Classification model of sheep echinococcosis and sheep brucellosis based on serum Raman spectroscopy, its establishment method, and detection device

Abstract

The present invention provides a classification model for sheep echinococcosis and sheep brucellosis based on serum Raman spectroscopy, a method for establishing the model, and a detection device. The method for establishing the classification model for sheep echinococcosis and sheep brucellosis based on serum Raman spectroscopy comprises: (1) collecting blood from healthy sheep and diseased sheep, extracting serum, and obtaining serum samples; (2) detecting the serum samples using a laser Raman spectrometer to obtain Raman spectral data; (3) performing noise reduction, baseline correction, and normalization on the Raman spectral data to obtain pre-processed spectral data; (4) performing principal component analysis (PCA) dimensionality reduction on the processed spectral data, and then establishing a linear discriminant analysis (LDA) or support vector machine (SVM) classification model to obtain a classification model for sheep echinococcosis and sheep brucellosis. The classification model and detection device established by the present invention based on machine learning algorithms and Raman spectroscopy have potential in screening sheep infected with echinococcosis and sheep infected with brucellosis.

Description

Technical Field

[0001] The present invention specifically relates to a classification model of sheep echinococcosis and sheep brucellosis based on serum Raman spectroscopy, an establishment method thereof, and a detection device. Background Art

[0002] Echinococcosis, also known as echinococcosis, is a zoonotic parasitic disease caused by the larvae of the tapeworm Echinococcus, which parasitizes even-toed ungulates or humans. Sheep are the primary intermediate hosts of Echinococcus. Effective and proactive prevention and control measures for Echinococcosis in sheep are crucial for halting the spread and progression of the disease. According to the "Action Plan for the Prevention and Control of Echinococcosis (2010-2015)," over 50 million livestock in my country are affected annually, resulting in direct economic losses exceeding 3 billion yuan. Currently, etiological, immunological, and ultrasound methods are the primary diagnostic methods for Echinococcosis in sheep. These methods, however, have drawbacks such as difficulty isolating and culturing the pathogen, difficulty standardizing between batches, and the risk of cross-reactivity. Ultrasound imaging requires the removal of wool from the test area, the need for two staff members to secure the sheep's body for the ultrasound technician to perform the examination, and the high cost of the equipment.

[0003] Brucellosis (abbreviated as brucellosis) is one of the most widespread and serious zoonotic diseases in the world, severely impacting the healthy development of livestock farming, national public health, and food safety. High-incidence areas of brucellosis in my country are primarily agricultural and pastoral areas in Inner Mongolia, Qinghai, Xinjiang, and Tibet. Healthy animals are primarily infected through contact with secretions and aborted fetuses from infected animals. Brucella infection in livestock severely impairs reproductive and production performance, with first-birth miscarriage rates as high as 50-80%, resulting in a 15-20% reduction in milk and meat production. Contact with infected animals and consumption of contaminated milk and meat are the primary routes of transmission to humans. Currently, the main diagnostic methods for brucellosis in sheep and goats include bacteriology, serology, and molecular biology. Isolation and culture of Brucella is the gold standard for diagnosis, but it has drawbacks such as time-consuming, difficult culture conditions, low detection rates, and potential for infection during laboratory operations. Molecular biological testing, on the other hand, requires expensive specialized diagnostic equipment and reagents, as well as high-level biosafety laboratories. Serology, as the most commonly used detection method for sheep pox screening, still has many shortcomings, such as the detection "window period", difficulty in standardization between product batches, and cross-reactions.

[0004] In view of this, the present invention proposes a classification model for sheep echinococcosis and sheep brucellosis based on serum Raman spectroscopy, an establishment method thereof, and a detection device. Summary of the Invention

[0005] The purpose of the present invention is to provide a method for establishing a classification model for sheep echinococcosis and sheep brucellosis based on serum Raman spectroscopy. Based on a machine learning algorithm and combined with Raman spectroscopy data, the method has great application potential for screening sheep infected with echinococcosis and sheep infected with brucellosis at one time.

