Brain glioma IDH gene prediction method based on metamaterial sensing

Abstract

The invention relates to a brain glioma IDH gene prediction method based on metamaterial sensing, which comprises the following steps: irradiating brain glioma with IDH gene and brain glioma without IDH gene by using terahertz metamaterial sensors with different resonant frequencies, and acquiring terahertz absorption signals of the brain glioma in different frequency bands; acquiring characteristic indexes of terahertz absorption signals of different frequency bands; calculating the relevancy between the feature indexes; the terahertz absorption signals with the relevancy smaller than a preset threshold value are removed, and screened characteristic indexes are obtained; and inputting the screened characteristic indexes as training samples into a neural network for training to obtain a brain glioma IDH gene prediction model. The metamaterial has highly customized resonance response to electromagnetic waves, can greatly enhance the sensing ability to weak molecular changes (such as molecular structure changes caused by IDH genes), and effectively improves the detection sensitivity and specificity in cooperation with multi-band signal acquisition and feature screening.

Description

Technical Field

[0001] The present invention relates to the technical field of gene prediction, and in particular to a method for predicting the IDH gene of brain glioma based on metamaterial sensing. Background Art

[0002] Glioma is a common malignant tumor of the central nervous system, characterized by high incidence, aggressiveness, and poor prognosis. Recent molecular pathology studies have revealed that the mutational status of the isocitrate dehydrogenase (IDH) gene is closely associated with the development, progression, and prognosis of glioma. Therefore, accurate and rapid detection of IDH gene status in glioma tissue is crucial for clinical classification, treatment decision-making, and prognostic assessment.

[0003] Currently, IDH gene testing primarily relies on methods such as histopathological immunohistochemistry, molecular biology PCR, and next-generation sequencing (NGS). While these technologies can provide relatively accurate genetic information, the testing process is generally plagued by issues such as difficult sampling, complex procedures, long testing cycles, and high costs. Furthermore, some technologies rely on highly specialized equipment and experimental environments, making rapid and convenient clinical application difficult. These limitations are particularly evident in real-time surgical assessment and bedside testing.

[0004] Metamaterials are a new type of material with artificially designed microstructures that can exhibit electromagnetic properties not seen in nature. In recent years, sensing technology based on metamaterials has been widely used in the fields of biomolecule detection and medical imaging due to its high sensitivity, fast response speed, and integration. The use of metamaterial sensors can achieve efficient detection of trace biomolecule changes, and has outstanding advantages such as non-destructive, real-time, and low-cost. Combining metamaterial sensing with IDH gene detection for brain glioma can provide a new detection method for molecular diagnosis, which is expected to overcome many shortcomings of existing molecular detection methods and achieve rapid, accurate, and convenient prediction of IDH genes for brain glioma. Therefore, the development of a method for predicting IDH genes in brain gliomas based on metamaterial sensing has important scientific significance and broad application prospects. Summary of the Invention

[0005] To solve the above problems, the embodiment of the present invention aims to provide a method for predicting IDH gene in brain glioma based on metamaterial sensing.

[0006] A method for predicting IDH genes in brain glioma based on metamaterial sensing, comprising:

[0007] Step 1: Use terahertz metamaterial sensors with different resonant frequencies to irradiate brain gliomas with and without the IDH gene, and obtain terahertz absorption signals of the gliomas at different frequency bands;

[0008] Step 2: Obtain characteristic indicators of terahertz absorption signals in different frequency bands;

[0009] Step 3: Calculate the correlation between the characteristic indices of the terahertz absorption signals of brain gliomas with IDH genes at different frequency bands and the characteristic indices of the terahertz absorption signals of brain gliomas without IDH genes at different frequency bands;

[0010] Step 4: Remove the terahertz absorption signals in the corresponding frequency bands whose correlation is less than a preset threshold to obtain the filtered characteristic indicators;

[0011] Step 5: Input the filtered characteristic indicators as training samples into the neural network for training to obtain the IDH gene prediction model for brain glioma;

[0012] Step 6: Use the brain glioma IDH gene prediction model to predict whether the target brain glioma carries the IDH gene to obtain a prediction result.

