A method and device for detecting IDH mutation in glioma based on Raman spectroscopy

Abstract

The present application discloses a method and device for detecting IDH mutations in gliomas based on Raman spectroscopy, and relates to the field of Raman spectroscopy detection. The method comprises: 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; inputting the intensity value at the characteristic displacement into a classification model to calculate a classification prediction value; the classification model is a multivariate linear equation of the classification prediction value and the intensity value at the characteristic displacement; the classification prediction value is used to assist in determining whether the glioma tissue sample to be tested is an IDH mutant glioma. The present application can realize rapid and accurate detection of the IDH mutation status of gliomas.

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 IDH mutations in gliomas based on Raman spectroscopy. Background Art

[0002] Glioma is a common central nervous system tumor, and accurate molecular testing is crucial for guiding treatment and prognosis assessment. Among them, isocitrate dehydrogenase (IDH) gene mutation is one of the common molecular variations in glioma, and is closely related to the occurrence, development and prognosis of glioma. IDH gene mutation is an active molecular change in glioma, and its clinical relevance is mainly reflected in: compared with IDH wild-type, IDH mutant glioma patients have a more favorable prognosis and enhanced sensitivity to chemotherapy and radiotherapy. Therefore, rapid detection of IDH mutation status in glioma is crucial for guiding surgical strategies and postoperative adjuvant therapy.

[0003] At present, conventional IDH mutation detection methods require a long time and complex experimental operations, and lack efficient, simple and rapid detection methods. Traditional intraoperative pathological detection methods include intraoperative frozen sections, rapid immunohistochemistry, intraoperative cytology, etc. Although the accuracy of these pathological diagnosis methods is relatively high, they are time-consuming and not conducive to the entire surgical process. For example, intraoperative frozen sections are the most widely used method and take about 30 to 40 minutes. Summary of the invention

[0004] The purpose of this application is to provide a method and device for detecting IDH mutations in gliomas based on Raman spectroscopy, so as to achieve rapid and accurate detection of the IDH mutation status of gliomas.

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

[0006] In the first aspect, the present application provides a method for detecting IDH mutations in gliomas based on Raman spectroscopy, 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 a characteristic displacement from the normalized Raman spectroscopy data; the characteristic displacement includes a Raman displacement of 1161 cm -1 、1200cm -1 、1412cm -1 、1518cm -1 、1524cm -1 、1592cm -1 、2839cm -1 、2916cm -1、2958cm -1 、2971cm -1 、3176cm -1 、3180cm -1 and 3184cm -1 ; Input the intensity value at the characteristic displacement into the classification model to calculate the classification prediction value; the classification model is a multivariate linear equation of the classification prediction value and the intensity value at the characteristic displacement; the classification prediction value is used to assist in determining whether the glioma tissue sample to be tested is an IDH mutant glioma.

[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; performing spectrum smoothing on the Raman spectrum data with fluorescence background removed to obtain smoothed Raman spectrum data; and normalizing the smoothed Raman spectrum data to obtain normalized Raman spectrum data.

[0008] Optionally, the classification model is determined based on the LASSO regression algorithm; the determination process of the classification model includes: obtaining a number of Raman spectrum samples marked with the IDH mutation status of glioma, and preprocessing and extracting characteristic spectrum information to construct a sample data set; the sample data set includes characteristic spectrum information and glioma IDH mutation status corresponding to a number of normalized Raman spectrum samples; the characteristic spectrum information includes intensity values ​​at a number of Raman shifts; the glioma IDH mutation status includes 1 and 0, wherein 1 represents IDH mutant glioma and 0 represents IDH wild-type glioma; a linear regression model is constructed with the glioma IDH mutation status as the target variable and the characteristic spectrum information as the characteristic variable; according to the sample data set, a 5-fold cross-validation is used to determine the optimal regularization parameter value, and based on the optimal regularization parameter value, the LASSO regression algorithm is used to train the linear regression model, and the coefficient of the irrelevant Raman shift in the linear regression model is gradually reduced to zero to obtain a characteristic shift and a classification model.

