Automatic tissue section staining system based on raman spectroscopy and application method

Abstract

The application discloses an automatic tissue section staining system based on a Raman spectrum technology and an application method, and relates to the technical field of tissue section staining. The application discloses an automatic tissue section staining system based on a Raman spectrum technology and an application method, and relates to the technical field of tissue section staining. The application discloses an automatic tissue section staining system based on a Raman spectrum technology and an application method, and relates to the technical field of tissue section staining.

Description

Technical Field

[0001] This invention relates to the field of image processing technology. More specifically, this invention relates to an automated tissue section staining system and application method based on Raman spectroscopy. Background Technology

[0002] Cancer is a common challenge facing humanity, and accurate tumor classification is fundamental to cancer treatment. Currently, the gold standard in clinical practice is histopathology, primarily based on immunohistochemistry (IHC) for staining tissue sections. It relies on antigen-antibody immune reactions, linking antibodies to dye molecules. The specific binding of these antibodies to the antigen stains the locations where the antigen is present, allowing doctors to determine the tumor type based on the stained images.

[0003] Currently, in hospital pathology departments, performing histopathological diagnosis on a tissue section requires multiple staining processes. First, HE staining distinguishes between cell nuclei and cytoplasm by staining them with different colors, identifying the tumor region based on the rapid differentiation and dense nuclei of tumor cells. The second step is IHC staining, using various antibody-conjugated staining agents to stain the tissue section and identify the expression levels of corresponding antigens (tumor markers), thus enabling tumor subtyping. Taking breast cancer as an example, breast cancer can be classified into subtypes such as Luminal A, Luminal B, HER2-overexpressing, and triple-negative. The diagnosis of these subtypes requires determining the expression levels of four marker proteins: HER2, ER, PR, and Ki67. Using IHC requires four adjacent tissue sections, each stained with a dye conjugated to a different marker antibody. Therefore, this method involves numerous steps and requires significant manpower, resources, and time, especially the various antibodies and staining reagents, which represent a substantial expense. Summary of the Invention

[0004] One object of the present invention is to solve at least the above-mentioned problems and / or defects, and to provide at least the advantages described below.

[0005] To achieve these objectives and other advantages of the present invention, an automated tissue section staining system based on Raman spectroscopy is provided, comprising: a microconfocal Raman spectrometer constructed by means of a microscope and a spectrometer thereof, wherein the microscope is provided with a spraying unit for Raman signal enhancement at a position that cooperates with the stage of the tissue section sample;

[0006] The slide sample stage is adjustable in height via a telescopic assembly.

[0007] Preferably, the spraying unit includes:

[0008] A housing for storing Raman signal enhancement reagents;

[0009] A bracket fixed to one side;

[0010] Atomizing nozzle is installed at the free end of the bracket;

[0011] The atomizing nozzle is connected to the housing via pipes, valves, and a pump, and the pipes are in clearance fit with the bottom of the housing.

[0012] The bracket and the atomizing nozzle are connected by a connector, and the connector is adjusted so that the atomizing nozzle can move horizontally along the stage of the sliced ​​sample.

[0013] Preferably, it also includes a control circuit and control panel for switching the operating states of the telescopic components, adjusting mechanisms, valves, and pumps.

[0014] An application method for an automated tissue section staining system based on Raman spectroscopy includes:

[0015] S1. Sample preparation: The sample slides on the stage are photographed using a microscope, and the obtained optical images are acquired by conventional Raman spectroscopy area scanning to distinguish tumor tissue from other normal tissues through the generated conventional Raman staining images.

[0016] S2. The control panel controls the spraying unit to spray nano-gold particles onto the sample to complete the Raman signal enhancement operation.

[0017] S3. Based on the tumor region marked in the conventional Raman staining image in S1, position the microscope lens at the corresponding position on the sample and perform surface-enhanced Raman spectroscopy on the sample using a spectrometer.

[0018] S4. Generate surface-enhanced Raman staining images based on the colors corresponding to the types of marker proteins in the spectrum, so as to distinguish the expression of various markers on tumor tissue.

