A method and device for detecting methylation of MGMT promoter in glioblastoma

Abstract

The present application discloses a method and device for detecting methylation of MGMT promoter of glioblastoma, which relates to the field of Raman spectrum detection. The method comprises: performing Raman spectrum scanning on a glioma tissue sample to be tested to obtain Raman spectrum data; preprocessing the Raman spectrum data to obtain normalized Raman spectrum data; extracting the intensity value at the characteristic displacement from the normalized Raman spectrum data; inputting the intensity value at the characteristic displacement into a classification model, calculating and obtaining a classification prediction value, which is used to assist in determining the methylation state of MGMT promoter of the glioma tissue sample to be tested. The present application can realize low-cost, non-destructive and accurate detection of methylation state of MGMT promoter of glioblastoma.

Description

Technical Field

[0001] The present application relates to the field of Raman spectroscopy detection, and in particular to a method and device for detecting methylation of the glioblastoma MGMT promoter. Background Art

[0002] Gliomas are the most common tumors of the central nervous system, accounting for about 70% of all primary brain malignancies. They are one of the major brain diseases with high mortality and disability rates. The median survival time of patients with highly malignant glioblastomas is only 12 to 14 months. There is no satisfactory treatment for glioblastoma so far, and patients' resistance to chemotherapy drugs is one of the important reasons for the poor treatment effect.

[0003] The methylation status of the O6-methylguanine-DNA methyltransferase (MGMT) promoter is significantly correlated with chemotherapeutic resistance in brain gliomas. MGMT promoter methylation is not only an important basis for evaluating whether glioblastoma patients are sensitive to alkylating agents and thus formulating personalized and precise treatment plans, but also a reference indicator for evaluating prognosis and distinguishing recurrence from pseudoprogression. Currently, the detection of MGMT promoter methylation relies on genetic analysis after surgical removal of pathological tissue, which has the disadvantages of high cost, invasiveness, and inaccurate results. Patients generally find it difficult to accept the high cost of testing. Summary of the invention

[0004] The purpose of the present application is to provide a method and device for detecting methylation of the glioblastoma MGMT promoter, which can realize low-cost, non-destructive and accurate detection of the methylation status of the glioblastoma MGMT promoter.

[0005] To achieve the above objectives, this application provides the following solutions.

[0006] In a first aspect, the present application provides a method for detecting methylation of the MGMT promoter in glioblastoma, comprising: performing Raman spectroscopy scanning on a glioma tissue sample to be tested to obtain Raman spectroscopy data; preprocessing the Raman spectroscopy data to obtain normalized Raman spectroscopy data; extracting the intensity value at the characteristic displacement from the normalized Raman spectroscopy data; the characteristic displacement includes a Raman displacement of 1518 cm -1 、1664cm -1 、2895cm -1 、2945cm -1 and 3176cm -1; The intensity value at the characteristic displacement is input into the classification model to calculate the classification prediction value; the classification model characterizes the relationship between the classification prediction value and the intensity value at the characteristic displacement; the classification prediction value is used to assist in determining the MGMT promoter methylation status of the glioma tissue sample to be tested; the MGMT promoter methylation status includes MGMT promoter methylation positive and MGMT promoter methylation negative.

[0007] Optionally, the Raman spectrum data is preprocessed to obtain normalized Raman spectrum data, including: removing fluorescence background from the Raman spectrum data to obtain Raman spectrum data with fluorescence background removed; smoothing and denoising the Raman spectrum data with fluorescence background removed to obtain smoothed and denoised Raman spectrum data; and normalizing the smoothed and denoised Raman spectrum data to obtain normalized Raman spectrum data.

