Microfluidic Chip For High-Throughput Detection Of Single-Cell Mechanical Properties Based On Shear-Induced Deformation

Since modern times,cancer has become one of the major threats to human life and health.The early screening and diagnosis of cancer is of great significance for improving the survival rate and alleviating the suffering.The study of various physical and biochemical properties of tumor cells also greatly promotes cancer diagnosis and treatment.Among them,the cell mechanical properties are closely related to the physiological states and functions of cells.Early studies on the cell mechanical properties were limited by technical conditions,and only the elastic or shear modulus of the cell population could be obtained,which results in the loss of the mechanical phenotypes of the heterogeneous cells.Therefore,this paper used the fluid shear force in the microfluidic channel to deform the cells,and obtained the mechanical properties and size differences between cells,thereby realizing high-throughput cell identification.In addition,there are a large number of interfering cells and impurities in body fluids such as ascites,pleural effusion,and blood of cancer patients,which makes it difficult to directly detect tumor cells.Therefore,this paper integrated the inertial and viscoelastic sorting chip with the detection chip,which greatly reduces the detection load and shortens the detection time.The main research contents of this paper are as follows:(1)Design and performance verification of multi-channel detection chip based on shearinduced deformation.In order to obtain the mechanical properties between cells,a multichannel microfluidic detection chip based on shear deformation was proposed.Firstly,the fluid-structure interaction simulation was used to investigate the force generated by the fluid in the microfluidic channel on the cells and the deformation of the cells under the stimulation of external force under the two-dimensional model.According to the simulation results,the cross-sectional size of the chip was determined to be 20 × 20 μm,and the single-channel flow rate was determined to be 15 μL/min.Secondly,a high-speed microscopic imaging system was built to capture the deformation of cells in the microfluidic channel,and a cell contour extraction algorithm based on Open CV and image background subtraction was designed to extract the contour features of cells deformed in the microfluidic channel.In addition,in order to further improve the detection throughput,this paper designed a parallel 8-channel detection chip,which increases the detection throughput several times.Moreover,the ability of the chip to distinguish different cells was verified by comparing the difference in deformability between drug-treated tumor cells and normal tumor cells,and between low-invasive and highinvasive tumor cells.Finally,in order to further distinguish cells at the single-cell level,a variety of cell deformation parameters were extracted from the cell contour,and different machine learning classification algorithms were applied to single-cell identification.The results showed that the identification accuracy of two types of cells with different mechanical properties was about 80%.(2)Design and performance verification of single-cell detection chip with integrated inertial sorting.In order to filter out a large number of interfering cells and impurities in body fluids,a microfluidic chip for detecting the mechanical properties of single cells integrated with inertial sorting was proposed.Firstly,fluorescent microspheres with different diameters were used to simulate leukocytes and tumor cells for preliminary verification of the sorting performance of the dual-port inertial spiral chip.Secondly,the asymmetric sinusoidal and serpentine flow resistance structure structure were used to realize the flow matching between the sorting and the detection chip,and it can also enrich the sorted tumor cells.Thirdly,the sorting and concentration chip was integrated with the 8-channel parallel detection chip,which realizes the high-throughput separation of tumor cells and leukocytes.In addition,an 8-channel detection chip was used for the single-cell identification of tumor cells and leukocytes,and the identification accuracy rate exceeds 95%.At the same time,cell lines with different mechanical properties in tumor cells can be further distinguished.Finally,the ability of the integrated chip to discriminate between tumor cells and leukocytes was verified by the comparison of immunofluorescence counting and detection counting.(3)Design and performance verification of single-cell detection chip with integrated viscoelastic sorting.In order to expand the application field of the integrated sorting and detection chip,a microfluidic chip based on viscoelastic fluid was designed.Firstly,a viscoelastic sorting chip with a symmetric shrinkage-expandable structure was designed and manufactured based on 0.1% HA solution,and its sorting performance and applicable flow rate range were explored by using fluorescent microspheres with different diameters.Secondly,according to the selected sorting flow rate,a 6-channel parallel detection chip was designed and integrated with the viscoelastic sorting chip,and the feasibility of the integrated chip for tumor cell and leukocyte sorting was verified.Furthermore,the identification accuracy between white blood cells and tumor cells was more than 94% by using the 6-channel detection chip,and the chip can further distinguish tumor cell lines with different mechanical properties in three classifications.Finally,by comparing the immunofluorescence counting and the detection counting,the single-cell discrimination ability of the integrated chip based on viscoelastic fluid for tumor cells and leukocytes was verified.