A method and system for diagnosing and predicting prognosis of glioma based on LOXL1 and LOXL2

Abstract

The application provides a method and system for diagnosing and predicting prognosis of glioma based on LOXL1 and LOXL2. The method for diagnosing glioma comprises the following steps: obtaining expression level data of LOXL1 and LOXL2 of a subject; inputting the expression level data of LOXL1 and LOXL2 into a constructed classification model; outputting a classification result based on the classification model, obtaining a classification result of whether the subject has glioma, or obtaining a classification result of whether the subject does not have glioma. The application provides a brand-new strategy for diagnosis and prognosis of glioma.

Description

Technical Field

[0001] This invention belongs to the field of intelligent medical care, specifically relating to a method and system for diagnosing gliomas and predicting their prognosis based on LOXL1 and LOXL2. Background Technology

[0002] Gliomas, especially their most aggressive subtype glioblastoma (GBM), are the malignant tumors of the central nervous system with the worst prognosis. Currently, the standard clinical pathway relies on a multimodal comprehensive treatment framework: based on maximally safe surgical resection, followed by concurrent chemoradiotherapy (radiotherapy combined with temozolomide) and adjuvant chemoradiotherapy with temozolomide (i.e., the Stupp regimen). Although this regimen establishes a treatment benchmark, its efficacy has a significant ceiling; the vast majority of patients will eventually relapse, and long-term survival is extremely low.

[0003] At the level of diagnostic and treatment technology, the existing system faces a dual bottleneck. First, in terms of diagnosis and assessment, there are limitations in both "invasive" and "non-invasive" methods. Histopathological examination is the gold standard for diagnosis, but its invasive nature carries inherent risks and is not suitable for early screening and dynamic monitoring of diseases. Meanwhile, magnetic resonance imaging (MRI), as a primary non-invasive assessment tool, struggles to reliably distinguish between true tumor progression and treatment-related changes (such as pseudoprogression and radiation necrosis) in post-treatment monitoring, leading to uncertainty in clinical decision-making.

[0004] Secondly, there is a significant gap between theory and practice in the clinical translation of novel monitoring technologies. Liquid biopsy technology, which uses circulating tumor DNA and exosomes as analytes, theoretically offers the possibility of non-invasive, dynamic monitoring of tumor genomic information. However, in the application scenario of glioma, this technology is limited by low biomarker release concentrations due to the blood-brain barrier, insufficient detection sensitivity, and the lack of standardized operation and analysis procedures. It is still in the exploratory stage and has not yet been integrated into routine clinical practice.

[0005] Therefore, the current clinical management of gliomas urgently needs a new biomarker detection scheme that can take into account non-invasiveness, high sensitivity and high specificity, in order to break through the technical bottlenecks in the entire chain from early diagnosis, efficacy evaluation to recurrence monitoring. Summary of the Invention

[0006] In view of this, in order to overcome the shortcomings of the prior art, the present invention is proposed.

[0007] The first aspect of this invention provides a method for diagnosing gliomas, the method being computer-based and specifically comprising: Obtain expression level data of LOXL1 and LOXL2 from the subjects; The expression level data of LOXL1 and LOXL2 are input into the constructed classification model; Based on the classification model, the classification results are output to determine whether the subject has glioma or not.

[0008] In some implementations, the classification model is constructed using machine learning methods.

[0009] In some implementations, the machine learning method includes an algorithmic model developed using development tools.

[0010] In some implementations, the development tools include, but are not limited to, TensorFlow, Scikit-Learn, PyTorch, OpenNN, RapidMiner, Azure Machine Learning, Apache Mahout, Shogun, KNIME, VertexAI, H2Oai, Anaconda, Keras, Tableau, Fast.ai, Catalyst, Amazon ML, MLJAR, and Spell.

[0011] In some implementations, the algorithm model includes, but is not limited to, linear regression model, logistic regression model, Lasso regression model, Ridge regression model, linear discriminant analysis model, nearest neighbor model, decision tree model, perceptron model, neural network model, support vector machine model, Naive Bayes model, AdaBoost model, GBDT model, XGBoost model, LightGBM model, CatBoost model, and random forest model.

[0012] In some implementations, the algorithm model is selected from logistic regression models.

[0013] In some implementation schemes, the constant term and regression coefficients of the logistic regression model can be automatically determined using machine learning methods.

[0014] In some implementations, the sample source for the expression level data of LOXL1 and LOXL2 is blood.

[0015] In some implementations, the blood includes, but is not limited to, peripheral blood, whole blood, venous blood, or arterial blood.

[0016] In some implementations, the peripheral blood includes, but is not limited to, peripheral blood serum and peripheral blood plasma.

[0017] In some implementations, the blood is selected from peripheral blood serum.

[0018] A second aspect of the present invention provides a method for diagnosing gliomas, the method being computer-based and specifically comprising: Obtain expression level data of LOXL1 and LOXL2 from the subjects; Based on the expression level data of LOXL1 and LOXL2 in the subjects, classification prediction is performed to obtain the classification result of whether the subjects have glioma or not. If the expression levels of LOXL1 and LOXL2 are both higher than their respective thresholds, a classification result is obtained indicating that the subject has glioma; otherwise, a classification result is obtained indicating that the subject does not have glioma.