[0019] S5. Overlay the conventional Raman staining image from S1 with the surface-enhanced Raman staining image from S4 to obtain a Raman staining image corresponding to the tissue section.

[0020] Preferably, the area scanning acquisition in S1 and S3 refers to: within a predetermined area, moving the laser spot or the sample relative to each other in the horizontal and vertical dimensions at fixed spatial intervals to acquire the Raman spectrum of each point in that area;

[0021] In S1 and S4, Raman staining refers to analyzing Raman surface scan data using algorithms, classifying the Raman spectra of each pixel, obtaining the tissue and / or protein information of that pixel, and staining each pixel with different colors according to the classification, to obtain a color image that can reflect the distribution of tissue and protein.

[0022] Preferably, in S4, the surface-enhanced Raman staining image is realized based on the surface-enhanced Raman surface scanning spectroscopy processing algorithm, and the surface-enhanced Raman surface scanning spectroscopy processing algorithm is based on the characteristic peak information of the surface-enhanced Raman spectrum, and then determines which biomarker protein the spectrum belongs to.

[0023] The identification of a protein involves comparing the characteristic peaks of the measured Raman spectrum with the spectra in a database containing Raman spectra of various biomarker proteins to confirm its protein type.

[0024] Preferably, the comparison method includes:

[0025] S41. Extract the peak position, peak intensity, and peak width information of the characteristic peaks in the Raman spectrum of the analyte based on the peak feature extraction algorithm;

[0026] S42. Match the extracted characteristic peak position information with the peak characteristic information of the biomarker protein Raman spectrum stored in the database to find the spectral peaks that appear simultaneously on the analyte and the biomarker protein, and calculate the proportion C of peaks with matching positions. position ;

[0027] S43. Further compare the peak intensity and peak width information of the matched spectral peaks to obtain the matching degree C of each matched spectral peak. shape ;

[0028] S44. Obtain the consistency parameter C between the analyte spectrum and the biomarker protein spectrum based on the degree of matching. SERS This is used to determine whether the spectrum originates from a marker protein.

[0029] Preferably, in S42, the C position The formula for obtaining it is:

[0030]

[0031] In the above formula, n total n represents the total number of spectral peaks in the protein biomarker spectrum. match This represents the number of spectral peaks appearing simultaneously on the analyte and biomarker proteins.

[0032] Preferably, in S42, the C shape The formula for obtaining it is:

[0033]

[0034] In the above formula, and W i m This represents the intensity and width of the i-th matching peak in the spectrum of the analyte. and W i p This represents the intensity and width of the i-th matching peak in the marker protein spectrum.

[0035] Preferably, in S42, the C SERS The formula for obtaining it is:

[0036]

[0037] The present invention has at least the following beneficial effects: compared with existing immunohistochemistry methods, this system can achieve fully automated operation, eliminating the cumbersome operation procedures of existing methods and greatly saving time and reagent costs.

[0038] The Raman stained images obtained by this system can be used to distinguish between tumor tissue and normal tissue, and can further differentiate tumor types, providing doctors with accurate and reliable auxiliary references for diagnosing cancer patients.

[0039] Other advantages, objectives and features of the present invention will become apparent in part from the following description, and in part from those skilled in the art through study and practice of the invention. Attached Figure Description

[0040] Figure 1 This is a schematic diagram of the structure of the tissue section Raman staining system of the present invention;

[0041] Figure 2 for Figure 1 A schematic diagram of the structural layout of the spraying unit on a medium-sized microscope;

[0042] Figure 3 This is a schematic diagram of the system operation process of the present invention;

[0043] Figure 4 A schematic diagram of the process for generating stained images from conventional Raman surface scanning and data processing;

[0044] Figure 5 A schematic diagram of the process for generating stained images through surface-enhanced Raman scanning and data processing;

[0045] Figure 6 Raman stained images of breast cancer tissue sections;

[0046] Figure 7 Raman stained images of lung cancer tissue sections;

[0047] Figure 8This is a schematic diagram of a routine immunohistochemical assay.

