Single-cell spontaneous-stimulated raman spectroscopy detection axial automatic positioning method and system

Abstract

The present application relates to the technical field of cell detection, and particularly relates to a single-cell spontaneous-stimulated Raman spectrum detection axial automatic positioning method and system. The method comprises the following steps: building a positioning system; after positioning the cell through white light rough focusing, acquiring three-dimensional spatial tomographic imaging stack of the cell by using stimulated Raman spectrum imaging; determining the axial displacement distance of the precise displacement platform by using an equal interval method or a PID control algorithm, and adjusting the axial displacement distance H of the precise displacement platform; setting the position of the precise displacement platform as H+L after correction; switching the electric diaphragm, setting spontaneous Raman spectrum acquisition parameters, collecting the spontaneous Raman spectrum of the cell, and completing fine Raman spectrum acquisition of the cell; collecting the Raman spectrum of the known cell and establishing a cell type discrimination model; and performing discrimination analysis on the unknown type cell sample to realize accurate positioning. The method has the advantages of label-free and non-staining, and can obtain the spatial distribution of internal lipid molecules of the cell and realize accurate sampling of the cell Raman spectrum.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of cell detection, and in particular to a single-cell spontaneous-stimulated Raman spectrum detection axial automatic positioning method and system. BACKGROUND

[0002] Cells are the basic unit of life, mainly composed of proteins, lipids, nucleic acids and other molecules. Different cells and different developmental stages of the same cell have different molecular components and contents. For example, there are usually metabolic differences in the process of cell carcinogenesis. By detecting the molecular components in the cell, different types of cells can be identified. Raman spectrum is a molecular scattering spectrum that can reflect the intrinsic vibration information of molecular covalent bonds, and thus the molecular species and content information of the cell can be obtained through the cell Raman spectrum. Single-cell Raman spectrum technology has the technical advantages of no label, non-destructive and in-situ detection, and has important application value in the fields of life science and precision medicine such as cell differentiation and carcinogenesis. However, cells have complex structures and component differences. Whether in the transverse direction or in the axial direction, there are significant molecular differences in the collection of cell interiors. Such differences will cause significant uncertainty in the detection signal of the cell, resulting in a serious decline in the accuracy and repeatability of subsequent analysis. Therefore, to achieve precise analysis of single cells based on Raman spectrum detection, precise positioning of the spatial position of cell spectrum collection is required.

[0003] Current cell Raman spectrum detection usually uses spontaneous Raman combined with a white light microscope. The sample to be analyzed is placed under the microscope, and the microscope focuses the Raman excitation light spot on the cell. After the cell Raman signal is excited, the objective collects the scattered signal and transmits it to the spectrometer to obtain the cell Raman spectrum for subsequent analysis. In order to further obtain the spatial image information of the cell, conventional Raman spectrum technology is integrated with emerging imaging technology for detection. Commonly used technologies include fluorescence spectrum analysis technology, optical coherence tomography technology, and stimulated Raman technology. Fluorescence labeling and staining methods are mature technical means for providing cell image information. Fluorescence labeling and staining methods use dyes or fluorescent groups as exogenous markers to modify cells, increase the contrast of intracellular molecules, and obtain spatial position information of the cell. Optical coherence tomography (OCT) is an emerging optical imaging technology that analyzes the optical coherence of the returned photons of the medium and the reference light to obtain information such as the refractive index of biological tissue, and can realize non-contact three-dimensional tomography of living tissues. Stimulated Raman is based on two different wavelength pulsed lasers that amplify the Raman scattering signal through their third-order nonlinear effect with the molecules of the substance, and combines the cell structure and molecular spectrum information to obtain the image information inside the cell.

[0004] The spot size of the Raman excitation light focused by the microscope objective is usually smaller than the size of the cell, and the traditional microscope acquires a planar structure image of a certain layer in the cell. The cell sample is positioned on the focal spot by a galvanometer or a translation stage for horizontal (lateral) positioning. In the horizontal direction, the cell profile has a clear boundary with the external environment, which can assist in cell positioning. However, in the direction of the optical axis of the microscope, the focal length of the microscope is manually adjusted by the detection personnel to image different layers of the cell and to position the cell in the axial (vertical) direction. The cell interior is a nearly transparent sample, and the cell Raman detection cannot be stained. The contrast of the internal structure of the cell is low, and the differences between the layers are almost indistinguishable. Therefore, the ordinary white light microscope cannot effectively distinguish the morphological differences of the cell in the axial direction, i.e., different layers, and the axial resolution is low. Only the experimenter's experience can be relied on for judgment, resulting in low axial positioning accuracy of the cell Raman excitation spot, and the consistency of the excitation spectrum cannot be guaranteed. Sampling errors are introduced in the subsequent analysis, resulting in a decrease in the overall technical analysis accuracy. Therefore, other imaging techniques are needed to provide the axial position information of the cell.

[0005] The use of dyes or fluorescent groups as exogenous markers in the method of fluorescent labeling and staining will seriously interfere with the life activities of the cell and affect the metabolic activities of small molecules in the cell. Even the observation of living cells is impossible. The combination of fluorescent imaging and Raman technology will destroy the technical advantages of label-free detection of Raman spectroscopy. Optical coherence tomography is a label-free detection technology, but the specificity of the technology comes from the difference in the refractive index of different media, which cannot reflect the intrinsic properties of molecules. The imaging resolution, especially the axial resolution, is on the order of 10 microns, and the size of a single cell is also about 10 microns. Therefore, it is difficult to meet the demand for fine analysis of the internal structure of a single cell.

[0006] Stimulated Raman is based on the nonlinear optical effect to enhance the Raman scattering efficiency, which can realize video rate Raman spectral imaging with sub-micron imaging resolution. The spectral signal is consistent with the spontaneous Raman spectrum, but the spectral resolution is low due to the limitation of the spectral characteristics of femtosecond laser, which cannot meet the demand for cell fingerprint spectrum analysis. It is often used for high-wavenumber spectral imaging with low spectral resolution demand. Our group previously proposed a method for obtaining cell images based on stimulated Raman spectral imaging for lateral positioning and detection of cell substructures to improve the accuracy of analysis. However, this method still needs to rely on manual adjustment of the microscope focal length by the experimenter to judge the axial position of the detection. The axial focusing position of stimulated Raman is different from that of spontaneous Raman, and the experimenter still needs to rely on experience to judge the axial focusing position of spontaneous Raman, resulting in a decrease in the axial detection accuracy and a decrease in the uncertainty and repeatability of the analysis results. SUMMARY

[0007] The application provides a single-cell spontaneous-stimulated Raman spectrum detection axial automatic positioning method and system.