[0008] Optionally, the classification model is determined based on a logistic regression algorithm and a maximum relevance minimum redundancy algorithm; the determination process of the classification model includes: obtaining a number of Raman spectrum samples annotated with the methylation status of the MGMT promoter, and preprocessing and extracting characteristic spectrum information to construct a sample data set; the sample data set includes characteristic spectrum information and MGMT promoter methylation status corresponding to a number of normalized Raman spectrum samples; the characteristic spectrum information includes intensity values ​​at a number of Raman shifts; the MGMT promoter methylation status includes 1 and 0, wherein 1 indicates that the MGMT promoter is positively methylated and 0 indicates that the MGMT promoter is negatively methylated; a linear regression model is constructed with the MGMT promoter methylation status as the target variable and the characteristic spectrum information as the characteristic variable; according to the sample data set, the maximum relevance minimum redundancy algorithm is used to screen the characteristic shifts, and the logistic regression algorithm is used to train the linear regression model to obtain a classification model.

[0009] Optionally, the classification model is expressed as: ;s = -0.2675 + (-1.9564 × I 1518 ) + 0.7958 × I 1664 + 0.5511× I 2895 + 1.3966 × I 2945 + (-1.4758× I 3176 ), where s represents the output value of the regression algorithm; σ( ) represents the sigmoid function; e represents the natural constant; P represents the classification prediction value. When P ≥ 0.5, it is determined to be a glioblastoma with positive MGMT promoter methylation, and when P < 0.5, it is determined to be a glioblastoma with negative MGMT promoter methylation; I 1518 Indicates that the Raman shift is 1518 cm -1The intensity value at I 1664 Indicates the Raman shift of 1664 cm -1 The intensity value at I 2895 Indicated at Raman shift 2895cm -1 The intensity value at I 2945 Indicates the Raman shift of 2945 cm -1 The intensity value at I 3176 Indicates that the Raman shift is 3176cm -1 The regression coefficient before the intensity value at each characteristic displacement represents the importance of the characteristic displacement.

[0010] Optionally, the Raman spectrum data is processed to remove the fluorescence background to obtain the Raman spectrum data with the fluorescence background removed, including: using an adaptive iterative reweighted penalized least squares method to process the Raman spectrum data to remove the fluorescence background to obtain the Raman spectrum data with the fluorescence background removed.

[0011] Optionally, performing smoothing and denoising processing on the Raman spectrum data from which the fluorescence background is removed to obtain smoothed and denoised Raman spectrum data includes: performing smoothing and denoising processing on the Raman spectrum data from which the fluorescence background is removed using a Savitzky-Golay smoothing algorithm to obtain smoothed and denoised Raman spectrum data.

[0012] Optionally, normalizing the smoothed and denoised Raman spectrum data to obtain normalized Raman spectrum data includes: normalizing the smoothed and denoised Raman spectrum data using a maximum-minimum normalization algorithm to obtain normalized Raman spectrum data.

[0013] In a second aspect, the present application provides a glioblastoma MGMT promoter methylation detection device, which adopts the above-mentioned glioblastoma MGMT promoter methylation detection method, including: a laser, used to emit laser to the glioma tissue sample to be tested; a spectrometer, used to collect Raman light scattered by the glioma tissue sample to be tested to obtain Raman spectrum data; a computer, connected to the spectrometer, used to preprocess the Raman spectrum data to obtain normalized Raman spectrum data, extract the intensity value at the characteristic displacement from the normalized Raman spectrum data, and input the intensity value at the characteristic displacement into a classification model to calculate the classification prediction value.

[0014] Optionally, the glioblastoma MGMT promoter methylation detection device also includes: an optical fiber probe, connected to the laser and the spectrometer respectively, for transmitting the laser emitted by the laser to the surface of the glioma tissue sample to be tested, and transmitting the Raman light scattered by the surface of the glioma tissue sample to be tested to the spectrometer.

[0015] Optionally, the glioblastoma MGMT promoter methylation detection device further includes: a display connected to the computer, for displaying the classification prediction value and the auxiliary determination result; the auxiliary determination result is determined based on the classification prediction value.

[0016] Optionally, the glioblastoma MGMT promoter methylation detection device further comprises: a detection platform for placing the glioma tissue sample to be detected, and adjusting the position and height of the glioma tissue sample to be detected.