[0048] Figure 9 This is a schematic diagram of the enhanced Raman detection method of the present invention. Detailed Implementation

[0049] The present invention will now be described in further detail with reference to the accompanying drawings, so that those skilled in the art can implement it based on the description.

[0050] Raman spectroscopy reflects the vibrational modes of chemical bonds in molecules, serving as a fingerprint spectroscopy technique. It can also be used to identify tumor marker proteins on tissue sections. Raman spectroscopy has a wide range but relatively narrow peaks, allowing the characteristic peaks of different tumor marker proteins to be simultaneously represented in a single Raman spectrum without overlapping. Therefore, by establishing a Raman spectral database of tumor marker proteins, the expression levels of multiple tumor markers can be measured in a single detection, thus providing auxiliary reference for the diagnosis of tumor molecular subtyping.

[0051] This invention aims to provide a fully automated tissue section staining system based on Raman spectroscopy, which can replace the existing immunohistochemical methods used in hospital pathology departments for multi-step staining of tissue sections. It eliminates the need for staining reagents, simplifies the cumbersome procedures, and makes tumor diagnosis more efficient and accurate. In practical applications, the method of this fully automated tissue section staining system based on Raman spectroscopy first identifies normal and tumor tissues using conventional Raman spectroscopy, and then identifies the types and expression levels of tumor marker proteins using surface-enhanced Raman spectroscopy. The results are presented in the form of Raman stained images, enabling auxiliary diagnosis of tumors and their molecular subtyping.

[0052] like Figure 1-2The hardware device in this invention system is based on a microscopic confocal Raman spectrometer (i.e., a combination of spectrometer 1 and microscope 2), with a reagent spraying device 3 installed. The spraying device is located on one side of the microscope base 20 and below the lens 21. It is a modified microporous atomizing element based on system requirements and structural design. It consists of an atomizing nozzle 30, a nozzle support 31, a control circuit (not shown), and a reagent kit 32. The atomizing nozzle is a microporous atomizing sheet that generates high-frequency oscillations through piezoelectric effect, atomizing the liquid reagent before spraying. The nozzle support controls the nozzle position; it extends only when spraying is needed and retracts at other times. The extended position is fixed at 5 cm above the slice, with the horizontal position located at the center of the identified slice's outer contour. The control circuit controls the spraying time and number of repetitions to control the amount of sprayed reagent. This amount is determined based on the slice area calculated from the center of the identified slice's outer contour. The kit contains reagents to be sprayed, such as colloidal gold solution, which are connected to the nozzle via a tubing. Specifically, in implementation, the atomizing nozzle is connected to the kit (also called the box or barrel) via tubing 33, valve 34, and pump 35, with the tubing having a clearance fit with the bottom of the kit. The support is connected to the atomizing nozzle via a connector, and the connector allows the atomizing nozzle to move horizontally along the sample stage via an adjustment mechanism (such as the cooperation of a lead screw and a motor or a hydraulic cylinder to achieve reciprocating horizontal position adjustment). It also includes a control circuit (not shown) for switching the working states of the telescopic component, adjustment mechanism, valve, and pump (control panel).

[0053] like Figure 2 The specific operating procedures of this system include:

[0054] S1: Sample preparation: Place an ultra-flat gold-plated silicon substrate (hereinafter referred to as ultra-flat gold substrate) on a glass slide, and then place the paraffin section of the tissue to be tested flat on the ultra-flat gold substrate for dewaxing.

[0055] S2: Place the sample on the system stage, the system automatically focuses, captures an optical image of the slice, and identifies the outer contour of the slice (manual selection function is also retained). Then, the slice area is scanned and acquired using conventional Raman spectroscopy.

[0056] S3: After the routine Raman scan, the built-in algorithm automatically processes the data and generates a routine Raman stained image, distinguishing tumor tissue from other normal tissues. The specific process is shown in the attached document. Figure 3 As shown.

[0057] S4: The system automatically sprays nano-gold particles onto the slices.

[0058] S5: The system automatically locates the corresponding position on the slice based on the tumor region marked in the conventional Raman staining image and performs surface-enhanced Raman spectroscopy scanning acquisition.