[0006] In order to achieve the above objectives, the technical solutions adopted are:

[0007] A method for establishing a classification model for sheep echinococcosis and sheep brucellosis based on serum Raman spectroscopy comprises the following steps:

[0008] (1) Collect fresh blood from healthy sheep and diseased sheep, let it stand and then centrifuge it to extract serum to obtain serum samples;

[0009] (2) detecting the serum sample using a laser Raman spectrometer to obtain Raman spectrum data;

[0010] (3) performing preprocessing such as noise reduction, baseline correction, and normalization on the Raman spectral data to obtain preprocessed spectral data;

[0011] (4) Select a meaningful spectral range (400-1800cm) for the pre-processed spectral data -1 ) data were subjected to principal component analysis (PCA) dimensionality reduction, and a PCA-LDA or PCA-SVM classification model was established to obtain the classification model of sheep echinococcosis and sheep brucellosis.

[0012] Furthermore, in the step (1), the diseased sheep are sheep with echinococcosis and / or brucellosis.

[0013] Furthermore, serum samples from healthy sheep were collected from at least 244 cases;

[0014] The diseased sheep are Echinococcus granulosus-infected sheep, and serum samples from at least 231 diseased sheep are collected;

[0015] The diseased sheep are brucellosis-infected sheep, and serum samples of at least 106 diseased sheep are collected.

[0016] Furthermore, in step (2), the detection integration time is 3 seconds and the resolution is 6 cm -1 , laser power is 200mW, average scanning times is 3 times, in the range of 0-3500cm -1 Raman spectral data were acquired within the scanning range.

[0017] Furthermore, in the step (4), principal component analysis is used to reduce the dimension of the pre-processed data within the modeling spectrum range.

[0018] Furthermore, the modeling spectrum range is 400-1800cm -1 .

[0019] Furthermore, in step (4), an SVM classification model is established.

[0020] One object of the present invention is to provide a classification model for sheep echinococcosis and sheep brucellosis, which is obtained using the above-mentioned establishment method and has application potential in screening sheep infected with echinococcosis and sheep infected with brucellosis.

[0021] Another object of the present invention is to provide a detection device for sheep echinococcosis and sheep brucellosis based on serum Raman spectroscopy, which can be used for screening sheep echinococcosis and sheep brucellosis, and can realize the classification screening of sheep infected with echinococcosis and sheep infected with brucellosis at one time.

[0022] A device for detecting sheep echinococcosis and sheep brucellosis based on serum Raman spectroscopy, comprising: a laser, a Raman spectrometer, and a computing and analysis system;

[0023] The laser Raman spectrometer scans each sample to obtain serum Raman spectrum data, and transmits the obtained serum Raman spectrum data to the calculation and analysis system;

[0024] The computing and analysis system receives serum Raman spectrum data collected by the laser Raman spectrometer and performs pre-processing of the data such as noise reduction, baseline correction and normalization;

[0025] The computational analysis system uses principal component analysis (PCA) to reduce the dimensionality of the pre-processed serum Raman spectral data, takes the principal component with the largest cumulative variance contribution rate and statistical difference and inputs it into the support vector machine (SVM) and linear discriminant analysis (LDA) algorithms to establish a classification model, and randomly divides all samples into training sets and test sets, and finally outputs the classification results of sheep infected with echinococcosis granulosus, sheep infected with brucellosis, and healthy sheep.

[0026] Compared with the prior art, the present invention has the following beneficial effects:

[0027] This invention is the first to apply serum Raman spectroscopy to the detection of sheep infected with Echinococcosis granulosus and brucellosis. Currently, the diagnosis of Echinococcosis granulosus in sheep primarily relies on imaging (ultrasound, CT, etc.) and serological immunology. Imaging requires expensive ultrasound equipment, CT scanners, and specialized imaging technicians. The diagnostic process is also relatively cumbersome, and the diagnosis results require the judgment of a professional radiologist. Before the ultrasound imaging test, the wool at the test site must be removed, and two staff members must secure the sheep's body so that the ultrasound imaging physician can perform the test. Although sheep serum is relatively easy to collect, the serum immunology test kit for Echinococcosis granulosus in sheep has disadvantages such as long testing time, difficulty in standardizing between product batches, and cross-reactivity. The main diagnostic techniques for brucellosis in sheep include bacteriological testing, serological testing, and molecular testing. Bacterial isolation and culture is currently the gold standard for diagnosing brucellosis, but its disadvantages are the long bacterial culture cycle and the potential risk of laboratory-acquired brucellosis during the identification of growing bacteria. While the serological test, the red tiger plate agglutination test (RBPT), offers advantages such as speed and affordability, it also creates a "window period" during the early stages of illness when antibodies are not produced. Furthermore, it is susceptible to cross-reactions with other bacteria, resulting in false-positive results. The serum agglutination test (SAT) is often used for retrospective diagnosis, typically weeks or months after the acute phase of illness. False-negative results can occur when SAT titers do not change significantly in the early stages of illness, making the test performance dependent on the SAT titer. ELISA offers advantages such as high sensitivity and a short processing time, but also has disadvantages such as lack of standardization of reagents and the potential for false-positive results due to cross-reactions. Furthermore, sheep in pastoral areas suffer from both echinococcosis and brucellosis, but there is currently no comprehensive diagnostic technology that can distinguish between echinococcosis and brucellosis simultaneously, resulting in economic losses for the pastoral livestock industry and harm to human health.