[0008] The application aims to provide a single-cell spontaneous-stimulated Raman spectrum detection axial automatic positioning method, which specifically comprises the following steps:

[0009] S1. A single-cell spontaneous-stimulated Raman spectrum detection axial automatic positioning system is built, which comprises a Raman excitation light source unit, a signal excitation and collection unit, a spatial position scanning unit, a signal processing and system control unit; a cell sample is placed on the object table of the signal excitation and collection unit, the white light source and the camera are turned on, and the field of view is roughly focused under the white light mode so that the cell appears in the field of view;

[0010] S2. The stimulated Raman excitation light source motorized aperture of the Raman excitation light source unit is turned on, the spontaneous Raman light source motorized aperture of the Raman excitation light source unit is turned off, and the stimulated Raman detection waveband is set as the Raman frequency shift of lipid molecules;

[0011] S3. The motion range and single-step range of the precision displacement platform of the axial position scanning mechanism are set, the precision displacement platform drives the microscopic objective lens of the signal excitation and collection unit to perform submicron-level stepping along the axial direction, the precision displacement platform steps once, and the cell is subjected to stimulated Raman spectrum imaging once to obtain a stimulated Raman image; after the motion is completed, a three-dimensional stimulated Raman tomographic imaging stack of the cell is obtained;

[0012] S4. The axial displacement distance of the precision displacement platform is determined by an equal-interval method or a PID control algorithm, the axial displacement distance H of the precision displacement platform is adjusted, and the cell is subjected to stimulated Raman spectrum imaging, and the spontaneous Raman detection position is positioned at the light spot where the lipid molecules are focused in the stimulated Raman image;

[0013] S5. The spontaneous Raman and stimulated Raman axial focusing position compensation method is used to calculate the spontaneous Raman and stimulated Raman axial focusing position correction amount L, and the position of the precision displacement platform is set as H+L;

[0014] S6. The stimulated Raman excitation light source motorized aperture of the Raman excitation light source unit is turned off, the white light source is turned off, the spontaneous Raman light source motorized aperture of the Raman excitation light source unit is turned on, the spontaneous Raman spectrum acquisition parameters are set, the cell spontaneous Raman spectrum acquisition is performed, and the fine Raman spectrum acquisition of the cell is completed;

[0015] S7. The Raman spectrum of a cell sample of a known type is acquired according to steps S1-S6, a Raman spectrum database of different types of cells is established, and a cell discrimination model of different types is obtained; then, each cell discrimination model is trained and optimized;

[0016] S8. Collecting the Raman spectrum of the unknown type of cell sample to be tested according to steps S1-S6, inputting the Raman spectrum into the cell discrimination model to obtain a discrimination result, and realizing accurate analysis of the cell.

[0017] Preferably, the Raman excitation light source unit comprises a stimulated Raman excitation light source module and a spontaneous Raman excitation light source module, which correspond to two spectral detection modes of the system; the stimulated Raman excitation light source module is used to emit two beams of femtosecond pulsed laser signals with different center wavelengths, which are transmitted to the signal excitation and collection unit after modulation to excite stimulated Raman signals; the spontaneous Raman excitation light source module is used to emit narrowband laser with fixed wavelength, which is transmitted to the signal excitation and collection unit to excite cell sample Raman signals; the stimulated Raman excitation light source module and the spontaneous Raman excitation light source module both comprise a motorized aperture, and the switching of the spontaneous Raman light source and the stimulated Raman excitation light source is realized by switching of the motorized aperture;

[0018] The signal excitation and collection unit is used to focus the stimulated Raman excitation light on the sample to change the excitation energy, and then modulate the excitation light to obtain stimulated Raman signals;

[0019] The spatial position scanning unit comprises a transverse position scanning mechanism and an axial position scanning mechanism; three-dimensional spatial tomographic imaging of the cell is realized by cooperation of the transverse position scanning mechanism and the axial position scanning mechanism;

[0020] The system control unit comprises a computer, which is used to coordinate the cooperative work of each unit.

[0021] Preferably, the stimulated Raman excitation light source module comprises a femtosecond laser, a first optical path adjustment group, a second optical path adjustment group, a beam combining mirror group and a stimulated Raman excitation light source motorized aperture; the femtosecond laser emits two beams of femtosecond pulsed laser signals with different center wavelengths, which are transmitted to the signal excitation and collection unit after spatial beam combining by the beam combining mirror group after passing through the first optical path adjustment group and the second optical path adjustment group respectively to excite stimulated Raman signals;

[0022] The spontaneous Raman excitation light source module comprises a continuous wave laser and a spontaneous Raman light source motorized aperture; the continuous wave laser is used to emit narrowband laser with fixed wavelength, which is transmitted to the signal excitation and collection unit to excite cell sample Raman signals;

[0023] By controlling the switching of the spontaneous Raman light source motorized aperture and / or the stimulated Raman excitation light source motorized aperture, it is determined whether the spontaneous Raman light source or the stimulated Raman excitation light source can be delivered to the signal excitation and collection unit for Raman signal excitation.

[0024] Preferably, the first optical path adjustment group comprises, in sequence along the optical path direction, a half-wave plate, a polarization beam splitter prism and a dispersive medium;

[0025] The second light path adjustment group comprises, in sequence along the light path direction, a half-wave plate, a polarization beam splitter prism, a mirror, a delay line, a mirror, a modulator, and a dispersive medium.

[0026] The beam combiner is configured to combine the femtosecond pulse laser signals that have passed through the first light path adjustment group and the second light path adjustment group.

[0027] In the first light path adjustment group and the second light path adjustment group, the half-wave plate is configured to rotate the polarization angle of the light, the polarization beam splitter prism is configured to selectively transmit the light component of a fixed polarization direction and reflect the light component of other directions, the combination of the half-wave plate and the polarization beam splitter prism is configured to adjust the output power of the femtosecond laser, and the dispersive medium is configured to adjust the chirp degree of the two femtosecond pulse laser signals.

[0028] In the second light path adjustment group, the delay line is configured to change the characteristic wavelength of the stimulated Raman detection mode by translation.

[0029] Preferably, the modulator is connected with a power amplifier and a signal generator in sequence, the spatial light modulator is configured to receive the high-frequency modulation signal output by the signal generator and amplified by the power amplifier and perform amplitude high-frequency modulation, and the modulator is a spatial light modulator, an electro-optical modulator, or an acousto-optical modulator.

[0030] Preferably, the signal excitation and collection unit comprises a microscope objective, an object table, a condenser, a filter, a photodetector, and a lock-in amplifier.