[0009] Optionally, the classification model is expressed as: y = 0.7273 + (-0.3157 × I 1161 ) + 0.2600 × I 1200 + 0.1042 × I 1412 + (-0.1198 × I 1518 ) + (-0.5139 × I 1524 ) + (-0.1682 × I 1592 ) + 0.0769 × I 2839 + 0.1626 × I2916 + 0.0212 × I 2958 + 0.1187 × I 2971 + (-0.2708 × I 3176 ) + (-0.2196 × I 3180 ) + (-0.0287 × I 3184 ), where y represents the classification prediction value. When |y|≥0.5, it is determined to be an IDH mutant glioma, and when |y|<0.5, it is determined to be an IDH wild-type glioma. 1161 Indicates that the Raman shift is 1161cm -1 The intensity value at I 1200 Indicates the Raman shift of 1200cm -1 The intensity value at I 1412 Indicates that the Raman shift is 1412 cm -1 The intensity value at I 1518 Indicates the Raman shift of 1518 cm -1 The intensity value at I 1524 Indicates the Raman shift of 1524 cm -1 The intensity value at I 1592 Indicated at Raman shift 1592cm -1 The intensity value at I 2839 Indicated at Raman shift 2839cm -1 The intensity value at I 2916 Indicated at Raman shift 2916cm -1 The intensity value at I 2958 Indicated at Raman shift 2958cm -1 The intensity value at I 2971 Indicated at Raman shift 2971cm -1 The intensity value at I 3176 Indicates that the Raman shift is 3176cm -1 The intensity value at I 3180 Indicates the Raman shift of 3180cm -1 The intensity value at I 3184 Indicated at Raman shift 3184cm -1 The intensity value at .

[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 the background removal function provided in the RamPy library of python 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 spectral smoothing processing on the Raman spectral data after removing the fluorescence background to obtain smoothed Raman spectral data includes: performing spectral smoothing processing on the Raman spectral data after removing the fluorescence background using a Savitzky-Golay smoothing algorithm to obtain smoothed Raman spectral data.

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

[0013] In the second aspect, the present application provides a glioma IDH mutation detection device based on Raman spectroscopy, which adopts the above-mentioned glioma IDH mutation detection method based on Raman spectroscopy, 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 glioma IDH mutation detection device based on Raman spectroscopy 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 glioma IDH mutation detection device based on Raman spectroscopy 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.

[0016] Optionally, the glioma IDH mutation detection device based on Raman spectroscopy also includes: a movable slide for placing the glioma tissue sample to be tested, and adjusting the position and height of the glioma tissue sample to be tested.

[0017] 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 IDH mutations in gliomas based on Raman spectroscopy, firstly, a Raman spectroscopy scan is performed on the glioma tissue sample to be tested to obtain Raman spectroscopy data, then the Raman spectroscopy 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 predetermined classification model to calculate the classification prediction value to assist in determining whether the glioma tissue sample to be tested is an IDH mutant glioma. Compared with the prior art, the present application adopts Raman spectroscopy technology, does not require complex pretreatment and preparation processes for the glioma tissue sample to be tested, is simple to operate, has a short detection time, and can obtain results quickly; in addition, the present application uses a classification model to calculate the classification prediction value based on the intensity value at the Raman displacement that has a high correlation with the IDH mutation state, which can achieve accurate identification of the IDH mutation state of gliomas. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] 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.

[0019] Figure 1 A flow chart of the method for detecting IDH mutations in glioma based on Raman spectroscopy provided in this application.

[0020] Figure 2 The average Raman spectra of IDH mutant glioma and IDH wild-type glioma provided in this application.

[0021] Figure 3 This is a prediction performance curve of the classification model provided in this application.