[0017] Optionally, the excitation wavelength of the laser is 532 nm and the integration time is 3 seconds.

[0018] According to the specific embodiments provided by the present application, the present application discloses the following technical effects: the present application provides a method and device for detecting methylation of the MGMT promoter of glioblastoma, firstly, a Raman spectrum scan is performed on the glioma tissue sample to be tested to obtain Raman spectrum data, then the Raman spectrum data is preprocessed and feature extracted to obtain the intensity value at the characteristic displacement, and finally the intensity value at the characteristic displacement is input into a pre-trained classification model to calculate the classification prediction value to assist in determining the methylation state of the MGMT promoter of the glioma tissue sample to be tested. Compared with the prior art, the present application utilizes Raman spectroscopy technology, performs Raman spectroscopy analysis on the glioma tissue sample to be tested, and combines the trained classification model to achieve low-cost, non-destructive, and accurate MGMT promoter methylation state detection, which can partially replace the high-cost tumor gene detection and analysis function. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0020] Figure 1 A flow chart of the method for detecting methylation of the glioblastoma MGMT promoter provided in this application.

[0021] Figure 2 Schematic diagram of the average methylation and unmethylation spectra of the glioblastoma MGMT promoter provided in this application.

[0022] Figure 3 This is a graph showing the prediction performance of the MGMT promoter methylation status of the classification model provided in this application.

[0023] Figure 4This is a structural diagram of the glioblastoma MGMT promoter methylation detection device provided in this application.

[0024] Reference numerals: laser-1, spectrometer-2, computer-3, display-4, detection platform-5. DETAILED DESCRIPTION

[0025] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0026] The present application provides a method and device for detecting methylation of the MGMT promoter in glioblastoma. The method uses Raman spectroscopy technology to identify the methylation state of the MGMT gene promoter by performing Raman spectroscopy analysis on the glioma tissue sample to be tested, thereby realizing rapid detection of glioblastoma with positive MGMT promoter methylation and guiding the formulation of treatment plans.

[0027] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0028] like Figure 1 As shown, the method for detecting methylation of the glioblastoma MGMT promoter provided in the present application includes the following steps S1 to S4.

[0029] Step S1: Perform Raman spectrum scanning on the glioma tissue sample to be tested to obtain Raman spectrum data.

[0030] Among them, the glioma tissue samples to be tested can be collected directly during the operation without any slicing processing. Tissue fragments with a long diameter of about 2 to 20 mm can be taken and placed on a glass slide.

[0031] For example, a Raman spectrum excitation device (such as a laser with an excitation wavelength of 532 nm) is used to perform Raman spectrum scanning on the glioma tissue sample to be tested, and a spectrometer is used to collect Raman spectrum data. The integration time of the Raman spectrum scanning is controlled within 3 seconds.

[0032] Step S2: preprocessing the Raman spectrum data to obtain normalized Raman spectrum data.

[0033] Exemplarily, the Raman spectrum data is first processed to remove the fluorescence background to obtain the Raman spectrum data with the fluorescence background removed; secondly, the Raman spectrum data with the fluorescence background removed is smoothed and denoised to obtain the smoothed and denoised Raman spectrum data; finally, the smoothed and denoised Raman spectrum data is normalized to obtain the normalized Raman spectrum data.

[0034] Preferably, the fluorescence background removal process uses an adaptive iterative re-weighted penalized least squares (airPLS) method.

[0035] The airPLS algorithm removes the background by iteratively adjusting the weight of the signal. In each iteration, it calculates the difference between the signal and the background and updates the weight so that the background better fits the data. The airPLS algorithm includes initialization and iteration processes.

[0036] Initialization: Set the initial weights and input parameters. For example, the initial weight w=1, and the input parameters include the signal g, the penalty parameter μ (4 in this application), and the maximum number of iterations itermax (15 in this application).

[0037] Iterative process: For each iteration, background smoothing, residual calculation, weight update and convergence check are performed in sequence.