[0031] After the stimulated Raman excitation light is focused on the sample on the object table by the microscope objective to generate the stimulated Raman scattering phenomenon, the excitation energy changes, then the excitation light passes through the sample and is reflected by the beam splitter, the excitation light modulated by the modulator is filtered out by the filter, the other excitation light is collected and detected by the photodetector, and the signal is transmitted to the lock-in amplifier to extract the stimulated Raman signal.

[0032] The plane of the transverse position scanning mechanism is perpendicular to the optical axis of the microscope objective, and the axis of the axial position scanning mechanism is parallel to the optical axis of the microscope objective; the transverse position scanning mechanism comprises a two-dimensional galvanometer group and a 4-f lens group composed of two lenses, and is configured to perform stimulated Raman spectrum scanning imaging on the cells; the axial position scanning mechanism comprises a precision displacement platform, and the precision displacement platform drives the microscope objective to move axially along the optical axis with sub-micron precision.

[0033] Preferably, the step S4 determines the axial displacement distance of the precision displacement platform by the equal-interval method, and adjusts the axial displacement distance H of the precision displacement platform; and the step S4 specifically comprises the following sub-steps:

[0034] S401. The method of equal-interval step scanning: calculate the variance A of the pixel intensity of each point in the acquired stimulated Raman image; assuming that the pixel intensity is denoted as Ixy, the pixel intensity variance A is calculated according to the following formula:

[0035] ;

[0036] wherein m represents the total number of pixels in the x dimension, n represents the total number of pixels in the y dimension, denotes the average intensity of all pixels in the image, I xy denotes the pixel intensity; The calculation formula is as follows:

[0037] ;

[0038] wherein m represents the total number of pixels in the x dimension, n represents the total number of pixels in the y dimension, I xy denotes the pixel intensity;

[0039] The focusing coefficient F of the stimulated Raman image is set as follows:

[0040] ;

[0041] S402. Select the precise displacement platform axial displacement H corresponding to the maximum value of the focusing coefficient, and set the axial displacement distance of the precise displacement platform as H;

[0042] S403. Perform stimulated Raman spectrum imaging on the cells, and position the spontaneous Raman detection position at the lipid molecule focusing spot in the stimulated Raman image.

[0043] Preferably, step S4 determines the axial displacement distance of the precise displacement platform through a PID control algorithm, and adjusts the axial displacement distance H of the precise displacement platform; specifically including the following sub-steps:

[0044] S401. Solving the focusing coefficient of the current stimulated Raman image: F(t) represents the focusing coefficient of the current stimulated Raman image, and F(t-1) represents the focusing coefficient of the image collected at the previous moving position of the precise displacement platform; t=1 before the precise displacement platform moves for the first time; F(0) is set to 0, and the position of the precise displacement platform is h at this time; the focusing coefficient F(t) of the current stimulated Raman image is calculated according to the following formula:

[0045] Calculate the variance A of the pixel intensity of each point in the current stimulated Raman image; assuming that the pixel intensity is denoted as Ixy, the pixel intensity variance A is calculated according to the following formula:

[0046] ;

[0047] wherein m represents the total number of pixels in the x-dimension, n represents the total number of pixels in the y-dimension, represents the average intensity of all pixels in the image, I xy represents the pixel intensity; The calculation formula is as follows:

[0048] ;

[0049] wherein m represents the total number of pixels in the x-dimension, n represents the total number of pixels in the y-dimension, I xy represents the pixel intensity;

[0050] The current stimulated Raman image focusing coefficient F(t) is represented as follows:

[0051] ;

[0052] The current focusing coefficient difference E(t) = F(t) - F(t-1); E(t-1) represents the focusing coefficient difference corresponding to the last precise displacement platform movement;

[0053] S402. Estimate the next movement distance Z of the precise displacement platform, and the calculation formula is as follows:

[0054] ;

[0055] wherein k p , k i , and k d are constants set according to actual conditions;

[0056] Set the precise displacement platform movement position as H = h + Z; let h = H, F(t-1) = F(t); E(t-1) = E(t); repeat the above process, and when E(t) or Z is less than a set threshold, stop; at this time, H is the axial positioning position of the precise displacement platform;

[0057] S403. Select the H value obtained by iteration of the PID control process, and set the axial displacement distance of the precise displacement platform as H;

[0058] S404. Perform stimulated Raman spectrum imaging on the cells, and position the spontaneous Raman detection position at the lipid molecule focusing light spot in the stimulated Raman image.

[0059] Preferably, the initial focusing coefficient difference E(0) in step S402 is 0; the value of the next movement distance Z of the precise displacement platform includes a positive value and a negative value, and the positive value and the negative value respectively represent the movement directions of the precise displacement platform as upward and downward.

[0060] Preferably, the precision displacement platform is a piezoelectric displacement device; the spontaneous Raman and stimulated Raman axial focusing position compensation method in the step S5 specifically comprises the following steps:

[0061] S501. Adjusting the position of the piezoelectric displacement device so that the spontaneous Raman signal intensity of the collected lipid molecules reaches the maximum, and recording the displacement distance H1 of the piezoelectric displacement device at this time;

[0062] S502. Calculating the spontaneous Raman and stimulated Raman axial focusing position correction amount L, and the expression is as follows: L = H1 - H;

[0063] In the formula, H1 is the displacement distance of the piezoelectric displacement device when the spontaneous Raman signal intensity of the lipid molecules reaches the maximum, and H is the axial displacement of the piezoelectric displacement device corresponding to the maximum pixel intensity variance;

[0064] The step S8 adopts a principal component analysis method or a linear discriminant analysis method to discriminate the unknown type of the cell sample to be measured.

[0065] Compared with the prior art, the present application can achieve the following beneficial effects:

[0066] Through the positioning system provided by the present application, the spatial three-dimensional distribution of the lipid molecules in the cells can be obtained in a label-free and non-staining manner, the axial position of the lipid molecule layer in each cell sample is used as a reference to realize the spatial three-dimensional position calibration of the cell Raman collection, and the Raman spectrum is accurately sampled. Compared with the traditional method, the positioning method provided by the present application selects the lipid molecule layer in the cells as the axial position detection reference for the Raman spectrum detection of different cells, reduces the sampling error of the axial position in the Raman spectrum detection process of different cells, greatly improves the consistency and repeatability of the measurement signals in the cell identification process based on the principal component analysis-linear discriminant analysis data-driven method, and thus greatly improves the accuracy of the cell identification method based on the principal component analysis-linear discriminant analysis data-driven method. At the same time, the method of the present application only needs the participation of a small amount of experimental personnel, and the collection process can be automatically performed by a computer control system, so that the collection efficiency is greatly improved. BRIEF DESCRIPTION OF DRAWINGS

[0067] Figure 1 is a schematic diagram of an optical path of a single-cell spontaneous-stimulated Raman spectrum detection axial automatic positioning system according to an embodiment of the present application.