[0028] The technology invented this time has the advantages of being fast, easy to operate and accurate. It can meet the shortcomings of the current conventional diagnostic technology for sheep echinococcosis and sheep brucellosis, and has great application potential for distinguishing sheep echinococcosis and sheep brucellosis in one go. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 The results of B-ultrasound examination of the liver of sheep diagnosed with Echinococcosis granulosus;

[0030] Figure 2 Diagram of the diagnosis of brucellosis by using the tiger red plate agglutination test and the sheep brucellosis tube agglutination test for serum collected from sheep that need to be tested for brucellosis;

[0031] Figure 3The device for detecting sheep echinococcosis and sheep brucellosis based on serum Raman spectroscopy used in the experiment;

[0032] Figure 4 This is a diagram showing the preprocessing process of denoising, baseline correction and normalization for the collected sheep serum Raman spectrum data;

[0033] Figure 5 The averaged Raman spectra of serum from sheep infected with Echinococcus granulosus and healthy sheep;

[0034] Figure 6 The scatter plot is the principal component score matrix of the serum Raman spectroscopy data of sheep infected with Echinococcus granulosus and healthy sheep;

[0035] Figure 7 Averaged Raman spectra of sera from brucellosis-infected and healthy sheep;

[0036] Figure 8 The scatter plot of the principal component score matrix of the serum Raman spectroscopy data of sheep infected with Brucellosis and healthy sheep is shown;

[0037] Figure 9 The Raman spectra of the serum of sheep infected with Echinococcosis granulosus, sheep infected with Brucellosis and healthy sheep are averaged.

[0038] Figure 10 Matrix scatter plot of the significant principal component scores in the serum Raman spectroscopy data of sheep infected with Echinococcosis granulosus, sheep infected with Brucellosis, and healthy sheep. DETAILED DESCRIPTION

[0039] To further illustrate the present invention's serum Raman spectroscopy-based classification model for echinococcosis and brucellosis in sheep, its establishment method, and detection device, and to achieve the intended purpose of the invention, the following describes in detail, in conjunction with preferred embodiments, a classification model for echinococcosis and brucellosis in sheep, its establishment method, and detection device, as well as its specific implementation, structure, features, and efficacy. In the following description, different "one embodiment" or "embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.

[0040] The following will further introduce in detail a classification model for sheep echinococcosis and sheep brucellosis based on serum Raman spectroscopy, its establishment method, and detection device in conjunction with specific embodiments of the present invention:

[0041] The present invention uses serum as a biological sample to diagnose sheep echinococcosis and sheep brucellosis using Raman spectroscopy technology. Experimental results show that the classification model established by the present invention has great application potential for one-time screening of sheep echinococcosis and sheep brucellosis.

[0042] The technical solution of the present invention is:

[0043] (I) Raman spectroscopy detection model for sheep echinococcosis serum based on machine learning algorithm. We first tested the serum of sheep infected with echinococcosis and healthy sheep. The detailed steps are as follows:

[0044] (1) Blood sampling: Select the sheep to be tested, perform classification diagnosis by B-ultrasound examination, and divide them into sheep infected with echinococcosis and healthy sheep. Then use a 10ml sheep blood collector (without anticoagulant) to collect blood from the left jugular vein of the sheep.

[0045] (2) Blood sample processing: The collected blood samples were allowed to stand at room temperature for 2-3 hours to allow the blood to completely coagulate. The samples were then centrifuged at 4000 rpm. The separated serum (the upper milky yellow supernatant) was aspirated with a pipette and transferred to a 2.5 ml EP tube. The tube was then stored at -80°C until use.

