Flow Cytometric Detection Of Leukemic Cells With Light-Sheet Illumination Two-dimensional Light Scattering And High Magnification Bright-field Imaging

Leukemia is a common tumor disease,which can be divided into acute lymphoid leukemia(ALL),acute myeloid leukemia(AML),chronic lymphoid leukemia(CLL)and chronic myeloid leukemia(CML)according to the natural course of disease and cell origin.Research on the diagnosis and treatment of leukemia subtype has always been the focus of attention.Flow cytometry can be used to classify and diagnose leukemia.Traditional flow cytometry mainly relies on fluorescence spectroscopy,which is complicated and has certain influence on cells.Fluorescence imaging can be achieved by flow cytometry,but is usually limited to low optical magnification.It is of great significance to explore the techniques of unlabeled analysis and high magnification imaging flow cytometry for leukemia detection.In this paper,the flow detection is realized based on variable slice lighting technology.The variable light-sheet illumination system studied in this paper can provide lamellar laser with controllable thickness in the range of 3～50μm.By combining microfluidic technology and two-dimensional light scattering technology,the two-dimensional light scattering flow detection system of variable light-sheet illumination is built in this paper to realize two-dimensional light scattering imaging of samples in the flow state.Secondly,high magnification bright-field imaging is studied in this paper.By improving the general imaging optical path and introducing the method of secondary magnification,the high magnification imaging of more than 100 times optical magnification can be achieved in the long working distance of the experimental samples,which can be applied in the flow detection experiment.A high magnification optical flow cytometer was constructed by combining this technique with bright-field imaging and microfluidic technology.This kind of flow cytometer can collect high magnification bright-field original image of cells and observe the cell morphology.Finally,the combination of flow cytometry and machine learning was studied in this paper.2D light scattering images and high magnification bright-field images of normal myeloid leukocytes and chronic myeloid leukemia cell lines were obtained by flow cytometry.In this paper,the automatic learning classification of these two kinds of cells based on different images was studied by machine learning method.The accuracy of light scattering image and bright-field image classification reached 91.7%and 86.1%respectively.In summary,in this paper,the flow cytometry detection of leukemia cells by 2d light scattering and high magnification bright-field imaging was successfully realized,and automatic learning classification was realized by combining machine learning.The method presented in this paper is a feasible improvement on the existing flow cytometry and has a promising application in the detection of leukemia cells.