[0068] Figure 2 is a detection result of the single-cell spontaneous-stimulated Raman spectrum detection axial automatic positioning according to an embodiment of the present application; in the figure, A is stimulated Raman axial tomography data, B is the axial distribution of the lipid molecule layer in the cell, C is the stimulated Raman transverse spectrum imaging of the cell after axial positioning, and D is the collected spontaneous Raman spectrum after three-dimensional spatial positioning.

[0069] Figure 3Figure is a cell classification scatter diagram obtained by Raman identification analysis of two cell lines DU145 and SIHA according to an embodiment of the present application; Figure A is a cell classification scatter diagram obtained by non-axial positioning detection, and Figure B is a classification scatter diagram established by collecting Raman spectrum data by using the method of the present application.

[0070] Figure 4 Figure is a PID positioning flow chart according to an embodiment of the present application.

[0071] Reference signs:

[0072] 1: femtosecond laser; 2: first half-wave plate; 3: first polarization beam splitter prism; 4: first dispersive medium; 5: spectrometer; 6: electric shutter of stimulated Raman excitation light source; 7: continuous wave laser; 8: electric shutter of spontaneous Raman light source; 9: second dichroic mirror; 10: third dichroic mirror; 11: two-dimensional galvanometer group; 12: first lens; 13: second lens; 14: first beam splitter; 15: camera; 16: piezoelectric positioner; 17: microscope objective; 18: stage; 19: condenser; 20: second beam splitter; 21: white light source; 22: filter; 23: photodetector; 24: computer; 25: lock-in amplifier; 26: signal generator; 27: power amplifier; 28: delay line; 29: first mirror; 30: second polarization beam splitter prism; 31: second half-wave plate; 32: second mirror; 33: spatial light modulator; 34: second dispersive medium; 35: third mirror; 36: first dichroic mirror. DETAILED DESCRIPTION

[0073] Hereinafter, embodiments of the present application will be described with reference to the accompanying drawings. In the following description, the same modules are denoted by the same reference numerals. In the case of the same reference numerals, their names and functions are also the same. Therefore, detailed descriptions thereof will not be repeated.

[0074] In order to make the objects, technical solutions and advantages of the present application clearer, the present application will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and do not constitute a limitation on the present application.

[0075] Reference Figure 1 The present application provides a single-cell spontaneous-stimulated Raman spectrum detection axial automatic positioning system, comprising: a Raman excitation light source unit, a spatial position scanning unit, a signal excitation and collection unit, a signal processing and system control unit.

[0076] The Raman excitation light source unit includes a stimulated Raman excitation light source module and a spontaneous Raman excitation light source module, corresponding to two spectral detection modes of the system; in operation, first, the stimulated Raman excitation light source module is used to obtain a three-dimensional image of cells, and the excitation position is automatically positioned to the region where lipid molecules are aggregated, and then the spontaneous Raman excitation light source module is switched to, so as to realize automatic and accurate spontaneous Raman detection.

[0077] The stimulated Raman excitation light source module includes a femtosecond laser 1, a first optical path adjustment group, a second optical path adjustment group, a beam combination lens group and a stimulated Raman excitation light source motorized iris 6; the core of the stimulated Raman excitation light source module is a femtosecond laser 1;

[0078] The femtosecond laser 1 emits two beams of femtosecond pulse laser signals with different central wavelengths, which enter the first optical path adjustment group and the second optical path adjustment group, respectively;

[0079] The first optical path adjustment group includes, in sequence along the optical path direction, a first half-wave plate 2, a first polarization beam splitter prism 3 and a first dispersion medium 4; the second optical path adjustment group includes, in sequence along the optical path direction, a second half-wave plate 31, a second polarization beam splitter prism 30, a first mirror 29, a delay line 28, a second mirror 32, a spatial light modulator 33 and a second dispersion medium 34; the half-wave plate is used for rotating the polarization angle of light, the polarization beam splitter prism is used for selectively transmitting light components with a fixed polarization direction and reflecting light components with other directions; the combination of the half-wave plate and the polarization beam splitter prism is used for adjusting the output power of the femtosecond laser 1; the dispersion medium is used for adjusting the chirp degree of the two beams of femtosecond pulse laser signals; the delay line 28 is composed of a motorized displacement platform carrying a pair of mutually perpendicular plane mirrors, and the motorized displacement platform can translate forward and backward along the arrow direction and change the characteristic wavelength of the stimulated Raman detection mode;

[0080] The signal generator 26, the power amplifier 27 and the spatial light modulator 33 are connected in sequence, and the spatial light modulator 33 is used for receiving high-frequency modulation signals output by the signal generator 26 and amplified by the power amplifier 27 and performing amplitude high-frequency modulation;

[0081] The spatial light modulator 33 can also be replaced by an electro-optic modulator or an acousto-optic modulator to realize the modulation function;

[0082] The beam combination lens group includes a third mirror 35 and a first dichroic mirror 36, and is used for spatially combining the femtosecond pulse laser signals passing through the first optical path adjustment group and the second optical path adjustment group;

[0083] The principle of the stimulated Raman excitation light source module is briefly described as follows: by adjusting the wavelength difference between the femtosecond pulses and changing the position of the mirror pair on the delay line 28, the characteristic wavelength of the stimulated Raman detection mode can be changed, and the lipid molecules in the cells are focused to a specific layer in the middle of the cells. In the present application, the axial position of the lipid molecule layer is innovatively used as a reference to calibrate the spatial axial position during detection of different cell samples, thereby improving the consistency of the detection data. Therefore, by setting the output wavelength of the femtosecond laser 1 and adjusting the delay line 28, the wavelength of the stimulated Raman spectrum imaging is set to the characteristic peak of the lipid (including but not limited to 2850cm -1 , 2930cm -1 , etc.). The output power of the femtosecond laser 1 is adjusted by the combination of the half-wave plate and the polarization beam splitter prism, and the chirp degree of the two femtosecond pulse laser signals is adjusted by the dispersive medium to improve the spectral resolution of the stimulated Raman detection. Any one of the two femtosecond pulses is high-frequency amplitude modulated by the spatial light modulator 33, and the modulation frequency is determined by the signal generator 26. Subsequently, the two femtosecond pulse lasers are spatially combined by the first dichroic mirror 36, and transmitted to the signal excitation and collection unit to excite the stimulated Raman signal.