[0046] (3) Serum Raman spectroscopy data acquisition: The experiment used a serum Raman spectroscopy detection device for sheep infected with echinococcosis and healthy sheep, with a detection integration time of 3 seconds and a resolution of 6 cm -1 , laser power is 200mW, average scanning times is 3 times, in the range of 0-3500cm -1 The Raman spectrum data of the serum sample obtained in step (2) are respectively acquired within the scanning range.

[0047] (4) Serum Raman spectral data preprocessing: The spectral data obtained in step (3) were preprocessed using MATLAB R2021a software to perform noise reduction, baseline correction, and normalization on all serum Raman spectral data, and then converted into xls format for extraction.

[0048] (5) Data dimensionality reduction: MATLAB R2021a software was used to perform principal component analysis (PCA) in the selected modeling spectral range (400-1800 cm -1 ) to reduce the dimension of the preprocessed data and select the principal component with the largest cumulative variance contribution and statistical difference.

[0049] (6) Establishment and verification of the diagnostic model: As described in step (5), the principal components of the serum data of sheep infected with echinococcosis and healthy sheep were randomly divided into training sets and validation sets, and input into the LDA and SVM algorithms, respectively, to obtain the binary classification diagnostic results of sheep infected with echinococcosis and healthy sheep based on the PCA-LDA and PCA-SVM models.

[0050] (II) Raman spectroscopy detection model for sheep brucellosis serum based on machine learning algorithm. We tested the serum of sheep infected with brucellosis and healthy sheep. The detailed steps are as follows:

[0051] (1) Blood sampling: Select the sheep to be tested, use a 10ml sheep blood collector (without anticoagulant) to collect blood from the left jugular vein of the sheep, and then perform the sheep brucellosis tiger red plate agglutination test and the sheep brucellosis tube agglutination test to divide the sheep into brucellosis-infected sheep and healthy sheep.

[0052] (2) Blood sample processing: The collected blood samples were allowed to stand at room temperature for 2-3 hours to allow the blood to completely coagulate. The samples were then centrifuged at 4000 rpm. The separated serum (the upper milky yellow supernatant) was aspirated with a pipette and transferred to a 2.5 ml EP tube. The tube was then stored at -80°C until use.

[0053] (3) Serum spectral data acquisition: The experiment used a Raman spectroscopy device for serum detection of brucellosis sheep and healthy sheep, with a detection integration time of 3 seconds and a resolution of 6 cm -1 , laser power is 200mW, average scanning times is 3 times, in the range of 0-3500cm -1 The Raman spectrum data of the serum sample obtained in step (1) are respectively acquired within the scanning range.

[0054] (4) Serum Raman spectral data preprocessing: The spectral data obtained in step (3) were preprocessed using MATLAB R2021a software to perform noise reduction, baseline correction, and normalization on all serum Raman spectral data, and then converted into xls format for extraction.

[0055] (5) Data dimensionality reduction: MATLAB R2021a software was used to perform principal component analysis (PCA) in the selected modeling spectral range (400-1800 cm -1 ) to reduce the dimension of the preprocessed data and select the principal component with the largest cumulative variance contribution and statistical difference.

[0056] (6) Establishment and verification of the diagnostic model: According to the method described in step (5), the principal components of the serum data of Brucella-infected sheep and healthy sheep were randomly divided into training sets and validation sets, and input into the LDA and SVM algorithms to obtain the binary classification diagnostic results of Brucella-infected sheep and healthy sheep based on the PCA-LDA and PCA-SVM models, respectively.

[0057] (III) Classification model of Raman spectra of sera of sheep infected with echinococcosis and brucellosis based on machine learning algorithm. We tested the sera of sheep infected with echinococcosis, sheep infected with brucellosis, and healthy sheep. The detailed steps are as follows:

[0058] (1) Blood sampling: Select the sheep to be tested, use a 10ml sheep blood collector (without anticoagulant) to collect blood from the left jugular vein of the sheep, and then perform sheep B-ultrasound examination, sheep brucellosis tiger red plate agglutination test and sheep brucellosis tube agglutination test to divide the sheep into sheep infected with echinococcosis, sheep infected with brucellosis and healthy sheep.