[0059] S6: After the surface-enhanced Raman scan is completed, the built-in algorithm automatically processes the data. Each point is stained with a different color according to the type of biomarker protein corresponding to its spectrum, generating a surface-enhanced Raman staining image, which displays the expression of various biomarkers on tumor tissue. The specific process is as follows: Figure 4 As shown.

[0060] S7: The system overlays conventional Raman staining images and surface-enhanced Raman staining images to generate Raman staining images of tissue sections for pathologists to use as a diagnostic reference.

[0061] In S2, Raman area scanning refers to the process of moving a laser spot (or a relatively moving sample) at fixed spatial intervals in both the horizontal and vertical dimensions within a certain area to collect the Raman spectra of each point within that area. Therefore, Raman area scanning data can also be viewed as an image, with each pixel displaying a Raman spectrum.

[0062] Raman staining in S3 and S6 refers to analyzing Raman surface scan data using algorithms, classifying each pixel after Raman spectral analysis, and obtaining tissue and protein information for that pixel (such as whether it belongs to tumor or normal tissue, and what kind of marker protein it belongs to). Based on the classification, each pixel is stained with a different color to obtain a color image that can reflect the distribution of tissue and protein.

[0063] The conventional Raman surface scanning spectroscopy processing algorithm used in S3 and the surface-enhanced Raman surface scanning spectroscopy processing algorithm used in S6 are described in detail below:

[0064] Both conventional and surface-enhanced Raman surface scanning spectroscopy algorithms require analysis of spectral peak characteristics for discrimination. To extract peak characteristic information from Raman spectra, we developed a Raman spectral peak characteristic extraction algorithm based on continuous wavelet transform (CWT). This algorithm obtains the peak position, intensity, and width information of all spectral peaks for subsequent analysis, including both the conventional Raman spectroscopy analysis described here and the enhanced Raman spectroscopy analysis discussed below. The definition of continuous wavelet transform (CWT) is as follows:

[0065]

[0066] The signal x(t) can be decomposed into a series of wavelet mother functions. The superposition of signals. It has a scale parameter 'a', which can change the size of its transform window, thereby enabling the analysis of local characteristics of the signal.

[0067] In practical applications, the value of 'a' needs to be set. As 'a' takes different values, the continuous wavelet transform returns a two-dimensional NA×NX wavelet coefficient matrix, where NA is the number of scaling parameters 'a' and NX is the signal length. The value at each point in the matrix can be considered as the similarity between the wavelet mother function with the smaller scaling parameter and the signal at that location. We choose the Mexican hat function (Mexh) as the wavelet mother function:

[0068]

[0069] A continuous wavelet transform is performed on the spectrum, with the scale parameter varying exponentially from 1.5 to 50, yielding a wavelet coefficient matrix. Subsequently, the ridge peak-finding method is used to search for spectral peaks. Ridges represent maxima that exist continuously across multiple scales. The location of a ridge is considered to indicate the presence of a peak. Within each ridge, the position where the scale parameter is maximized is the peak position, and the scale parameter represents the peak width. Valid peak positions are then selected by comparing the signal-to-noise ratio (SNR) of the peak positions identified by the ridges (the ratio of the peak position to the baseline standard deviation). This algorithm uses a scale parameter of: e. 0.4 ~e 4.0 With a minimum signal-to-noise ratio of 1.5, the baseline standard deviation is calculated by considering the 20% of points with the lowest intensity as the baseline, enabling effective searching for strong and weak peaks and extraction of peak width. Peak intensity can be obtained by finding the corresponding intensity value in the spectrum based on the identified peak position.