[0084] The spontaneous Raman excitation light source module includes a continuous wave laser 7 and a spontaneous Raman light source motorized iris 8. The continuous wave laser 7 is used to emit narrowband laser with fixed wavelength, which is transmitted to the signal excitation and collection unit to excite the cell sample Raman signal. The cell sample Raman signal is transmitted to the spectrometer 5 for Raman spectrum analysis. A filter is arranged at the entrance slit of the spectrometer 5, which is used to filter out the laser excitation signal. In order to facilitate system integration and reduce the volume, the light incident to the spectrometer 5 can be folded by a mirror.

[0085] By controlling the switch of the spontaneous Raman light source motorized iris 8 and / or the stimulated Raman excitation light source motorized iris 6, it is determined whether the spontaneous Raman light source and / or the stimulated Raman excitation light source can be delivered to the signal excitation and collection unit for Raman signal excitation.

[0086] In a specific embodiment, the first dichroic mirror 36, the stimulated Raman excitation light source motorized iris 6, the second dichroic mirror 9, and the third dichroic mirror 10 are arranged in sequence along the light path propagation direction. The second dichroic mirror 9 and the third dichroic mirror 10 are used to reflect light of a certain wavelength or waveband and transmit light of another wavelength or waveband, so as to facilitate the separation of light of different wavelengths or wavebands.

[0087] The signal excitation and collection unit comprises a microscope objective 17, an objective table 18, a condenser 19, a filter 22, a photodetector 23 and a lock-in amplifier 25. After the stimulated Raman scattering phenomenon is generated on the sample on the objective table 18 by focusing the excited Raman excitation light by the microscope objective 17, the excitation energy changes, and then the changed excitation light passes through the sample, is reflected by the second beam splitter 20, and the excitation light modulated by the spatial light modulator 33 is filtered out by the filter 22. Another excitation light is collected and detected by the photodetector 23, and then the signal is transmitted to the lock-in amplifier 25 to extract the stimulated Raman signal.

[0088] The spatial position scanning unit comprises a transverse position scanning mechanism and an axial position scanning mechanism, and the two are combined to realize three-dimensional spatial tomographic imaging of the cell. The plane of the transverse position scanning mechanism is perpendicular to the optical axis of the microscope objective 17, and the axis of the axial position scanning mechanism is parallel to the optical axis of the microscope objective 17. The transverse position scanning mechanism comprises a two-dimensional galvanometer group 11 and a 4-f lens group composed of two lenses (a first lens 12 and a second lens 13), which is used for stimulated Raman spectral scanning imaging of the cell.

[0089] The axial position scanning mechanism comprises a precision displacement platform, which drives the microscope objective 17 to move axially along the optical axis with sub-micron precision, so as to realize accurate positioning. The precision displacement platform is a piezoelectric displacement device 16.

[0090] The principle of the spatial position scanning unit is as follows: the transverse position scanning mechanism (i.e. the galvanometer transverse scanning mechanism) and the axial position scanning mechanism (i.e. the microscope objective axial scanning mechanism) are combined. The microscope objective axial scanning mechanism is composed of a piezoelectric displacement device and a conventional optical objective. The piezoelectric displacement device drives the optical objective to move axially with sub-micron precision, so as to determine a specific axial position. The detection wavelength of the stimulated Raman mode is set as the characteristic wavelength of the lipid molecules of the target cell. Then, the cell is scanned and imaged by the scanning galvanometer to obtain a Raman spectral image reflecting the spatial distribution of the lipid molecules at the axial position. The intensity variance value of each pixel point of the image is used as the judgment basis for the focusing degree of the cell image. The PID control algorithm or the equal-interval step scanning method is used to automatically search for the image axial position with the maximum pixel point intensity variance value, so as to find the specific position of the lipid molecule layer in the cell. The lipid molecule layer is used as the normalized positioning reference for the detection of different cells, so as to determine the spatial axial position of the sampling. Then, according to the cell substructure information reflected by the stimulated Raman spectral image at the axial position, the transverse sampling position of the spontaneous Raman spectrum is determined. Then, the spontaneous Raman spectrum unit is used to collect the spontaneous Raman spectrum at the position.

[0091] The system control unit comprises a computer 24, which is used for coordinating the cooperative work of the units. For example, Figure 1The computer 24 can be connected with the signal generator 26, the photodetector 23, and the piezoelectric positioner 16, respectively.

[0092] Referring to Figures 1-3 , based on the above positioning system, an automatic positioning method for single-cell spontaneous-stimulated Raman spectrum detection axis is provided, which specifically comprises the following steps:

[0093] S1. Place the cell sample on the stage 18, turn on the white light source 21 and the camera 15, and coarsely focus in the white light mode to make the cells appear in the field of view.

[0094] S2. Turn on the stimulated Raman excitation light source motorized diaphragm 6 and turn off the spontaneous Raman light source motorized diaphragm 8, and set the stimulated Raman detection waveband to the Raman shift of lipid molecules, such as 2850 cm -1 , 2930 cm -1 , etc.

[0095] S3. Set the motion range and single-step range of the piezoelectric positioner 16, and drive the microscope objective 17 to perform sub-micron fine stepping along the axis by the piezoelectric positioner 16, and perform stimulated Raman spectrum imaging (XY plane imaging) of the cell once for each step of the piezoelectric positioner 16 to obtain a stimulated Raman image; after the motion is completed, a stimulated Raman tomographic imaging stack of the cell in three-dimensional space is obtained, as shown in A of Figure 2 .

[0096] S4. Determine the axial displacement distance of the piezoelectric positioner 16 by the equal-interval method or the PID control algorithm, adjust the axial displacement H of the piezoelectric positioner 16, and perform stimulated Raman spectrum imaging of the cell to position the spontaneous Raman detection position at the light spot where the lipid molecules are focused in the stimulated Raman image; in this step, all axial positions of the stimulated Raman image are searched automatically by the PID control algorithm or the equal-interval step scanning method, and the maximum value of the focusing coefficient is found in the focusing coefficient values of the stimulated Raman images corresponding to all the axial positions; in specific embodiments, the two methods are as follows:

[0097] When the equal-interval method is used, the following sub-steps are specifically included:

[0098] S401. Calculate the focusing coefficient by the interval step scanning method: calculate the pixel intensity variance A of each stimulated Raman image obtained; assuming that the pixel intensity is Ixy, the pixel intensity variance A is calculated according to the following formula:

[0099] ;

[0100] In the formula, m represents the total number of pixels in the x dimension, n represents the total number of pixels in the y dimension, represents the average intensity of all pixels in the image, Ixy represents the pixel intensity; The calculation formula is as follows:

[0101] ;

[0102] In the formula, m represents the total number of pixels in the x dimension, n represents the total number of pixels in the y dimension, I xy represents the pixel intensity;

[0103] The focusing coefficient F of the stimulated Raman image is represented as follows:

[0104] ;

[0105] The axial distribution of the lipid molecule layer is shown as B in Figure 2 ;

[0106] S402. Select the axial displacement of the piezoelectric displacement device 16 corresponding to the maximum value of the focusing coefficient, and set the axial displacement distance of the piezoelectric displacement device 16 as H;

[0107] S403. Perform stimulated Raman spectrum imaging on the cell, and position the spontaneous Raman detection position at the lipid molecule focusing spot in the stimulated Raman image, as shown as C in Figure 2 .