[0013] Preferably, the step 3: calculating the correlation between the characteristic index of the terahertz absorption signal of the brain glioma with the IDH gene in different frequency bands and the characteristic index of the terahertz absorption signal of the brain glioma without the IDH gene in different frequency bands, comprises:

[0014] Step 3.1: Sort the characteristic indicators of the terahertz absorption signal by frequency band to form a frequency band sequence;

[0015] Step 3.2: Subtract adjacent frequency band sequences to form a specific frequency band detection sequence;

[0016] Step 3.3: Calculate the specific value of the characteristic index in each frequency band using the specific frequency band detection sequence;

[0017] Step 3.4: retain the frequency band sequence corresponding to the specific value within the preset range to form the terahertz absorption signal under the filtered frequency band;

[0018] Step 3.5: Calculate the correlation between the characteristic index of the terahertz absorption signal of the brain glioma with the IDH gene in the post-screening frequency band and the characteristic index of the terahertz absorption signal of the brain glioma without the IDH gene in the post-screening frequency band.

[0019] Preferably, in step 3.3, the calculation formula of the specific value is:

[0020]

[0021] Among them, Δ j (k)=|x j (k)-x j+1 (k)| represents the specific frequency band detection sequence, x j (k) represents the value of the kth characteristic index in the jth frequency band, x j+1 (k) represents the value of the kth characteristic index in the j+1th frequency band, m represents the minimum value in the specific frequency band detection sequence, M represents the maximum value in the specific frequency band detection sequence, ξ represents the preset coefficient, and θ represents the specific value of the characteristic index in the jth frequency band.

[0022] Preferably, the step 3.5: calculating the correlation between the characteristic index of the terahertz absorption signal of the brain glioma with the IDH gene in the post-screening frequency band and the characteristic index of the terahertz absorption signal of the brain glioma without the IDH gene in the post-screening frequency band includes:

[0023] Step 3.5.1: Calculate the similarity between the characteristic indices of the terahertz absorption signal of the brain glioma with the IDH gene in the post-screening frequency band and the characteristic indices of the terahertz absorption signal of the brain glioma without the IDH gene in the post-screening frequency band;

[0024] Step 3.5.2: construct a similarity matrix using the similarity;

[0025] Step 3.5.3: Get the eigenvalues ​​of the similarity matrix;

[0026] Step 3.5.4: Calculate the correlation corresponding to each feature indicator based on the eigenvalues ​​of the similarity matrix.

[0027] Preferably, in step 3.5.1, a similarity calculation formula is used to calculate the similarity between the characteristic index of the terahertz absorption signal of the brain glioma with the IDH gene in the frequency band after screening and the characteristic index of the terahertz absorption signal of the brain glioma without the IDH gene in the frequency band after screening; wherein, the similarity calculation formula is:

[0028]

[0029] Among them, r ij represents the similarity between the characteristic index with IDH gene in the i-th frequency band and the characteristic index without IDH gene in the j-th frequency band, x ki represents the kth characteristic index under the i-th frequency band, Represents the mean value of the characteristic index under the i-th frequency band, y kj represents the kth characteristic index under the jth frequency band, Represents the mean value of the characteristic index in the jth frequency band.

[0030] Preferably, in step 3.5.2, the similarity matrix constructed is:

[0031]

[0032] Among them, r 11 represents the similarity between the characteristic index with IDH gene in the first frequency band and the characteristic index without IDH gene in the first frequency band, r nm It represents the similarity between the characteristic index with IDH gene in the nth frequency band and the characteristic index without IDH gene in the nth frequency band.

[0033] Preferably, the step 3.5.4: calculating the correlation corresponding to each characteristic indicator based on the eigenvalue of the similarity matrix includes:

[0034] Obtain the eigenvalues ​​of the similarity matrix and arrange them from large to small to form a descending sequence. Calculate the correlation corresponding to each characteristic indicator based on the descending sequence. The correlation calculation formula is:

[0035]

[0036] in, Indicates the correlation corresponding to the i-th feature index, λ i represents the i-th value in the descending sequence, λ k Represents the kth value in a descending sequence.