[0022] Figure 4 This is a structural diagram of the glioma IDH mutation detection device based on Raman spectroscopy provided in this application.

[0023] Reference numerals: laser-1, spectrometer-2, computer-3, display-4. DETAILED DESCRIPTION

[0024] 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.

[0025] The purpose of this application is to provide a method and device for detecting IDH mutations in gliomas based on Raman spectroscopy. By using Raman spectroscopy technology, the mutation status of the IDH gene can be identified by performing Raman spectroscopy analysis on glioma tissue samples, thereby realizing rapid diagnosis of gliomas and formulation of treatment plans. This application provides a simple, efficient and accurate method and device for detecting IDH mutations in gliomas, providing a new technical means for the field of clinical medical diagnosis and treatment.

[0026] 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.

[0027] like Figure 1 As shown, the method for detecting IDH mutation in glioma based on Raman spectroscopy provided in the present application includes the following steps.

[0028] Step S1: Perform Raman spectrum scanning on the glioma tissue sample to be tested to obtain Raman spectrum data.

[0029] Among them, the glioma tissue samples to be tested can be directly collected during surgery without any slicing. They can be placed on a glass slide at a suitable size (2-20 mm in long diameter). In addition, the integration time of the Raman spectrum scan is controlled within 5 seconds.

[0030] Step S2: preprocessing the Raman spectrum data to obtain normalized Raman spectrum data.

[0031] Specifically, the Raman spectrum data is firstly 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 spectrally smoothed to obtain the smoothed Raman spectrum data; finally, the smoothed Raman spectrum data is normalized to obtain the normalized Raman spectrum data.

[0032] Preferably, the fluorescence background removal process uses the background removal function in the Python RamPy library. The smoothing process uses the Savitzky-Golay smoothing algorithm, and the Python function name is savgol_filter. The normalization process uses the maximum and minimum normalization algorithm, and the formula is ;in, is the normalized Raman spectral data, For smoothed Raman spectral data, is the minimum value of the smoothed Raman spectrum data, is the maximum value of the smoothed Raman spectrum data.

[0033] Step S3: Extract the intensity value at the characteristic shift from the normalized Raman spectrum data. The characteristic shift specifically includes the Raman shift 1161cm -1 、1200cm -1 、1412cm -1 、1518cm -1 、1524cm -1 、1592cm -1 、2839cm -1 、2916cm -1 、2958cm -1 、2971cm -1 、3176cm -1 、3180cm -1 and 3184cm -1 ,These Raman shifts are highly correlated with the IDH mutation status, facilitating the accurate identification of the IDH mutation status of glioma.

[0034] Step S4: Input the intensity value at the characteristic displacement into the classification model to calculate the classification prediction value. The classification model is determined based on machine learning or pattern recognition algorithm, specifically a multivariate linear equation of the classification prediction value and the intensity value at the characteristic displacement. The classification prediction value is used to assist in determining whether the glioma tissue sample to be tested is an IDH mutant glioma (or an IDH wild-type glioma).

[0035] Preferably, the present application determines a classification model based on a least absolute shrinkage and selection operator (LASSO) regression algorithm, and the specific steps include: obtaining a number of Raman spectrum samples annotated with the IDH mutation status of gliomas, and preprocessing and extracting characteristic spectrum information to construct a sample data set; the sample data set includes characteristic spectrum information and glioma IDH mutation status corresponding to a number of normalized Raman spectrum samples; the characteristic spectrum information includes intensity values ​​at a number of Raman shifts; the glioma IDH mutation status includes 1 and 0, wherein 1 represents an IDH mutant glioma and 0 represents an IDH wild-type glioma; a linear regression model is constructed using the glioma IDH mutation status as the target variable and the characteristic spectrum information as the characteristic variable; according to the sample data set, a 5-fold cross-validation is used to determine the optimal regularization parameter value, and based on the optimal regularization parameter value, the linear regression model is trained using the LASSO regression algorithm to gradually reduce the coefficient of the irrelevant Raman shift in the linear regression model to zero, and obtain the characteristic shift and classification model.