[0108] When the PID control algorithm is adopted, the specific steps include the following sub-steps:

[0109] S401. Solve the focusing coefficient of the current stimulated Raman image: F(t) represents the focusing coefficient of the current stimulated Raman image, and F(t-1) represents the focusing coefficient of the image collected at the previous moving position of the piezoelectric displacement device 16 before moving to the current position; t=1 before the piezoelectric displacement device 16 moves for the first time; F(0) is set to 0, and the position of the piezoelectric displacement device 16 at this time is h; The calculation method of the focusing coefficient F(t) of the current stimulated Raman image is as follows:

[0110] Calculate the pixel intensity variance A of each point in the current stimulated Raman image; assuming that the pixel intensity is represented as Ixy, the specific calculation formula of the pixel intensity variance A is as follows:

[0111] ;

[0112] In the formula, m represents the total number of pixels in the x dimension, n represents the total number of pixels in the y dimension, represents the average intensity of all pixels in the image, I xy represents the pixel intensity; The calculation formula is as follows:

[0113] ;

[0114] wherein m represents the total number of pixels in the x-dimension, n represents the total number of pixels in the y-dimension, I xy represents the pixel intensity;

[0115] The current stimulated Raman image focusing coefficient F(t) is represented as follows:

[0116] ;

[0117] The current focusing coefficient difference E(t) = F(t) - F(t-1); E(t-1) represents the corresponding focusing coefficient difference of the last movement of the piezoelectric positioner 16, and the initial focusing coefficient difference E(0) is 0;

[0118] S402. Estimate the next movement distance Z of the piezoelectric positioner 16, and the calculation formula is as follows:

[0119] ;

[0120] wherein k p , k i , and k d are constants set according to actual conditions; the value of Z is positive or negative, representing the up and down movement directions of the piezoelectric positioner 16, respectively;

[0121] Set the movement position of the piezoelectric positioner 16 as H = h + Z; let h = H, F(t-1) = F(t); E(t-1) = E(t); repeat the above process, and stop when E(t) or Z is less than a set threshold; at this time, H is the axial positioning position of the piezoelectric positioner 16; the PID positioning flow chart is shown in Figure 4 ;

[0122] S403. Select the H value obtained by iteration of the PID control process, and set the axial displacement distance of the piezoelectric positioner 16 as H;

[0123] S404. Perform stimulated Raman spectrum imaging on the cells, and position the spontaneous Raman detection position at the lipid molecule focusing spot in the stimulated Raman image.

[0124] S5. Calculate the spontaneous Raman and stimulated Raman axial focusing position correction amount L using the spontaneous Raman and stimulated Raman axial focusing position compensation method, and set the position of the piezoelectric positioner 16 as H + L;

[0125] The spontaneous Raman and stimulated Raman axial focusing position compensation method specifically includes the following steps:

[0126] S501. Adjust the position of the piezoelectric positioner 16 so that the collected spontaneous Raman signal intensity of the lipid molecules reaches the maximum, and record the displacement distance H1 of the piezoelectric positioner 16 at this time;

[0127] S502. Calculate the self-Raman and stimulated Raman axial focusing position correction amount L, expressed as follows: L = H1-H;

[0128] In the formula, H1 is the displacement distance of the piezoelectric displacement device when the self-Raman signal intensity of the lipid molecule reaches the maximum, and H is the axial displacement of the piezoelectric displacement device corresponding to the maximum pixel intensity variance.

[0129] S6. Turn off the stimulated Raman excitation light source motorized iris of the Raman excitation light source unit, turn off the white light source, turn on the self-Raman light source motorized iris of the Raman excitation light source unit, set the self-Raman spectrum acquisition parameters, collect the cell self-Raman spectrum, and complete the cell fine Raman spectrum acquisition.

[0130] The self-Raman spectrum acquisition parameters include the waveband, etc.

[0131] S7. Collect the Raman spectrum of the known type of cell sample according to steps S1-S6, establish a Raman spectrum database of different types of cells, and obtain a cell discrimination model of different types; then train and optimize each cell discrimination model.

[0132] Each cell discrimination model is trained by machine learning method.

[0133] S8. Collect the Raman spectrum of the unknown type of cell sample to be measured according to steps S1-S6, substitute it into the cell discrimination model for discrimination and obtain the discrimination result, and realize the precise analysis of the cell.

[0134] In this step, the unknown type of cell sample to be measured is discriminated based on principal component analysis and linear discriminant analysis.

[0135] The advantage is that after collecting data to establish a Raman spectrum database (standard sample library) of different types of cells, the Raman spectrum of the cell sample to be measured is collected by the same method, and the unknown cell sample is discriminated based on principal component analysis and linear discriminant analysis. Compared with traditional micro-Raman technology, the discrimination accuracy can be greatly improved.

[0136] In order to ensure the discrimination accuracy, the experimental conditions of Raman spectrum collection need to be consistent for the training set samples and the subsequent test set samples, so the positioning of the cell excitation position is crucial, but there is no cell Raman test axial positioning method at present, which cannot guarantee the consistency of the test, resulting in the reduction of the analysis accuracy. At the same time, when collecting the cell Raman spectrum, the optimal axial position is the center position of the cell, the molecular composition is rich at the center position of the cell, the characteristic difference is obvious, and the molecular concentration is different from the center position, so the uncertainty of the axial position will lead to the existence of large difference in the Raman spectrum of the same cell. Therefore, the present application uses the lipid molecular layer in the cell as the test calibration reference, adopts the spontaneous-stimulated Raman dual-mode spectrum detection axial automatic positioning and accurate identification method to normalize and standardize the sampling conditions of different cell Raman spectra, and further improves the analysis accuracy. The cell Raman identification analysis of two cultured cell lines, DU145 and SIHA, is carried out, as shown in the accompanying Figure 3 FIG. 1A shows the classification results of principal component-linear discriminant analysis of the cell Raman spectrum data collected without axial positioning guide, Figure 3 FIG. 1B shows the classification results of the present application after the axial position is normalized, Figure 3 FIG. 1C shows the classification results of the present application after the axial position is normalized.