[0059] (2) Blood sample processing: The collected blood samples were allowed to stand at room temperature for 2-3 hours to allow the blood to completely coagulate. The samples were then centrifuged at 4000 rpm. The separated serum (the upper milky yellow supernatant) was aspirated with a pipette and transferred to a 2.5 ml EP tube. The tube was then stored at -80°C until use.

[0060] (3) Serum spectral data acquisition: The experiment used a serum Raman spectroscopy detection device for sheep infected with echinococcosis, sheep infected with brucellosis, and healthy sheep, with a detection integration time of 3 seconds and a resolution of 6 cm -1 , laser power is 200mW, average scanning times is 3 times, in the range of 0-3500cm -1 The Raman spectrum data of the serum sample obtained in step (1) are respectively acquired within the scanning range.

[0061] (4) Serum Raman spectral data preprocessing: The spectral data obtained in step (3) were preprocessed using MATLAB R2021a software to perform noise reduction, baseline correction, and normalization on all serum Raman spectral data, and then converted into xls format for extraction.

[0062] (5) Data dimensionality reduction: MATLAB R2021a software was used to perform principal component analysis (PCA) in the selected modeling spectral range (400-1800 cm -1 ) to reduce the dimension of the preprocessed data and select the principal component with the largest cumulative variance contribution and statistical difference.

[0063] (6) Establishment and verification of the diagnostic model: According to the method described in step (5), the principal components of the serum data of sheep infected with Echinococcosis granulosus, sheep infected with Brucella and healthy sheep were randomly divided into training sets and validation sets and input into the LDA and SVM algorithms to obtain the three-category diagnostic results of sheep infected with Echinococcosis granulosus, sheep infected with Brucella and healthy sheep based on the PCA-LDA and PCA-SVM models, respectively.

[0064] Example 1.

[0065] The classification model of serum Raman spectra of sheep echinococcosis and sheep brucellosis based on machine learning algorithm was established. The specific steps are as follows:

[0066] (1) Sample collection: First, collect samples with clear diagnosis through B-ultrasound test ( Figure 1 Blood samples were collected from 231 sheep infected with Echinococcosis granulosus and 244 healthy sheep (shown). Blood samples were also collected from 300 sheep in flocks that needed to be tested for brucellosis.

[0067] (2) Obtaining serum samples: The collected blood samples were allowed to stand at room temperature for 2-3 hours to allow the blood to completely coagulate, and then placed in a centrifuge for centrifugation at 4000 rpm. The separated serum (the upper milky yellow supernatant) was aspirated with a pipette and transferred to a 2.5 ml EP tube. The serum collected from the sheep to be tested for brucellosis was subjected to the brucellosis red plate agglutination test and the brucellosis tube agglutination test ( Figure 2 As shown in Figure 3, 106 sheep infected with Brucellosis and 106 healthy sheep were screened out.

[0068] (3) Serum Raman spectroscopy data acquisition: The experiment used a serum Raman spectroscopy detection device (schematic diagram) of sheep infected with echinococcosis, sheep infected with brucellosis, and healthy sheep. Figure 3 As shown), set the detection integration time to 3 seconds and the resolution to 6 cm -1 , laser power is 200mW, average scanning times is 3 times, in the range of 0-3500cm -1 The Raman spectrum data of the serum sample obtained in step (2) are respectively acquired within the scanning range.

[0069] (4) Serum Raman spectroscopy data preprocessing: MATLAB R2021a software was used to perform denoising, baseline correction, and normalization on all serum Raman spectroscopy data ( Figure 4 shown).

[0070] (5) Data dimensionality reduction: MATLAB R2021a software was used to perform principal component analysis (PCA) in the range of 400-1800 cm -1 The preprocessed data were subjected to dimensionality reduction within the spectral range, and the principal component with the largest cumulative variance contribution and statistical difference was selected.

[0071] (6) Establishment and verification of diagnostic model: According to the method described in step (5), the principal components of serum Raman spectra of sheep infected with echinococcosis, sheep infected with brucellosis and healthy sheep were randomly divided into training set and validation set and input into LDA and SVM algorithms, respectively. Three classification diagnostic models and diagnostic results were obtained for serum Raman spectra of sheep infected with echinococcosis and healthy sheep, sheep infected with brucellosis and healthy sheep, and sheep infected with echinococcosis and healthy sheep based on PCA-LDA and PCA-SVM models. The specific results are as follows:

[0072] Classification results of sheep infected with Echinococcosis granulosus and healthy sheep: The two groups of serum samples were randomly divided into training set and validation set, and the results were analyzed in the range of 400-1800 cm -1 In the wavelength range ( Figure 5 As shown), we first extract the statistically significant principal components ( Figure 6 The meaningful principal components were input into the LDA and SVM models, and the classification results are shown in Table 1.