[0037] Preferably, the step 5: inputting the screened characteristic indicators as training samples into a neural network for training to obtain a brain glioma IDH gene prediction model includes:

[0038] The filtered feature indicators are used as training samples to input into the BP neural network for training to obtain the IDH gene prediction model for brain glioma; wherein, the loss function in the training process is:

[0039]

[0040] in, Indicates the loss value, y indicates the status of the IDH gene. When y is 1, it is positive, and when y is 0, it is negative. Indicates the label predicted by the BP neural network.

[0041] The present invention also provides a brain glioma IDH gene prediction system based on metamaterial sensing, comprising:

[0042] A signal acquisition module is used to irradiate brain gliomas with and without the IDH gene using terahertz metamaterial sensors with different resonance frequencies, and to acquire terahertz absorption signals of the brain gliomas at different frequency bands;

[0043] A characteristic index acquisition module is used to obtain characteristic indicators of terahertz absorption signals in different frequency bands;

[0044] A correlation calculation module is used to calculate the correlation between characteristic indicators of terahertz absorption signals of brain gliomas with IDH genes in different frequency bands and characteristic indicators of terahertz absorption signals of brain gliomas without IDH genes in different frequency bands;

[0045] A screening module is used to remove the terahertz absorption signals in the corresponding frequency band whose correlation is less than a preset threshold, and obtain the filtered characteristic indicators;

[0046] A training module is used to input the screened characteristic indicators as training samples into the neural network for training to obtain a brain glioma IDH gene prediction model;

[0047] The prediction module is used to use the brain glioma IDH gene prediction model to predict whether the target brain glioma carries the IDH gene to obtain a prediction result.

[0048] The present invention also provides an electronic device, comprising a bus, a transceiver, a memory, a processor, and a computer program stored in the memory and runnable on the processor, wherein the transceiver, the memory, and the processor are connected via the bus, and wherein the computer program, when executed by the processor, implements the steps of the above-mentioned method for predicting IDH genes in brain gliomas based on metamaterial sensing.

[0049] The present invention also provides a computer-readable storage medium having a computer program stored thereon, characterized in that when the computer program is executed by a processor, the steps in the above-mentioned method for predicting the IDH gene of brain glioma based on metamaterial sensing are implemented.

[0050] According to the specific embodiments provided by the present invention, the present invention discloses the following technical effects:

[0051] The present invention relates to a method for predicting the IDH gene in brain gliomas based on metamaterial sensing. Compared with the existing technology, the present invention uses terahertz metamaterial sensors to irradiate brain glioma tissue, which can achieve non-destructive and non-invasive detection of samples, avoiding the loss of tissue samples and complex processing procedures in traditional genetic testing, significantly shortening the detection cycle, and meeting real-time and rapid detection requirements. In addition, the metamaterial has a highly customized resonant response to electromagnetic waves, which can greatly enhance the perception of weak molecular changes (such as molecular structure changes caused by the IDH gene). Combined with the multi-band signal acquisition and feature screening in the present invention, the sensitivity and specificity of detection are effectively improved.

[0052] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0054] Figure 1 A flow chart of a method for predicting IDH genes in glioma based on metamaterial sensing provided by the present invention;

[0055] Figure 2 This is a schematic diagram of the principle of the IDH gene prediction system for glioma based on metamaterial sensing provided by the present invention. DETAILED DESCRIPTION

[0056] In the description of the present invention, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", "clockwise", "counterclockwise" and the like to indicate orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as limiting the present invention.

[0057] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of the technical features being referred to. Thus, a feature identified as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, "plurality" means two or more, unless otherwise specifically defined.

[0058] In the present invention, unless otherwise expressly specified or limited, the terms "mounted," "connected," "connect," "fixed," etc. should be understood broadly. For example, they may refer to fixed, detachable, or integral connections; mechanical or electrical connections; direct or indirect connections through an intermediary; or internal communication between two components. Those skilled in the art will understand the specific meanings of the above terms in the present invention based on specific circumstances.