[0137] It should be understood that the various forms of the flow shown above can be reordered, added to, or deleted from. For example, the steps recorded in the present disclosure can be executed in parallel, sequentially, or in different orders, as long as the desired results of the technical solutions of the present disclosure can be achieved, which is not limited herein.

[0138] The above specific embodiments do not constitute a limitation on the scope of protection of the present application. Those skilled in the art should understand that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent replacements and improvements made within the spirit and principles of the present application should be included in the scope of protection of the present application.

Claims

1. A method for automatic positioning of single-cell spontaneous-stimulated Raman spectroscopy detection axis, characterized in that: Specifically comprising the following steps: S1. Build a single-cell spontaneous-stimulated Raman spectrum detection axial automatic positioning system, including a Raman excitation light source unit, a signal excitation and collection unit, a spatial position scanning unit, a signal processing and system control unit; Place the cell sample on the sample stage of the signal excitation and collection unit, turn on the white light source and the camera, and roughly focus in white light mode to make the cells appear in the field of view; S2. Turn on the stimulated Raman excitation light source motorized iris of the Raman excitation light source unit, turn off the spontaneous Raman light source motorized iris of the Raman excitation light source unit, and set the stimulated Raman detection band to the lipid molecule Raman frequency shift; S3. Set the motion range and single-step range of the precision displacement platform of the axial position scanning mechanism, drive the microscope objective of the signal excitation and collection unit to perform sub-micron stepping along the axial direction, and obtain a stimulated Raman image by performing stimulated Raman spectrum imaging on the cell once for each step of the precision displacement platform; after the movement is completed, obtain the three-dimensional spatial stimulated Raman tomographic imaging stack of the cell; S4. Determine the axial displacement distance of the precision displacement platform by the equidistant method or the PID control algorithm, adjust the axial displacement distance H of the precision displacement platform, and perform stimulated Raman spectrum imaging on the cell to position the spontaneous Raman detection position at the lipid molecule focusing spot in the stimulated Raman image; S5. Calculate the spontaneous Raman and stimulated Raman axial focusing position correction amount L using the spontaneous Raman and stimulated Raman axial focusing position compensation method, and set the position of the precision displacement platform to H+L; S6. Turn off the stimulated Raman excitation light source motorized iris of the Raman excitation light source unit, turn off the white light source, turn on the spontaneous Raman light source motorized iris of the Raman excitation light source unit, set the spontaneous Raman spectrum acquisition parameters, and perform cell spontaneous Raman spectrum acquisition to complete the fine Raman spectrum acquisition of the cell; S7. Collect the Raman spectrum of the cell sample of a known type according to steps S1-S6, establish a Raman spectrum database of different types of cells, and obtain a cell discrimination model of different types; then train and optimize each cell discrimination model; S8. Collect the Raman spectrum of the unknown type of cell sample to be measured according to steps S1-S6, input it into the cell discrimination model for discrimination and obtain the discrimination result, and realize the precise analysis of the cell.

2. The method of claim 1, wherein the method is characterized by: The Raman excitation light source unit includes a stimulated Raman excitation light source module and a spontaneous Raman excitation light source module, which correspond to two spectrum detection modes of the system; the stimulated Raman excitation light source module is used to emit two femtosecond pulse laser signals with different center wavelengths, which are transmitted to the signal excitation and collection unit after modulation to excite stimulated Raman signals; the spontaneous Raman excitation light source module is used to emit a fixed-wavelength narrowband laser, which is transmitted to the signal excitation and collection unit to excite cell sample Raman signals; the stimulated Raman excitation light source module and the spontaneous Raman excitation light source module each include a motorized iris, and the switching of the spontaneous Raman light source and the stimulated Raman excitation light source is realized by the opening and closing of the motorized iris; The signal excitation and collection unit is used for focusing stimulated Raman excitation light on the sample to change the excitation energy, and then modulating the excitation light to obtain a stimulated Raman signal. The spatial position scanning unit comprises a transverse position scanning mechanism and an axial position scanning mechanism; and three-dimensional spatial tomographic imaging of the cells is realized through cooperation of the transverse position scanning mechanism and the axial position scanning mechanism. The system control unit comprises a computer, which is used for coordinating cooperative work of the units.

3. The method of claim 2, wherein the method further comprises: determining the axial position of the single cell based on the Raman spectrum of the single cell. The stimulated Raman excitation light source module comprises a femtosecond laser, a first optical path adjusting group, a second optical path adjusting group, a beam combining mirror group and a stimulated Raman excitation light source motorized aperture; the femtosecond laser emits two beams of femtosecond pulsed laser signals with different central wavelengths, which pass through the first optical path adjusting group and the second optical path adjusting group respectively, and then are spatially combined by the beam combining mirror group, and are transmitted to the signal excitation and collection unit to excite a stimulated Raman signal; The spontaneous Raman excitation light source module comprises a continuous wave laser and a spontaneous Raman light source motorized aperture; the continuous wave laser is used for emitting a narrowband laser with a fixed wavelength, which is transmitted to the signal excitation and collection unit to excite a cell sample Raman signal; By controlling the switch of the spontaneous Raman light source motorized aperture or / and the stimulated Raman excitation light source motorized aperture, whether the spontaneous Raman light source or the stimulated Raman excitation light source can be transmitted to the signal excitation and collection unit to excite a Raman signal is realized.

4. The method of claim 3, wherein the method further comprises: determining the axial position of the single cell based on the Raman spectrum of the single cell. The first optical path adjusting group comprises, in sequence along the optical path direction, a half-wave plate, a polarization beam splitter prism and a dispersive medium; The second optical path adjusting group comprises, in sequence along the optical path direction, a half-wave plate, a polarization beam splitter prism, a mirror, a delay line, a mirror, a modulator and a dispersive medium; The beam combining mirror group is used for spatially combining the femtosecond pulsed laser signals passing through the first optical path adjusting group and the second optical path adjusting group; In the first optical path adjusting group and the second optical path adjusting group: the half-wave plate is used for rotating the polarization angle of light; the polarization beam splitter prism is used for selectively transmitting light components with a fixed polarization direction and reflecting light components with other directions; and the combination of the half-wave plate and the polarization beam splitter prism is used for adjusting the output power of the femtosecond laser; The dispersive medium is used for adjusting the chirp degree of the two femtosecond pulsed laser signals; In the second optical path adjusting group, the delay line changes the characteristic wavelength of the stimulated Raman detection mode by translation.