[0019] In some implementations, the thresholds for the expression level data of LOXL1 and LOXL2 are obtained by the following method: ROC curves for diagnosing gliomas were plotted using the expression level data of LOXL1 and LOXL2, respectively. The optimal AUC value corresponding to each ROC curve was calculated. The LOXL1 expression level data corresponding to the optimal AUC value is the expression level threshold of LOXL1, and the LOXL2 expression level data corresponding to the optimal AUC value is the expression level threshold of LOXL2.

[0020] In some implementations, the sample source for the expression level data of LOXL1 and LOXL2 is blood.

[0021] In some implementations, the blood includes, but is not limited to, peripheral blood, whole blood, venous blood, or arterial blood.

[0022] In some implementations, the peripheral blood includes, but is not limited to, peripheral blood serum and peripheral blood plasma.

[0023] In some implementations, the blood is selected from peripheral blood serum.

[0024] A third aspect of the present invention provides a method for predicting the prognosis of gliomas, the method being computer-based and specifically comprising: Obtain expression level data of LOXL1 and LOXL2 from the subjects; The expression level data of LOXL1 and LOXL2 are input into the constructed classification model; Based on the classification model, the classification results are output, either yielding a long-term subject survival result or a short-term subject survival result.

[0025] In some implementations, the classification model is constructed using machine learning methods.

[0026] In some implementations, the machine learning method includes an algorithmic model developed using development tools.

[0027] In some implementations, the development tools include, but are not limited to, TensorFlow, Scikit-Learn, PyTorch, OpenNN, RapidMiner, Azure Machine Learning, Apache Mahout, Shogun, KNIME, VertexAI, H2Oai, Anaconda, Keras, Tableau, Fast.ai, Catalyst, Amazon ML, MLJAR, and Spell.

[0028] In some implementations, the algorithm model includes, but is not limited to, linear regression model, logistic regression model, Lasso regression model, Ridge regression model, linear discriminant analysis model, nearest neighbor model, decision tree model, perceptron model, neural network model, support vector machine model, Naive Bayes model, AdaBoost model, GBDT model, XGBoost model, LightGBM model, CatBoost model, and random forest model.

[0029] In some implementations, the algorithm model is selected from logistic regression models.

[0030] In some implementation schemes, the constant term and regression coefficients of the logistic regression model can be automatically determined using machine learning methods.

[0031] In some implementations, the sample source for the expression level data of LOXL1 and LOXL2 is blood.

[0032] In some implementations, the blood includes, but is not limited to, peripheral blood, whole blood, venous blood, or arterial blood.

[0033] In some implementations, the peripheral blood includes, but is not limited to, peripheral blood serum and peripheral blood plasma.

[0034] In some implementations, the blood is selected from peripheral blood serum.

[0035] A fourth aspect of the present invention provides a method for predicting the prognosis of gliomas, the method being computer-based and specifically comprising: Obtain expression level data of LOXL1 and LOXL2 from the subjects; Classification prediction is performed based on the expression level data of LOXL1 and LOXL2 of the subjects to obtain classification results of subjects with long survival or subjects with short survival. If the expression levels of LOXL1 and LOXL2 are both higher than their respective thresholds, a classification result of short subject survival is obtained; otherwise, a classification result of long subject survival is obtained.

[0036] In some implementations, the thresholds for the expression level data of LOXL1 and LOXL2 are obtained by the following method: ROC curves predicting glioma prognosis were plotted using LOXL1 and LOXL2 expression level data, respectively. The optimal AUC value corresponding to each ROC curve was calculated. The LOXL1 expression level data corresponding to the optimal AUC value is the LOXL1 expression level threshold, and the LOXL2 expression level data corresponding to the optimal AUC value is the LOXL2 expression level threshold.

[0037] In some implementations, the sample source for the expression level data of LOXL1 and LOXL2 is blood.

[0038] In some implementations, the blood includes, but is not limited to, peripheral blood, whole blood, venous blood, or arterial blood.

[0039] In some implementations, the peripheral blood includes, but is not limited to, peripheral blood serum and peripheral blood plasma.

[0040] In some implementations, the blood is selected from peripheral blood serum.

[0041] The fifth aspect of the present invention provides a system for diagnosing gliomas, as described in any of the following: (1) A system for diagnosing gliomas, specifically comprising: Acquisition Unit: Acquires expression level data of LOXL1 and LOXL2 from subjects; Prediction unit: Inputs the expression level data of LOXL1 and LOXL2 into the constructed classification model; Based on the classification model, the classification results are output to obtain the classification result of whether the subject has glioma or not. Output unit: Outputs the classification results; (2) A system for diagnosing gliomas, specifically comprising: Acquisition Unit: Acquires expression level data of LOXL1 and LOXL2 from subjects; Prediction Unit: Based on the expression level data of LOXL1 and LOXL2 of the subject, classification prediction is performed to obtain the classification result of whether the subject has glioma or not. If the expression levels of LOXL1 and LOXL2 are both higher than their respective thresholds, a classification result indicating that the subject has glioma is obtained; otherwise, a classification result indicating that the subject does not have glioma is obtained. Output unit: Outputs the classification results.