[0036] In a specific embodiment, the present application constructs a sample data set for training based on 2,393 spectral data from 620 patients retrospectively included, uses the LASSO regression algorithm for 5-fold cross-validation, selects the optimal regularization parameter λ value (λ=0.0046), gradually reduces the coefficient of unrelated Raman shifts to zero, and finally selects 13 characteristic shifts and generates a classification model to predict IDH mutation status.

[0037] When using the LASSO regression algorithm, first define its cost function expression as: .

[0038] Among them, n is the number of samples, p is the number of features, and y i is the observed value of the i-th sample, x ij is the jth feature of the i-th sample, β is the regression coefficient, β j is the regression coefficient of the jth feature, β0 is the initial value of the regression coefficient, and λ is the regularization parameter used to control the strength of regularization. While minimizing the cost function, a set of regression coefficients corresponding to each feature is obtained, so that the predicted value is closest to the actual observed value.

[0039] The specific algorithm steps are as follows.

[0040] Data preparation: 2393 spectral data of 620 patients were divided into training set and test set.

[0041] Standardization (preprocessing): Standardize the spectral data to ensure that different features have the same scale.

[0042] 5-fold cross validation: The training data is randomly divided into 5 parts, 4 of which are used for training each time and 1 for validation, and the cycle is repeated 5 times.

[0043] Model training: The LASSO regression algorithm is used to train the model in each cross-validation, and the parameter λ is adjusted to control feature selection and model complexity.

[0044] Select the best λ value: Through cross-validation, select the λ value that minimizes the prediction error on the validation set (here λ=0.0046).

[0045] Feature selection: The model was trained using the optimal λ value to gradually shrink the coefficients of irrelevant Raman shifts to zero, and finally 13 feature shifts were selected.

[0046] Model generation: Generate the final classification model based on the selected feature displacement.

[0047] The classification model constructed according to the above method has excellent performance on the validation set. The average Raman spectra of IDH mutant glioma and IDH wild-type glioma are shown in Figure 2. Figure 2As shown in the figure, the prediction performance curve of the classification model is as follows Figure 3 The area under the curve (AUC) of the receiver operating characteristic curve (ROC) of the classification model can reach up to 94.6%, and the sensitivity can reach 89.6%.

[0048] The 13 characteristic displacements finally selected can be divided into the following 8 characteristic peaks.

[0049] 1161cm -1 :It is formed by the CC bond and CN bond of β-carotene. Its existence itself can reduce lipid peroxidation. Its antioxidant property is mainly manifested in its ability to scavenge free radicals, prevent the chain reaction of free radicals, and thus reduce the damage of free radicals to cells.

[0050] 1200cm -1 : Formed by the mixing action of lipids and nucleotides.

[0051] 1412cm -1 : It is formed by the interaction of pyruvic acid and lipids, and is similar to 1200cm -1 There is a similar vibration mode. Pyruvate is an important intermediate in the sugar metabolism of all biological cells and the mutual conversion of various substances in the body.

[0052] 1518~1524cm -1 : 1161cm -1 The peaks have the same vibration mode and proportional signal intensity, and the spectrum of pure β-carotene also has a second peak here.

[0053] 1592cm -1 :This peak may be related to the Amide I band, which is mainly C=O stretching vibration. At the same time, it may also correspond to the characteristic vibration of aromatic compounds, such as tyrosine and tryptophan. Tryptophan is a synthetic precursor of many biologically active molecules, such as pigments, hormones and neurotransmitters.

[0054] 2839cm -1 : The Raman peak of the methyl group is caused by the stretching vibration of the CH bond in the methyl group and is related to the protein content.

[0055] 2916~2971cm -1 : Characteristic peak of methyl bond CH3, related to lipid content.