[0038] Background smoothing: Using the Whittaker smoothing algorithm, solve the background z = WhittakerSmooth(g,w,μ); where WhittakerSmooth( ) is the function of the Whittaker smoothing algorithm, which is achieved by minimizing the following objective function: min ||w·(gz)|| 2 +μ·||Δz|| 2 ; where Δ represents a difference operator applied to the background z to increase smoothness.

[0039] Calculate the residual: Calculate the difference between the signal and the background d k =g k -z k ; where k represents the index of the data point, g k represents the kth data point signal, z k represents the background of the kth data point, d k Represents the difference between the signal of the kth data point and the background, that is, the residual value.

[0040] Update weights: For the part where the signal is higher than or equal to the background (i.e., d k ≥0), set the corresponding weight to 0 to ignore these data points, that is, ; For the part where the signal is lower than the background (i.e., d k <0), update the weights according to the residual value: , where w k represents the weight of the kth data point, and dssn represents the sum of the absolute values ​​of the negative residuals, that is, , t represents the number of iterations, and exp() represents the exponential function with the natural constant e as the base.

[0041] Check convergence: If the absolute sum of negative residuals dssn is small enough or the number of iterations reaches the upper limit, exit the iteration.

[0042] Preferably, the smoothing and noise reduction process adopts the Savitzky-Golay smoothing algorithm, and the Python function name is savgol_filter.

[0043] The Savitzky-Golay smoothing algorithm is a filtering method used to smooth and differentiate digital signals. It achieves smoothing by fitting a polynomial to the data in a local window. The specific Python function is: scipy.signal.savgol_filter(a,window_length,polyorder).

[0044] Where a is the signal to be smoothed. window_length is the window width, which is an odd number and cannot exceed the length of a, len(a). The larger the value, the more obvious the smoothing effect, and the smaller the value, the closer to the original curve; the window width used in this application is 7. polyorder is the order of the polynomial fit. The smaller the value, the more obvious the smoothing effect, and the larger the value, the closer to the original curve; the order used in this application is 2.

[0045] The formula of the Savitzky-Golay smoothing algorithm can be expressed as: ; Among them, b u is the smoothed data point, a u is the original data point, a u+v for a u The data point after moving v units, m is half the window width (that is, the window width is 2m+1), c v are the coefficients of the polynomial fit.

[0046] Preferably, the normalization process adopts the maximum-minimum normalization algorithm. The formula is: x'=(x-min(x)) / (max(x)-min(x)); wherein x' is the normalized Raman spectrum data, x is the smoothed and denoised Raman spectrum data, min(x) is the minimum value of the smoothed and denoised Raman spectrum data, and max(x) is the maximum value of the smoothed and denoised Raman spectrum data.

[0047] Step S3: Extract the intensity value at the characteristic shift from the normalized Raman spectrum data. The characteristic shift includes the Raman shift of 1518 cm -1 、1664cm -1 、2895cm -1 、2945cm -1 and 3176cm -1 .

[0048] Step S4: Input the intensity value at the characteristic displacement into the classification model, calculate the classification prediction value, and use it to assist in determining the MGMT promoter methylation status of the glioma tissue sample to be tested. The MGMT promoter methylation status includes MGMT promoter methylation positive and MGMT promoter methylation negative. Among them, the classification model characterizes the relationship between the classification prediction value and the intensity value at the characteristic displacement. The classification model is determined using a machine learning or pattern recognition algorithm.

[0049] Preferably, the present application determines a classification model based on a logistic regression (LR) algorithm and a maximal relevance and minimal redundancy (mRMR) algorithm, and the determination process includes: obtaining a number of Raman spectrum samples annotated with the methylation status of the MGMT promoter, and preprocessing and extracting characteristic spectrum information to construct a sample data set; the sample data set includes characteristic spectrum information and MGMT promoter methylation status corresponding to a number of normalized Raman spectrum samples; the characteristic spectrum information includes intensity values ​​at a number of Raman shifts; the MGMT promoter methylation status includes 1 and 0, wherein 1 indicates that the MGMT promoter is positively methylated (i.e., the MGMT promoter is methylated), and 0 indicates that the MGMT promoter is negatively methylated (i.e., the MGMT promoter is not methylated); a linear regression model is constructed with the methylation status of the MGMT promoter as the target variable and the characteristic spectrum information as the characteristic variable; according to the sample data set, the mRMR algorithm is used to screen the characteristic shifts, and the LR algorithm is used to train the linear regression model to obtain a classification model.