[0070] The basic principle of the conventional Raman surface scan spectroscopy processing algorithm in S3 is based on the differences in peak intensity ratio and peak area in conventional Raman spectra to distinguish between tumor tissue and normal tissue. In our previous studies, we found that in the conventional Raman spectra of tumor tissue and normal tissue, the peak intensity ratio and peak area at 1330 cm⁻¹ are different. -1 / 1260cm -1 Peak intensity ratio and 1450cm -1 The peak areas exhibit relatively stable differences, which can be used to distinguish the spectra of tumors and normal tissues. To quantitatively represent these differences, we define a quantity C. normal The calculation formula is as follows:

[0071]

[0072] Among them I 1260 and I 1330 They represent 1260cm respectively -1 Peak and 1330cm -1 Peak intensity, S 1450 Indicates 1450cm -1The peak area, and all three factors can be obtained directly or by calculation using the spectral peak feature extraction algorithm proposed in Part 1. a1 and a2 are two adjustable coefficients, whose values ​​were obtained through calculation and analysis of conventional Raman spectra of a large number of tumor and normal tissue samples obtained in previous experiments. When C... normal A value ≥1 indicates that the spectrum originates from tumor tissue; otherwise, it originates from normal tissue. Therefore, for each spectrum obtained from a conventional Raman scan, C0 is calculated. normal The values ​​are used to distinguish between tumors and normal tissues, and then different colors are assigned to the pixels corresponding to the spectra of tumors and normal tissues to generate conventional Raman staining images.

[0073] The surface-enhanced Raman (SMR) scanning spectroscopy algorithm in S6 determines the type of biomarker protein a spectrum belongs to based on certain characteristic peaks. Since proteins are macromolecules, Raman spectra are complex, requiring multiple characteristic peaks for simultaneous discrimination. This algorithm first establishes a database containing Raman spectra of various biomarker proteins, then compares the characteristic peaks of the measured Raman spectra with the spectra in the database to confirm the most likely protein type.

[0074] The specific comparison method is as follows: First, the peak position, peak intensity, and peak width information of the characteristic peaks in the Raman spectrum of the analyte are extracted using a peak feature extraction algorithm. Then, it is compared with the peak feature information of the Raman spectra of biomarker proteins stored in the database: First, the peak positions are matched to find the peaks that appear simultaneously on the analyte and biomarker proteins, and the proportion of matching peaks is calculated; second, for the matching peaks, the peak intensity and peak width information are compared, and the degree of matching for each matching peak is calculated; third, the above information is combined to obtain the consistency parameter between the analyte spectrum and the biomarker protein spectrum, which is used to determine whether the spectrum originates from the biomarker protein. The entire calculation process can be expressed by the following formula: For each biomarker protein recorded in the database, the consistency parameter C between the analyte spectrum and the biomarker protein spectrum is calculated. SERS :

[0075]

[0076] Where C position The percentage of peaks in the spectrum indicating peak position matching is calculated as follows:

[0077]

[0078] n total n represents the total number of spectral peaks in the protein biomarker spectrum. match This represents the number of spectral peaks appearing simultaneously on the analyte and biomarker proteins.

[0079] C shape The degree of matching of the spectral peaks is indicated by the following calculation method:

[0080]

[0081] and W i m This represents the intensity and width of the i-th matching peak in the spectrum of the analyte. and W i p This represents the intensity and width of the i-th matching peak in the marker protein spectrum.

[0082] The more spectral peaks that match the peak positions, the higher the degree of matching, and the higher the consistency parameter C. SERS The higher the coefficient of agreement, the better the consistency between the spectrum of the analyte and the spectrum of the biomarker protein. In practical applications, the consistency coefficient between the spectrum of the analyte and the spectra of all biomarker proteins in the database is calculated. A discrimination threshold t is set. When the consistency coefficient is less than t, the analyte is considered not to belong to the biomarker protein. When the consistency coefficient exceeds t, the biomarker protein with the highest consistency coefficient is identified as the analyte.

[0083] In surface-enhanced Raman scanning, the spectrum of each point is analyzed using the algorithm described above to determine the type of marker protein. Then, the pixel is stained with the corresponding color based on the marker protein type, generating a surface-enhanced Raman stained image. For example, pixels identified as protein A are assigned red, pixels identified as protein B are assigned yellow, pixels identified as protein C are assigned green, and so on, using different colors to distinguish the distribution locations of different proteins.