[0059] See also Figure 1 , a method for predicting IDH gene in brain glioma based on metamaterial sensing, comprising:

[0060] Step 1: Use terahertz metamaterial sensors with different resonant frequencies to irradiate brain gliomas with and without the IDH gene, and obtain terahertz absorption signals of the gliomas at different frequency bands;

[0061] Step 2: Obtain characteristic indicators of terahertz absorption signals in different frequency bands;

[0062] In practical applications, the characteristic indicators of terahertz absorption signals include but are not limited to: absorption peak position, absorption peak intensity, absorption peak width, spectrum integral area, absorption curve slope, signal mean, signal variance, signal skewness and signal kurtosis.

[0063] Step 3: Calculate the correlation between the characteristic indices of the terahertz absorption signals of brain gliomas with IDH genes at different frequency bands and the characteristic indices of the terahertz absorption signals of brain gliomas without IDH genes at different frequency bands;

[0064] Furthermore, step 3 includes:

[0065] Step 3.1: Sort the characteristic indicators of the terahertz absorption signal by frequency band to form a frequency band sequence;

[0066] Step 3.2: Subtract adjacent frequency band sequences to form a specific frequency band detection sequence;

[0067] Step 3.3: Calculate the specific value of the characteristic index in each frequency band using the specific frequency band detection sequence;

[0068] In step 3.3, the calculation formula of the specific value is:

[0069]

[0070] Among them, Δ j (k)=|x j (k)-x j+1 (k)| represents the specific frequency band detection sequence, x j (k) represents the value of the kth characteristic index in the jth frequency band, x j+1 (k) represents the value of the kth characteristic index in the j+1th frequency band, m represents the minimum value in the specific frequency band detection sequence, M represents the maximum value in the specific frequency band detection sequence, ξ represents the preset coefficient, and θ represents the specific value of the characteristic index in the jth frequency band.

[0071] Step 3.4: retain the frequency band sequence corresponding to the specific value within the preset range to form the terahertz absorption signal under the filtered frequency band;

[0072] By sorting and subtracting frequency band sequences to generate specific frequency band detection sequences, this method can effectively identify subtle but critical differences in terahertz spectra between different glioma samples (with and without the IDH gene). Calculating specific values ​​and selecting frequency bands that fall within a preset range essentially preserves the frequency bands and signal features most valuable for distinguishing IDH gene status, effectively filtering out redundant or non-discriminative information and improving the accuracy and robustness of subsequent prediction models.

[0073] Step 3.5: Calculate the correlation between the characteristic index of the terahertz absorption signal of the brain glioma with the IDH gene in the post-screening frequency band and the characteristic index of the terahertz absorption signal of the brain glioma without the IDH gene in the post-screening frequency band;

[0074] Furthermore, step 3.5 includes:

[0075] Step 3.5.1: Calculate the similarity between the characteristic indices of the terahertz absorption signal of the brain glioma with the IDH gene in the post-screening frequency band and the characteristic indices of the terahertz absorption signal of the brain glioma without the IDH gene in the post-screening frequency band;

[0076] In step 3.5.1, a similarity calculation formula is used to calculate the similarity between the characteristic index of the terahertz absorption signal of the brain glioma with the IDH gene in the post-screening frequency band and the characteristic index of the terahertz absorption signal of the brain glioma without the IDH gene in the post-screening frequency band; wherein, the similarity calculation formula is:

[0077]

[0078] Among them, rij represents the similarity between the characteristic index with IDH gene in the i-th frequency band and the characteristic index without IDH gene in the j-th frequency band, x ki represents the kth characteristic index under the i-th frequency band, Represents the mean value of the characteristic index under the i-th frequency band, y kj represents the kth characteristic index under the jth frequency band, Represents the mean value of the characteristic index in the jth frequency band.

[0079] Step 3.5.2: Use the similarity to construct a similarity matrix; wherein the similarity matrix is:

[0080]

[0081] Among them, r 11 represents the similarity between the characteristic index with IDH gene in the first frequency band and the characteristic index without IDH gene in the first frequency band, r nm It represents the similarity between the characteristic index with IDH gene in the nth frequency band and the characteristic index without IDH gene in the nth frequency band.