[0050] Among them, the mRMR algorithm aims to select features from the sample data set that are most relevant to the target variable and have the least redundancy with each other. Its objective function is usually expressed as: , where S is the selected feature subset; I(f i ,c) is the feature f i Mutual information with the target variable c (representing correlation); I(f i ,f j ) is the feature f i With feature f jThe mutual information between (representing redundancy); max() represents the maximum value. This application is based on the Python pymrmr library and selects 5 features that are most relevant to the target variable and have the least redundancy, including the Raman shift of 1518cm -1 、1664cm -1 、2895cm -1 、2945cm -1 and 3176cm -1 , which facilitates the identification of MGMT promoter methylation in gliomas.

[0051] In this example, based on a total of 443 spectral data from 104 patients retrospectively included (including 178 MGMT promoter methylation spectra and 265 MGMT promoter unmethylation spectra, the average spectrum is as follows Figure 2 The samples were randomly divided into a training group and a validation group at a ratio of 8:2 for training and validation. The mRMR algorithm and the LR algorithm were used to generate a classification model to predict the methylation status of the MGMT promoter.

[0052] The trained classification model has excellent performance on the validation set, with the area under the receiver operating characteristic curve (ROC) reaching up to 86.4% and the sensitivity reaching 82.4%. Figure 3 shown.

[0053] In the process of spectral feature screening using the mRMR algorithm, the possible main characteristic peaks involved are as follows.

[0054] 1518cm -1 : This peak position is the characteristic peak position of β-carotene.

[0055] 1664cm -1 :This peak position often reflects the vibration of peptide bonds, namely the stretching vibration of C=O and NH, which can provide information about protein structure and conformation.

[0056] 2895cm -1 :The peak position in this area represents the Raman characteristic peak position of protein content, which is caused by the stretching vibration of the CH bond in the methyl group.

[0057] 2945cm -1 :This peak position usually represents the characteristic peak of methyl bond CH3, reflecting the lipid content in the tissue.

[0058] 3176cm -1:The peak position in this region is usually associated with the vibration of Amide B in the protein, involving the stretching of free NH bonds, and represents the characteristics of the protein secondary structure.

[0059] Specifically, the classification model is expressed as: ;s = -0.2675 + (-1.9564 ×I 1518 ) + 0.7958 × I 1664 + 0.5511× I 2895 + 1.3966 × I 2945 + (-1.4758× I 3176 ).

[0060] Where s represents the output value of the regression algorithm; σ( ) represents the sigmoid function; e represents the natural constant; P represents the classification prediction value, that is, the classification prediction probability. When P ≥ 0.5, it is judged as a glioblastoma with positive MGMT promoter methylation, and when P < 0.5, it is judged as a glioblastoma with negative MGMT promoter methylation; I 1518 Indicates that the Raman shift is 1518 cm -1 The intensity value at I 1664 Indicates the Raman shift of 1664 cm -1 The intensity value at I 2895 Indicated at Raman shift 2895cm -1 The intensity value at I 2945 Indicates the Raman shift of 2945 cm -1 The intensity value at I 3176 Indicates that the Raman shift is 3176cm -1 The regression coefficient before the intensity value at each characteristic displacement represents the importance of the characteristic displacement.