[0084] The specific steps of S4 are as follows:

[0085] S41: Move the stage to lower the slice height to the minimum height without moving the horizontal position during the process.

[0086] S42: The electric support of the spraying device controls the spray head to move to the center of the slice outline.

[0087] S43: The nozzle sprays reagent onto the slide. The amount of reagent used is determined based on the slide area calculated from the center of the identified outer contour.

[0088] S44: The electric support of the spraying device controls the nozzle to move out of the microscope's field of view.

[0089] S45: Move the stage to return the slice to the height before spraying and ensure the correct height through autofocus.

[0090] Example 1:

[0091] Using this invention, Raman staining of breast tissue sections revealed the distribution of tumor and normal tissues, as well as the expression levels of four marker proteins: HER2, Ki67, ER, and PR. Figure 6 The images shown are Raman stained images of breast cancer tissue sections (the left image is a conventional Raman stained image, with the orange area representing the tumor tissue; the right image is an enhanced Raman stained image, with the purple area representing HER2, the brown area representing Ki67, the pink area representing ER, and the red area representing PR).

[0092] Example 2:

[0093] The present invention was used to Raman stain lung tissue sections to show the differences between lung squamous cell carcinoma and lung adenocarcinoma tissues. For example... Figure 7 The images shown are Raman stained sections of lung cancer tissue. The left image is lung squamous cell carcinoma tissue, with the red area representing lung squamous cell carcinoma marker proteins. The right image is lung adenocarcinoma tissue, with the pink area representing lung adenocarcinoma marker proteins.

[0094] As can be seen from Examples 1 and 2, the Raman staining images obtained by this system can distinguish between tumor tissue and normal tissue, and can further distinguish tumor subtypes, providing doctors with accurate and reliable auxiliary references for diagnosing cancer patients.

[0095] Furthermore, all detection operations in this invention are completed by the system in one go, eliminating the need for cumbersome manual operations and a wide variety of antibodies and staining reagents. Compared with existing immunohistochemical methods, this greatly simplifies the operation process and reduces reagent costs. Specifically, for example... Figure 8 The schematic diagram of the routine immunohistochemistry test shows that the operation process is complicated. From the fixation step at point A to the final mounting and observation, it requires about 14 steps. The operation time of each step needs to be strictly controlled. Under such complicated operation and process control, it takes about 5 hours to complete the test. At the same time, multi-target detection requires multiple rounds of staining. One test procedure costs about 700 yuan.

[0096] And such Figure 9 As shown, the enhanced Raman detection of this invention takes up 0.5 hours, primarily for Raman surface scanning, a step that is automatically completed by the system without manual intervention. This simplifies operation, and multi-target detection can be completed in a single surface scan. The consumable cost for a single test procedure is less than 10 yuan. Therefore, compared to existing technologies, this invention offers a simpler, easier-to-implement, faster, and more cost-effective operation.

[0097] The above solution is merely an illustration of a preferred example and is not limited thereto. When implementing this invention, appropriate substitutions and / or modifications can be made according to the user's needs.

[0098] The number of devices and processing scale described herein are for the purpose of simplifying the description of the invention. Applications, modifications, and variations of the invention will be readily apparent to those skilled in the art.

[0099] Although embodiments of the present invention have been disclosed above, they are not limited to the applications listed in the specification and embodiments. It can be applied to various fields suitable for the present invention. Other modifications can be readily made by those skilled in the art. Therefore, without departing from the general concept defined by the claims and their equivalents, the present invention is not limited to the specific details and examples shown and described herein.