[0056] 3176~3184cm -1: These peaks are often associated with Amide B vibrations in proteins, involving NH stretching. Vibrations in this region may represent features of protein secondary structure.

[0057] The final classification model is expressed as follows.

[0058] y = 0.7273 + (-0.3157 × I 1161 ) + 0.2600 × I 1200 + 0.1042 × I 1412 + (-0.1198 × I 1518 ) + (-0.5139 × I 1524 ) + (-0.1682 × I 1592 ) + 0.0769 × I 2839 +0.1626 × I 2916 + 0.0212 × I 2958 + 0.1187 × I 2971 + (-0.2708 × I 3176 ) + (-0.2196× I 3180 ) + (-0.0287 × I 3184 ).

[0059] Wherein, y represents the classification prediction value. When |y|≥0.5, it is judged as IDH mutant glioma, and when |y|<0.5, it is judged as IDH wild-type glioma. 1161 Indicated at Raman shift 1161cm -1 The intensity value at I 1200 Indicates the Raman shift of 1200cm -1 The intensity value at I 1412 Indicates that the Raman shift is 1412 cm -1 The intensity value at I 1518 Indicates the Raman shift of 1518 cm -1 The intensity value at I 1524 Indicates the Raman shift of 1524 cm -1 The intensity value at I 1592 Indicated at Raman shift 1592cm -1 The intensity value at I 2839 Indicated at Raman shift 2839cm -1 The intensity value at I 2916 Indicated at Raman shift 2916cm -1 The intensity value at I 2958 Indicated at Raman shift 2958cm -1 The intensity value at I 2971Indicated at Raman shift 2971cm -1 The intensity value at I 3176 Indicates that the Raman shift is 3176cm -1 The intensity value at I 3180 Indicates the Raman shift of 3180cm -1 The intensity value at I 3184 Indicated at Raman shift 3184cm -1 The intensity value at .

[0060] like Figure 4 As shown, the glioma IDH mutation detection device based on Raman spectroscopy provided by the present application includes a laser 1, a spectrometer 2, a computer 3 and a display 4. Among them, the laser 1 is used to emit a laser to the glioma tissue sample to be tested. The spectrometer 2 is used to collect Raman 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. The display 4 is connected to the computer 3, and is used to display the classification prediction value and the auxiliary judgment result determined based on the classification prediction value.

[0061] Furthermore, the above 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 light 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.

[0062] Furthermore, the above device also includes a movable slide, which is used to place the glioma tissue sample to be tested and adjust the position and height of the glioma tissue sample to be tested.

[0063] The Raman spectroscopy-based glioma IDH mutation detection method and device provided in this application have the following advantages.

[0064] 1. Fast: Using Raman spectroscopy technology, the detection time is short and the results are obtained quickly.

[0065] 2. Accurate: The analysis using characteristic spectral information can accurately identify the mutation status of the IDH gene.

[0066] 3. Non-destructive: No additional processing or staining of tissue samples is required, and the integrity of the samples can be maintained.

[0067] 4. Simple: easy to operate, no need for complicated experimental conditions and professional skills.

[0068] In summary, the present application provides a method and device for detecting IDH mutations in glioma based on Raman spectroscopy, which has the advantages of being fast, accurate, and non-destructive, and is suitable for clinical medical diagnosis and scientific research.

[0069] 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.