[0061] like Figure 4 As shown, the glioblastoma MGMT promoter methylation detection device provided by the present application includes: a laser 1, a spectrometer 2 and a computer 3. Among them, the laser 1 is used to emit laser (i.e., 532nm excitation light) to the glioma tissue sample to be tested; the spectrometer 2 is used to collect Raman light (i.e., scattered light) scattered by the glioma tissue sample to be tested to obtain Raman spectrum data; the computer 3 is connected to the spectrometer 2, and is used to pre-process the Raman spectrum data to obtain normalized Raman spectrum data, extract the intensity value at the characteristic displacement from the normalized Raman spectrum data, and input the intensity value at the characteristic displacement into the classification model to calculate the classification prediction value.

[0062] Preferably, the excitation wavelength of the laser 1 is 532 nm, and the integration time is 3 seconds.

[0063] Furthermore, the above-mentioned glioblastoma MGMT promoter methylation detection device also includes: an optical fiber probe. The optical fiber probe is connected to the laser 1 and the spectrometer 2 respectively, and is used to transmit the laser emitted by the laser 1 to the surface of the glioma tissue sample to be tested, and transmit the Raman light scattered by the surface of the glioma tissue sample to be tested to the spectrometer 2.

[0064] Furthermore, the above-mentioned glioblastoma MGMT promoter methylation detection device further comprises: a display 4. The display 4 is connected to the computer 3 and is used to display the classification prediction value and the auxiliary determination result; the auxiliary determination result is determined based on the classification prediction value.

[0065] Furthermore, the above-mentioned glioblastoma MGMT promoter methylation detection device further comprises: a detection platform 5. The detection platform 5 is used to place the glioma tissue sample to be detected, and to adjust the position and height of the glioma tissue sample to be detected.

[0066] The glioblastoma MGMT promoter methylation detection method and device provided in the present application have the following advantages.

[0067] 1. Fast: Using resonance Raman spectroscopy technology, the detection time is short and the results are reported quickly.

[0068] 2. Accurate: Utilize characteristic spectral information for analysis to accurately identify the methylation status of the MGMT promoter.

[0069] 3. Non-destructive: No additional processing of tissue samples is required, maintaining the integrity of the samples.

[0070] 4. Simple: easy to operate, no need for complicated experimental conditions and professional skills.

[0071] In summary, the present application provides a method and device for detecting methylation of the MGMT promoter in glioblastoma, which has the advantages of being rapid, accurate, and non-destructive, and is suitable for clinical medicine and scientific research.

[0072] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0073] This article uses specific examples to illustrate the principles and implementation methods of this application. The description of the above embodiments is only used to help understand the method and core ideas of this application. At the same time, for those skilled in the art, according to the ideas of this application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.

Claims

1. A method for detecting methylation of the MGMT promoter in glioblastoma, characterized in that: include: Performing Raman spectrum scanning on the glioma tissue sample to be tested to obtain Raman spectrum data; Preprocessing the Raman spectrum data to obtain normalized Raman spectrum data; Extract the intensity value at the characteristic shift from the normalized Raman spectrum data; the characteristic shift includes the Raman shift 1518cm -1 、1664cm -1 、2895cm -1 、2945cm -1 and 3176cm -1 ; Inputting the intensity value at the characteristic displacement into the classification model to calculate the classification prediction value; The classification model characterizes the relationship between the classification prediction value and the intensity value at the characteristic displacement; the classification prediction value is used to assist in determining the MGMT promoter methylation state of the glioma tissue sample to be tested; the MGMT promoter methylation state includes MGMT promoter methylation positive and MGMT promoter methylation negative; The expression of the classification model is: ； s = -0.2675 + (-1.9564 × I 1518 ) + 0.7958 × I 1664 + 0.5511× I 2895 + 1.3966× I 2945 + (-1.4758× I 3176 ); Where s represents the output value of the regression algorithm; σ( ) represents the sigmoid function; e represents the natural constant; P represents the classification prediction value. When P ≥ 0.5, it is judged as a glioblastoma with positive MGMT promoter methylation; when P < 0.5, it is judged as a glioblastoma with negative MGMT promoter methylation; I 1518 Indicates that the Raman shift is 1518 cm -1 The intensity value at I 1664 Indicates the Raman shift of 1664 cm -1 The intensity value at I 2895 Indicated at Raman shift 2895cm -1 The intensity value at I 2945 Indicates the Raman shift of 2945 cm -1 The intensity value at I 3176 Indicates that the Raman shift is 3176cm -1 The regression coefficient before the intensity value at each characteristic displacement represents the importance of the characteristic displacement.