Claims

1. An application method for an automated tissue section staining system based on Raman spectroscopy, wherein the application method is for non-therapeutic and / or non-diagnostic purposes, and the system comprises: A microconfocal Raman spectrometer constructed using a microscope and a spectrometer used in conjunction with it is characterized in that the microscope is provided with a spraying unit for Raman signal enhancement at a position that cooperates with the stage of the sliced ​​sample. The slide sample stage is adjustable in height via a telescopic assembly. The application method includes: S1. Sample preparation: The sample slides on the stage are photographed using a microscope, and the obtained optical images are acquired by conventional Raman spectroscopy area scanning to distinguish tumor tissue from other normal tissues through the generated conventional Raman staining images. S2. The control panel controls the spraying unit to spray nano-gold particles onto the sample to complete the Raman signal enhancement operation. S3. Based on the tumor region marked in the conventional Raman staining image in S1, position the microscope lens at the corresponding position on the sample and perform surface-enhanced Raman spectroscopy on the sample using a spectrometer. S4. Generate surface-enhanced Raman staining images based on the colors corresponding to the types of marker proteins in the spectrum, so as to distinguish the expression of various markers on tumor tissue. S5. Overlay the conventional Raman staining image in S1 with the surface-enhanced Raman staining image in S4 to obtain a Raman staining image corresponding to the tissue section. In S4, the surface-enhanced Raman staining image is realized based on the surface-enhanced Raman surface scanning spectroscopy processing algorithm, and the surface-enhanced Raman surface scanning spectroscopy processing algorithm is based on the characteristic peak information of the surface-enhanced Raman spectrum, and then determines which biomarker protein the spectrum belongs to. Protein classification involves comparing the characteristic peaks of the measured Raman spectrum with spectra in a database containing Raman spectra of various biomarker proteins to confirm the protein type to which it belongs. The comparison methods include: S41. Extract the peak position, peak intensity, and peak width information of the characteristic peaks in the Raman spectrum of the analyte based on the peak feature extraction algorithm; S42. Match the extracted characteristic peak position information with the peak characteristic information of the Raman spectra of biomarker proteins stored in the database to find the spectral peaks that appear simultaneously on the analyte and biomarker proteins, and calculate the proportion of peaks with matching peak positions. ; S43. Further compare the peak intensity and peak width information of the matched spectral peaks to obtain the degree of matching for each matched spectral peak. ; S44. Obtain the consistency parameter between the analyte spectrum and the biomarker protein spectrum based on the degree of matching. This is used to determine whether the spectrum originates from a marker protein; In S43, the The formula for obtaining it is: In the above formula, and Indicates the first i The intensity and width of each matched peak in the spectrum of the analyte. and Indicates the first i The intensity and width of each matching peak in the spectrum of the biomarker protein. The number of spectral peaks appearing simultaneously on the analyte and biomarker proteins; In S44, the The formula for obtaining it is: 。 2. The application method of the automated tissue section staining system based on Raman spectroscopy as described in claim 1, characterized in that, The spraying unit includes: A housing for storing Raman signal enhancement reagents; A bracket fixed to one side; Atomizing nozzle is installed at the free end of the bracket; The atomizing nozzle is connected to the housing via pipes, valves, and a pump, and the pipes are in clearance fit with the bottom of the housing. The bracket and the atomizing nozzle are connected by a connector, and the connector is adjusted so that the atomizing nozzle can move horizontally along the stage of the sliced ​​sample.

3. The application method of the automated tissue section staining system based on Raman spectroscopy as described in claim 1, characterized in that, It also includes control circuits and control panels for switching the operating states of telescopic components, regulating mechanisms, valves, and pumps.

4. The application method of the automated tissue section staining system based on Raman spectroscopy as described in claim 1, characterized in that, In S1 and S3, area scanning acquisition refers to: within a predetermined area, moving the laser spot or the sample relative to it in the horizontal and vertical dimensions at fixed spatial intervals to acquire the Raman spectrum of each point in that area. In S1 and S4, Raman staining refers to analyzing Raman surface scan data using algorithms, classifying the Raman spectra of each pixel, obtaining the tissue and / or protein information of that pixel, and staining each pixel with different colors according to the classification, to obtain a color image that can reflect the distribution of tissue and protein.

5. The application method of the automated tissue section staining system based on Raman spectroscopy as described in claim 1, characterized in that, In S42, the The formula for obtaining it is: In the above formula, This represents the total number of spectral peaks in the protein biomarker spectrum. This represents the number of spectral peaks appearing simultaneously on the analyte and biomarker proteins.

Images