system based on microfluidic chip.Compared with particles such as cells,nematodes are transparent and have a high degree of coincidence with the background,making image recognition difficult.Aiming at this problem,this paper builds an automated and highly integrated liquid flow control system and machine vision system,and develops special software to monitor it in real time.The machine learning algorithm is used to continuously optimize the real-time background image and achieve better foreground and background separation effects than traditional background subtraction algorithms.A new nematode recognition algorithm is proposed.Using mathematical methods such as numerical analysis,it can quickly extract.It has been classified and achieved excellent recognition effect.The single frame processing time of this algorithm from nematode classification is 15 ms.Twenty L1 and L2 stage nematodes are selected for injection,and the classification accuracy of the algorithm was 100%.A nematode sorting system is set up at the back of the microfluidic chip.Through further experiments,high-speed separation of nematodes in different periods can be achieved.In this paper,the complex target recognition algorithm in high-speed flow is extended to the microemulsion system,which can realize accurate recognition and real-time counting.