5. The method of claim 4, wherein the method further comprises: determining the axial position of the single cell based on the Raman spectrum of the single cell. The modulator is connected with a power amplifier and a signal generator in sequence; the modulator is a spatial light modulator, an electro-optic modulator or an acousto-optic modulator; the spatial light modulator is used for receiving a high-frequency modulation signal output by the signal generator and amplified by the power amplifier and performing amplitude high-frequency modulation.

6. The method of claim 5, wherein the method further comprises: The signal excitation and collection unit comprises a microscope objective, a stage, a condenser, a filter, a photodetector and a lock-in amplifier; After the stimulated Raman excitation light is focused on the sample on the stage by the microscope objective to produce a stimulated Raman scattering phenomenon, the excitation energy changes, then the excitation light passes through the sample and is reflected by the beam splitter, the excitation light modulated by the modulator is filtered out by the filter, the other excitation light is collected and detected by the photodetector, and then the signal is transmitted to the lock-in amplifier to extract a stimulated Raman signal; The plane of the transverse position scanning mechanism is perpendicular to the optical axis of the microscope objective, and the axis of the axial position scanning mechanism is parallel to the optical axis of the microscope objective; the transverse position scanning mechanism comprises a two-dimensional galvanometer group and a 4-f lens group composed of two lenses, and is used for stimulated Raman spectrum scanning imaging of cells; the axial position scanning mechanism comprises a precision displacement platform, and the precision displacement platform drives the microscope objective to move axially along the optical axis with sub-micron precision.

7. The method of claim 1, wherein the method further comprises: automatically positioning the single cell in the focal plane of the objective lens based on the axial position of the single cell. The step S4 determines the axial displacement distance of the precision displacement platform by an equal interval method, and adjusts the axial displacement distance H of the precision displacement platform; and the step S4 specifically comprises the following sub-steps: S401. An equal interval step scanning method is used to calculate the pixel intensity variance A of each point in the obtained stimulated Raman image; assuming that the pixel intensity is Ixy, the pixel intensity variance A is calculated according to the following formula: ； where m represents the total number of pixels in the x-dimension, n represents the total number of pixels in the y-dimension, represents the average intensity of all pixels in the image, I xy represents the pixel intensity; The calculation formula is as follows: ； In the formula, m represents the total number of image elements in the x dimension, n represents the total number of image elements in the y dimension, I xy represents the pixel intensity; A focus coefficient F of the stimulated Raman image is set, and is expressed as follows: ； S402. The axial displacement H of the precision displacement platform corresponding to the maximum focus coefficient is selected, and the axial displacement distance of the precision displacement platform is set as H; S403. The stimulated Raman spectrum imaging of the cell is performed, and the spontaneous Raman detection position is positioned at the lipid molecule focusing spot in the stimulated Raman image.

8. The method of claim 1, wherein the method is a method of automatic positioning of single-cell spontaneous-stimulated Raman spectroscopy detection along an axis. The step S4 determines the axial displacement distance of the precision displacement platform by a PID control algorithm, and adjusts the axial displacement distance H of the precision displacement platform; and the step S4 specifically comprises the following sub-steps: S401. The focus coefficient of the current stimulated Raman image is solved; F(t) represents the focus coefficient of the current stimulated Raman image, and F(t-1) represents the focus coefficient of the image collected at the previous moving position of the precision displacement platform; t=1 before the precision displacement platform moves for the first time; F(0) is set as 0, and the position of the precision displacement platform is h; the focus coefficient F(t) of the current stimulated Raman image is calculated according to the following formula: The pixel intensity variance A of each point in the current stimulated Raman image is calculated; assuming that the pixel intensity is Ixy, the pixel intensity variance A is calculated according to the following formula: ； where m represents the total number of pixels in the x-dimension, and n represents the total number of pixels in the y-dimension, represents the average intensity of all pixels in the image, I xy represents the pixel intensity; The calculation formula is as follows: ； In the formula, m represents the total number of image elements in the x dimension, n represents the total number of image elements in the y dimension, I xy represents the pixel intensity; The focus coefficient F(t) of the current stimulated Raman image is expressed as follows: ； The current focus coefficient difference E(t)=F(t)-F(t-1); E(t-1) represents the focus coefficient difference corresponding to the last movement of the precision displacement platform; S402. The movement distance Z of the next precision displacement platform is estimated, and the calculation formula is as follows: ； wherein k p , k i , k d are constants set according to actual conditions; The moving position of the precision displacement platform is set as H=h+Z; h=H, F(t-1)=F(t), and E(t-1)=E(t); the above process is repeated, and when E(t) or Z is less than a set threshold value, the process is stopped; at this time, H is the axial positioning position of the precision displacement platform; S403. The H value obtained through the PID control process iteration is selected, and the axial displacement distance of the precision displacement platform is set as H; S404. The stimulated Raman spectrum imaging of the cell is performed, and the spontaneous Raman detection position is positioned at the lipid molecule focusing spot in the stimulated Raman image.

9. The method of claim 8, wherein the method further comprises: determining the axial position of the single cell based on the Raman spectrum of the single cell. The initial focus coefficient difference value E(0) in the step S402 is 0; the value of the next moving distance Z of the precision displacement platform includes a positive value and a negative value, and the positive value and the negative value respectively represent that the moving direction of the precision displacement platform is upward and downward.

10. The method of claim 1, wherein the method is an automatic positioning method for single-cell spontaneous-stimulated Raman spectroscopy detection in the axial direction. The precision displacement platform is a piezoelectric displacement device; the spontaneous Raman and stimulated Raman axial focusing position compensation method in the step S5 specifically includes the following steps: S501. Adjusting the position of the piezoelectric displacement device so that the collected spontaneous Raman signal intensity of the lipid molecules reaches the maximum, and recording the displacement distance H1 of the piezoelectric displacement device at this time; S502. Calculating the spontaneous Raman and stimulated Raman axial focusing position correction amount L, and the expression is as follows: L=H1-H; In the formula, H1 is the displacement distance of the piezoelectric displacement device when the spontaneous Raman signal intensity of the lipid molecules reaches the maximum, and H is the axial displacement of the piezoelectric displacement device corresponding to the maximum pixel intensity variance; The step S8 adopts a principal component analysis method or a linear discriminant analysis method to discriminate the unknown type of the to-be-tested cell sample.

Images