[0082] Step 3.5.3: Get the eigenvalues ​​of the similarity matrix;

[0083] Step 3.5.4: Calculate the correlation corresponding to each feature indicator based on the eigenvalue of the similarity matrix;

[0084] In step 3.5.4, the eigenvalues ​​of the similarity matrix are obtained and arranged from large to small to form a descending sequence. The correlation corresponding to each characteristic index is calculated based on the descending sequence; wherein the correlation calculation formula is:

[0085]

[0086] in, Indicates the correlation corresponding to the i-th feature index, λ i represents the i-th value in the descending sequence, λ k Represents the kth value in a descending sequence.

[0087] Step 4: Remove the terahertz absorption signals in the corresponding frequency bands whose correlation is less than a preset threshold to obtain the filtered characteristic indicators;

[0088] This method uses a similarity matrix and eigenvalue analysis method to deeply optimize the filtered frequency band features, effectively identifying and eliminating highly correlated or redundant features while retaining key feature indicators with complementary information and strong independence. This process not only improves the accuracy and generalization ability of the final neural network model, but also enhances the model's interpretability and practicality, greatly improving the overall effectiveness and reliability of the method for predicting IDH gene status in brain gliomas.

[0089] Step 5: Input the filtered characteristic indicators as training samples into the neural network for training to obtain the IDH gene prediction model for brain glioma;

[0090] In step 5, the filtered characteristic indicators are input as training samples into the BP neural network for training to obtain the IDH gene prediction model for glioma; wherein the loss function during the training process is:

[0091]

[0092] in, Indicates the loss value, y indicates the status of the IDH gene. When y is 1, it is positive, and when y is 0, it is negative. Indicates the label predicted by the BP neural network.

[0093] Step 6: Use the brain glioma IDH gene prediction model to predict whether the target brain glioma carries the IDH gene to obtain a prediction result.

[0094] The present invention uses terahertz metamaterial sensors to irradiate brain glioma tissue, which can achieve non-destructive and non-invasive detection of samples, avoiding the loss of tissue samples and complex processing procedures in traditional genetic testing, significantly shortening the detection cycle, and realizing real-time and rapid detection needs; in addition, metamaterials have a highly customized resonant response to electromagnetic waves, which can greatly enhance the perception of weak molecular changes (such as molecular structure changes caused by IDH genes). Combined with the multi-band signal acquisition and feature screening in the present invention, the sensitivity and specificity of detection are effectively improved.

[0095] See also Figure 2 The present invention also provides a brain glioma IDH gene prediction system based on metamaterial sensing, comprising:

[0096] A signal acquisition module is used to irradiate brain gliomas with and without the IDH gene using terahertz metamaterial sensors with different resonance frequencies, and to acquire terahertz absorption signals of the brain gliomas at different frequency bands;

[0097] A characteristic index acquisition module is used to obtain characteristic indicators of terahertz absorption signals in different frequency bands;

[0098] A correlation calculation module is used to calculate the correlation between characteristic indicators of terahertz absorption signals of brain gliomas with IDH genes in different frequency bands and characteristic indicators of terahertz absorption signals of brain gliomas without IDH genes in different frequency bands;

[0099] A screening module is used to remove the terahertz absorption signals in the corresponding frequency band whose correlation is less than a preset threshold, and obtain the filtered characteristic indicators;

[0100] A training module is used to input the screened characteristic indicators as training samples into the neural network for training to obtain a brain glioma IDH gene prediction model;

[0101] The prediction module is used to use the brain glioma IDH gene prediction model to predict whether the target brain glioma carries the IDH gene to obtain a prediction result.

[0102] Compared with the prior art, the beneficial effects of the brain glioma IDH gene prediction system based on metamaterial sensing provided by the present invention are the same as the beneficial effects of the brain glioma IDH gene prediction method based on metamaterial sensing described in the above technical solution, which will not be repeated here.