[0070] 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 IDH mutations in gliomas for non-diagnostic purposes based on Raman spectroscopy, 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 1161cm -1 、1200cm -1 、1412cm -1 、1518cm -1 、1524cm -1 、1592cm -1 、2839cm -1 、2916cm -1 、2958cm -1 、2971cm -1 、3176cm -1 、3180cm -1 and 3184cm -1 ; Inputting the intensity value at the characteristic displacement into a classification model to calculate a classification prediction value; the classification model is a multivariate linear equation of the classification prediction value and the intensity value at the characteristic displacement; the classification prediction value is used to assist in determining whether the glioma tissue sample to be tested is an IDH mutant glioma; The expression of the classification model is: y = 0.7273 + (-0.3157 × I 1161 ) + 0.2600 × I 1200 + 0.1042 × I 1412 + (-0.1198 × I 1518 ) + (-0.5139 × I 1524 ) + (-0.1682 × I 1592 ) + 0.0769 × I 2839 +0.1626 × I 2916 + 0.0212 × I 2958 + 0.1187 × I 2971 + (-0.2708 × I 3176 ) + (-0.2196 × I 3180 ) + (-0.0287 × I 3184 ); Wherein, y represents the classification prediction value. When |y|≥0.5, it is judged as IDH mutant glioma, and when |y|<0.5, it is judged as IDH wild-type glioma. 1161 Indicates that the Raman shift is 1161cm -1 The intensity value at I 1200 Indicates the Raman shift of 1200cm -1 The intensity value at I 1412 Indicates that the Raman shift is 1412 cm -1 The intensity value at I 1518 Indicates the Raman shift of 1518 cm -1 The intensity value at I 1524 Indicates the Raman shift of 1524 cm -1 The intensity value at I 1592 Indicated at Raman shift 1592cm -1 The intensity value at I 2839 Indicated at Raman shift 2839cm -1 The intensity value at I 2916 Indicated at Raman shift 2916cm -1 The intensity value at I 2958 Indicated at Raman shift 2958cm -1 The intensity value at I 2971 Indicated at Raman shift 2971cm -1 The intensity value at I 3176 Indicates that the Raman shift is 3176cm -1 The intensity value at I 3180 Indicates the Raman shift of 3180cm -1 The intensity value at I 3184 Indicated at Raman shift 3184cm -1 The intensity value at .

2. The method for detecting IDH mutations in gliomas for non-diagnostic purposes based on Raman spectroscopy 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 spectral smoothing processing on the Raman spectral data after removing the fluorescence background to obtain smoothed Raman spectral data; The smoothed Raman spectrum data is normalized to obtain normalized Raman spectrum data.

3. The method for detecting IDH mutation in glioma for non-diagnostic purposes based on Raman spectroscopy according to claim 1, characterized in that: The classification model is determined based on the LASSO regression algorithm; the determination process of the classification model includes: Acquire several Raman spectrum samples annotated with the IDH mutation status of glioma, perform preprocessing and extract characteristic spectrum information, and construct a sample data set; the sample data set includes characteristic spectrum information and glioma IDH mutation status corresponding to several normalized Raman spectrum samples; the characteristic spectrum information includes intensity values ​​at several Raman shifts; the glioma IDH mutation status includes 1 and 0, wherein 1 represents IDH mutant glioma and 0 represents IDH wild-type glioma; A linear regression model is constructed using the glioma IDH mutation status as the target variable and the characteristic spectrum information as the characteristic variable; According to the sample data set, a 5-fold cross validation is used to determine the optimal regularization parameter value, and based on the optimal regularization parameter value, the linear regression model is trained using a LASSO regression algorithm, and the coefficient of the uncorrelated Raman shift in the linear regression model is gradually reduced to zero to obtain a characteristic shift and classification model.

4. The method for detecting IDH mutations in gliomas for non-diagnostic purposes based on Raman spectroscopy according to claim 2, characterized in that: Performing spectral smoothing on the Raman spectral data after removing the fluorescence background to obtain smoothed Raman spectral data includes: The Savitzky-Golay smoothing algorithm is used to perform spectral smoothing on the Raman spectral data after removing the fluorescence background, so as to obtain smoothed Raman spectral data.

5. The method for detecting IDH mutations in gliomas for non-diagnostic purposes based on Raman spectroscopy according to claim 2, characterized in that: The smoothed Raman spectrum data is normalized to obtain normalized Raman spectrum data, including: The smoothed Raman spectrum data is normalized using a maximum-minimum normalization algorithm to obtain normalized Raman spectrum data.

Images