Font Size: a A A

Research On The Portable Microfluidic Cell Imaging Detection System

Posted on:2022-12-26Degree:MasterType:Thesis
Country:ChinaCandidate:R J WangFull Text:PDF
GTID:2480306605496844Subject:Electronics and Communications Engineering
Abstract/Summary:PDF Full Text Request
Cell detection is widely used in life science research and medical diagnosis.The existing imaging flow cytometer usually needs special optical microscope system,high-speed camera,and precision injection pump.Such equipment is bulky,expensive,and needs personnel with certain professional knowledge to operate.Although this method is reliable,it greatly limits the application of cell imaging in many places without special equipment,and it is difficult to support the use of most clinics in the field.In recent years,some portable cell detection systems have been proposed,but most of these systems only miniaturize the imaging part,and still need to rely on external personal computers in data storage,processing,and display.In order to meet this challenge,an independent portable microfluidic cell imaging detection system is proposed in this thesis.The system has an embedded computing function and can be used for point of care testing(POCT).The completely independent system integrates a portable microfluidic image acquisition module,light source,embedded processing and control module,micropump,touch screen,and power supply module to realize the functions of biological sample detection,processing,and result display.The weight of the whole system is 1.9kg and the volume size is 20×10×15cm~3.Through experimental tests,the system can clearly detect microspheres with different diameters.Under the flow rate of 1?L/min,the classification and counting accuracy of the system can reach over 99.7%for 10?m and 15?m particles in either single or mixed microbead sample solution,and reach95.96%for Hep G2(Human hepatocellular carcinomas)cells in mixed cell and microbead sample solution.For the function of cell recognition and classification,this thesis first completes the model training on the PC,and then builds a deep learning network on a Raspberry Pi to realize the real-time classification of cells.Therefore,the system can collect samples and analyze data around patients,and realize the function of real-time reporting of cell test results.The research of the portable cell analysis equipment will facilitate the early analysis of diseases and POC diagnosis.
Keywords/Search Tags:Microfluidic, cell imaging, deep learning, cell counting and classification, POCT
PDF Full Text Request
Related items