[0103] The present invention also provides an electronic device, comprising a bus, a transceiver, a memory, a processor, and a computer program stored in the memory and runnable on the processor, wherein the transceiver, the memory, and the processor are connected via the bus, and is characterized in that when the computer program is executed by the processor, the steps in the above-mentioned method for predicting IDH genes in brain gliomas based on metamaterial sensing are implemented. Compared with the prior art, the beneficial effects of the electronic device provided by the present invention are the same as the beneficial effects of the method for predicting IDH genes in brain gliomas based on metamaterial sensing described in the above-mentioned technical solution, and will not be elaborated here.

[0104] The present invention also provides a computer-readable storage medium having a computer program stored thereon, characterized in that when the computer program is executed by a processor, the steps of the above-mentioned method for predicting the IDH gene of glioma based on metamaterial sensing are implemented. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided by the present invention are the same as the beneficial effects of the method for predicting the IDH gene of glioma based on metamaterial sensing described in the above technical solution, and will not be repeated here.

[0105] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any technical solution that can be easily conceived by a person skilled in the art within the technical scope disclosed in the present invention should be included in the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be based on the scope of protection of the claims.

Claims

1. A method for predicting IDH gene in brain glioma based on metamaterial sensing, characterized in that: include: Step 1: Use terahertz metamaterial sensors with different resonant frequencies to irradiate brain gliomas with and without the IDH gene, and obtain terahertz absorption signals of the gliomas at different frequency bands; Step 2: Obtain characteristic indicators of terahertz absorption signals in different frequency bands; Step 3: Calculate the correlation between the characteristic indices of the terahertz absorption signals of brain gliomas with IDH genes at different frequency bands and the characteristic indices of the terahertz absorption signals of brain gliomas without IDH genes at different frequency bands; Step 4: Remove the terahertz absorption signals in the corresponding frequency bands whose correlation is less than a preset threshold to obtain the filtered characteristic indicators; Step 5: Input the filtered characteristic indicators as training samples into the neural network for training to obtain the IDH gene prediction model for brain glioma; Step 6: Use the brain glioma IDH gene prediction model to predict whether the target brain glioma carries the IDH gene to obtain a prediction result.

2. The method for predicting IDH gene in glioma based on metamaterial sensing according to claim 1, characterized in that: The step 3: calculating the correlation between characteristic indices of terahertz absorption signals of brain gliomas with IDH genes at different frequency bands and characteristic indices of terahertz absorption signals of brain gliomas without IDH genes at different frequency bands, including: Step 3.1: Sort the characteristic indicators of the terahertz absorption signal by frequency band to form a frequency band sequence; Step 3.2: Subtract adjacent frequency band sequences to form a specific frequency band detection sequence; Step 3.3: Calculate the specific value of the characteristic index in each frequency band using the specific frequency band detection sequence; Step 3.4: retain the frequency band sequence corresponding to the specific value within the preset range to form the terahertz absorption signal under the filtered frequency band; Step 3.5: Calculate the correlation between the characteristic index of the terahertz absorption signal of the brain glioma with the IDH gene in the post-screening frequency band and the characteristic index of the terahertz absorption signal of the brain glioma without the IDH gene in the post-screening frequency band.

3. The method for predicting IDH gene in glioma based on metamaterial sensing according to claim 2, characterized in that: In step 3.3, the calculation formula of the specific value is: Among them, Δ j (k)=|x j (k)-x j+1 (k)| represents the specific frequency band detection sequence, x j (k) represents the value of the kth characteristic index in the jth frequency band, x j+1 (k) represents the value of the kth characteristic index in the j+1th frequency band, m represents the minimum value in the specific frequency band detection sequence, M represents the maximum value in the specific frequency band detection sequence, ξ represents the preset coefficient, and θ represents the specific value of the characteristic index in the jth frequency band.