[0073] Table 1

[0074]

[0075] Classification results of sheep infected with Brucellosis and healthy sheep: The two groups of serum samples were randomly divided into training set and validation set, and the -1 In the wavelength range ( Figure 7 As shown), we first extract the statistically significant principal components ( Figure 8 Then the meaningful principal components are input into LDA and SVM models, and the classification results are shown in Table 2.

[0076] Table 2

[0077]

[0078]

[0079] Classification results of sheep infected with echinococcosis, sheep infected with brucellosis and healthy sheep: All serum samples were randomly divided into training set and validation set, and the serum samples were collected at 400-1800 cm -1 In the wavelength range ( Figure 9 As shown), we first extract the statistically significant principal components ( Figure 10 Then the statistically significant principal components were input into the LDA and SVM models, and the classification results are shown in Table 3.

[0080] Table 3

[0081]

[0082] This study is the first to apply serum Raman spectroscopy to the detection of echinococcosis and brucellosis in sheep. Based on a machine learning algorithm and Raman spectroscopy, the authors designed PCA-LDA and PCA-SVM classification models for echinococcosis and brucellosis in sheep. Using validation set accuracy as the evaluation basis, experimental results demonstrated that both the PCA-LDA and PCA-SVM models were highly effective in screening and diagnosing sheep infected with echinococcosis and brucellosis, with the SVM model showing superior performance.

[0083] The above is merely a preferred embodiment of the present invention and does not constitute any form of limitation to the embodiments of the present invention. Any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the embodiments of the present invention are still within the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for establishing a classification model for sheep echinococcosis and sheep brucellosis based on serum Raman spectroscopy, characterized in that: The following steps are involved: (1) Collecting fresh blood from healthy sheep and diseased sheep, centrifuging the blood after standing, extracting serum, and obtaining serum samples; the diseased sheep are sheep with echinococcosis and sheep with brucellosis; (2) The serum sample is tested using a laser Raman spectrometer to obtain Raman spectrum data; the detection integration time is 3 seconds and the resolution is 6 cm -1 , laser power is 200mW, average scanning times is 3 times, in the range of 0-3500cm -1 Acquire Raman spectrum data within the scanning range; (3) performing preprocessing of the Raman spectral data by noise reduction, baseline correction, and normalization to obtain preprocessed spectral data; (4) The pre-processed spectral data in the range of 400-1800 cm -1 After PCA dimensionality reduction was performed on the modeled spectral range, an LDA classification model was established to obtain the classification models of sheep echinococcosis and sheep brucellosis.

2. The establishment method according to claim 1, characterized in that At least 244 serum samples were collected from healthy sheep; The diseased sheep are Echinococcus granulosus-infected sheep, and serum samples from at least 231 diseased sheep are collected; The diseased sheep are brucellosis-infected sheep, and serum samples of at least 106 diseased sheep are collected.

3. A classification model for sheep echinococcosis and sheep brucellosis based on serum Raman spectroscopy, characterized in that: The method is obtained by using the method according to any one of claims 1 to 2.

4. A detection device for sheep echinococcosis and sheep brucellosis based on serum Raman spectroscopy, characterized in that: The detection device adopts the establishment method according to any one of claims 1-2; the detection device comprises: a laser, a Raman spectrometer, and a computing and analysis system; The laser Raman spectrometer scans each sample to obtain serum Raman spectrum data, and transmits the obtained serum Raman spectrum data to the calculation and analysis system; The computing and analysis system receives serum Raman spectrum data collected by the laser Raman spectrometer and performs noise reduction, baseline correction and normalization preprocessing on the data; The computational analysis system uses the PCA algorithm to reduce the dimension of the preprocessed serum Raman spectral data, takes the principal component with the largest cumulative variance contribution rate and statistical difference and inputs it into the LDA algorithm to establish a classification model, and randomly divides all samples into training sets and test sets, and finally outputs the classification results of the diseased group and the healthy group.

Images