2. The method for detecting methylation of the glioblastoma MGMT promoter according to claim 1, characterized in that: Preprocessing the Raman spectrum data to obtain normalized Raman spectrum data includes: Performing fluorescence background removal processing on the Raman spectrum data to obtain Raman spectrum data with fluorescence background removed; Performing smoothing and noise reduction processing on the Raman spectrum data after the fluorescence background is removed to obtain smoothed and noise-reduced Raman spectrum data; The smoothed and denoised Raman spectrum data is normalized to obtain normalized Raman spectrum data.

3. The method for detecting methylation of the glioblastoma MGMT promoter according to claim 1, characterized in that: The classification model is determined based on a logistic regression algorithm and a maximum relevance minimum redundancy algorithm; the determination process of the classification model includes: Acquire several Raman spectrum samples annotated with MGMT promoter methylation status, perform preprocessing and extract characteristic spectrum information, and construct a sample data set; the sample data set includes characteristic spectrum information and MGMT promoter methylation status corresponding to several normalized Raman spectrum samples; the characteristic spectrum information includes intensity values ​​at several Raman shifts; the MGMT promoter methylation status includes 1 and 0, wherein 1 indicates that the MGMT promoter is positively methylated, and 0 indicates that the MGMT promoter is negatively methylated; A linear regression model is constructed using the methylation state of the MGMT promoter as a target variable and the characteristic spectrum information as a characteristic variable; According to the sample data set, the maximum relevance minimum redundancy algorithm is used to screen the feature displacement, and the linear regression model is trained using a logistic regression algorithm to obtain a classification model.

4. The method for detecting methylation of the glioblastoma MGMT promoter according to claim 2, characterized in that: Adopting adaptive iterative reweighted penalty least squares method to remove fluorescence background from the Raman spectrum data; The Savitzky-Golay smoothing algorithm is used to perform smoothing and noise reduction processing on the Raman spectrum data after removing the fluorescence background; and the maximum and minimum normalization algorithm is used to perform normalization processing on the smoothed and noise-reduced Raman spectrum data.

5. A glioblastoma MGMT promoter methylation detection device, using the glioblastoma MGMT promoter methylation detection method according to claim 1, characterized in that: include: A laser, used for emitting laser light to the glioma tissue sample to be tested; A spectrometer, used to collect Raman light scattered by the glioma tissue sample to be tested to obtain Raman spectrum data; A computer is connected to the spectrometer and is used to preprocess the Raman spectrum data to obtain normalized Raman spectrum data, extract intensity values ​​at characteristic displacements from the normalized Raman spectrum data, and input the intensity values ​​at characteristic displacements into a classification model to calculate a classification prediction value.

6. The glioblastoma MGMT promoter methylation detection device according to claim 5, characterized in that: Also includes: The optical fiber probe is connected to the laser and the spectrometer respectively, and is used to transmit the laser light emitted by the laser to the surface of the glioma tissue sample to be tested, and to transmit the Raman light scattered by the surface of the glioma tissue sample to be tested to the spectrometer.

7. The glioblastoma MGMT promoter methylation detection device according to claim 5, characterized in that: Also includes: A display, connected to the computer, for displaying the classification prediction value and the auxiliary determination result; The auxiliary determination result is determined based on the classification prediction value.

8. The glioblastoma MGMT promoter methylation detection device according to claim 5, characterized in that: Also includes: The detection platform is used to place the glioma tissue sample to be detected and adjust the position and height of the glioma tissue sample to be detected.

9. The glioblastoma MGMT promoter methylation detection device according to claim 5, characterized in that: The excitation wavelength of the laser is 532 nm, and the integration time is 3 seconds.

Images