4. The method for predicting IDH gene in glioma based on metamaterial sensing according to claim 3, characterized in that: The step 3.5: calculating the correlation between the characteristic index of the terahertz absorption signal of the brain glioma with the IDH gene in the frequency band after screening and the characteristic index of the terahertz absorption signal of the brain glioma without the IDH gene in the frequency band after screening, includes: Step 3.5.1: Calculate the similarity between the characteristic indices of the terahertz absorption signal of the brain glioma with the IDH gene in the post-screening frequency band and the characteristic indices of the terahertz absorption signal of the brain glioma without the IDH gene in the post-screening frequency band; Step 3.5.2: construct a similarity matrix using the similarity; Step 3.5.3: Get the eigenvalues ​​of the similarity matrix; Step 3.5.4: Calculate the correlation corresponding to each feature indicator based on the eigenvalues ​​of the similarity matrix.

5. The method for predicting IDH gene in glioma based on metamaterial sensing according to claim 4, characterized in that: In step 3.5.1, a similarity calculation formula is used to calculate the similarity between the characteristic index of the terahertz absorption signal of the brain glioma with the IDH gene in the post-screening frequency band and the characteristic index of the terahertz absorption signal of the brain glioma without the IDH gene in the post-screening frequency band; wherein, the similarity calculation formula is: Among them, r ij represents the similarity between the characteristic index with IDH gene in the i-th frequency band and the characteristic index without IDH gene in the j-th frequency band, x ki represents the kth characteristic index under the i-th frequency band, Represents the mean value of the characteristic index under the i-th frequency band, y kj represents the kth characteristic index under the jth frequency band, Represents the mean value of the characteristic index in the jth frequency band.

6. The method for predicting IDH gene in glioma based on metamaterial sensing according to claim 5, characterized in that: In step 3.5.2, the similarity matrix constructed is: Among them, r 11 represents the similarity between the characteristic index with IDH gene in the first frequency band and the characteristic index without IDH gene in the first frequency band, r nm It represents the similarity between the characteristic index with IDH gene in the nth frequency band and the characteristic index without IDH gene in the nth frequency band.

7. The method for predicting IDH gene in glioma based on metamaterial sensing according to claim 6, characterized in that: The step 3.5.4: calculating the correlation corresponding to each characteristic indicator based on the eigenvalue of the similarity matrix, includes: Obtain the eigenvalues ​​of the similarity matrix and arrange them from large to small to form a descending sequence. Calculate the correlation corresponding to each characteristic indicator based on the descending sequence. The correlation calculation formula is: in, Indicates the correlation corresponding to the i-th feature index, λ i represents the i-th value in the descending sequence, λ k Represents the kth value in a descending sequence.

8. The method for predicting IDH gene in glioma based on metamaterial sensing according to claim 1, characterized in that: Step 5: Inputting the filtered characteristic indicators as training samples into a neural network for training to obtain a brain glioma IDH gene prediction model, including: The filtered feature indicators are used as training samples to input into the BP neural network for training to obtain the IDH gene prediction model for brain glioma; wherein, the loss function in the training process is: in, Indicates the loss value, y indicates the status of the IDH gene. When y is 1, it is positive, and when y is 0, it is negative. Indicates the label predicted by the BP neural network.

9. A brain glioma IDH gene prediction system based on metamaterial sensing, characterized in that: include: A signal acquisition module is used to irradiate brain gliomas with and without the IDH gene using terahertz metamaterial sensors with different resonance frequencies, and to acquire terahertz absorption signals of the brain gliomas at different frequency bands; A characteristic index acquisition module is used to obtain characteristic indicators of terahertz absorption signals in different frequency bands; A correlation calculation module is used to calculate the correlation between characteristic indicators of terahertz absorption signals of brain gliomas with IDH genes in different frequency bands and characteristic indicators of terahertz absorption signals of brain gliomas without IDH genes in different frequency bands; A screening module is used to remove the terahertz absorption signals in the corresponding frequency band whose correlation is less than a preset threshold, and obtain the filtered characteristic indicators; A training module is used to input the screened characteristic indicators as training samples into the neural network for training to obtain a brain glioma IDH gene prediction model; The prediction module is used to use the brain glioma IDH gene prediction model to predict whether the target brain glioma carries the IDH gene to obtain a prediction result.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method for predicting IDH gene in brain glioma based on metamaterial sensing are implemented as described in any one of claims 1 to 8.

Images