[0042] The sixth aspect of this invention provides a system for predicting the prognosis of gliomas, according to any of the following: (1) A system for predicting the prognosis of glioma, characterized in that it specifically includes: Acquisition Unit: Acquires expression level data of LOXL1 and LOXL2 from subjects; Prediction unit: Inputs the expression level data of LOXL1 and LOXL2 into the constructed classification model; Based on the classification model, the classification results are output to obtain either a classification result indicating a long survival time for the subject or a classification result indicating a short survival time for the subject. Output unit: Outputs the classification results; (2) A system for predicting the prognosis of glioma, specifically comprising: Acquisition Unit: Acquires expression level data of LOXL1 and LOXL2 from subjects; Prediction Unit: Based on the expression level data of LOXL1 and LOXL2 of the subjects, classification prediction is performed to obtain classification results of subjects with long survival or subjects with short survival. If the expression levels of LOXL1 and LOXL2 are both higher than their respective thresholds, a classification result of short subject survival is obtained; otherwise, a classification result of long subject survival is obtained. Output unit: Outputs the classification results.

[0043] The seventh aspect of the present invention provides a method for constructing a classification model for diagnosing gliomas, the method comprising constructing the model using the expression levels of LOXL1 and LOXL2 in a subject as input variables.

[0044] In some implementations, the method includes constructing using machine learning methods.

[0045] In some implementations, the machine learning method includes an algorithmic model developed using development tools.

[0046] In some implementations, the development tools include, but are not limited to, TensorFlow, Scikit-Learn, PyTorch, OpenNN, RapidMiner, Azure Machine Learning, Apache Mahout, Shogun, KNIME, VertexAI, H2Oai, Anaconda, Keras, Tableau, Fast.ai, Catalyst, Amazon ML, MLJAR, and Spell.

[0047] In some implementations, the algorithm model includes, but is not limited to, linear regression model, logistic regression model, Lasso regression model, Ridge regression model, linear discriminant analysis model, nearest neighbor model, decision tree model, perceptron model, neural network model, support vector machine model, Naive Bayes model, AdaBoost model, GBDT model, XGBoost model, LightGBM model, CatBoost model, and random forest model.

[0048] In some implementations, the algorithm model is selected from logistic regression models.

[0049] The eighth aspect of the present invention provides a method for constructing a classification model for predicting the prognosis of glioma, the method comprising constructing the model using the expression levels of LOXL1 and LOXL2 in the subject as input variables.

[0050] In some implementations, the method includes constructing using machine learning methods.

[0051] In some implementations, the machine learning method includes an algorithmic model developed using development tools.

[0052] In some implementations, the development tools include, but are not limited to, TensorFlow, Scikit-Learn, PyTorch, OpenNN, RapidMiner, Azure Machine Learning, Apache Mahout, Shogun, KNIME, VertexAI, H2Oai, Anaconda, Keras, Tableau, Fast.ai, Catalyst, Amazon ML, MLJAR, and Spell.

[0053] In some implementations, the algorithm model includes, but is not limited to, linear regression model, logistic regression model, Lasso regression model, Ridge regression model, linear discriminant analysis model, nearest neighbor model, decision tree model, perceptron model, neural network model, support vector machine model, Naive Bayes model, AdaBoost model, GBDT model, XGBoost model, LightGBM model, CatBoost model, and random forest model.

[0054] In some implementations, the algorithm model is selected from logistic regression models.

[0055] The ninth aspect of the present invention provides a computer device / program product, including a memory and a processor, wherein the memory is used to store program instructions, and the processor is used to invoke the program instructions, wherein when the program instructions are executed, they are used to perform the method for diagnosing glioma as described in the first or second aspect of the present invention, or the method for predicting the prognosis of glioma as described in the third or fourth aspect of the present invention.

[0056] The tenth aspect of this invention provides any of the following applications: (1) Application of reagents for detecting subjects LOXL1 and LOXL2 in the preparation of products for diagnosing glioma; (2) Application of reagents for detecting subjects LOXL1 and LOXL2 in the preparation of products for predicting the prognosis of glioma; In some implementations, the reagents include those for detecting the expression levels of biomarker proteins.

[0057] In some embodiments, the reagent for detecting the expression level of the biomarker protein includes antibodies, antibody fragments, high-affinity polymers, peptides, peptide mimics, and nucleic acid aptamers that specifically bind to the full length of the protein encoded by the biomarker or a fragment thereof.

[0058] The advantages and beneficial effects of this invention are as follows: This invention provides a method and system for diagnosing gliomas and predicting their prognosis based on LOXL1 and LOXL2. The method of this invention allows for non-invasive diagnosis of gliomas, dynamic monitoring of recurrence, and prediction of patient prognosis. Furthermore, the combined use of LOXL1 and LOXL2 improves diagnostic and predictive efficacy. This invention has broad prospects for clinical translation. Attached Figure Description

[0059] Figure 1 This is a schematic flowchart of a method for diagnosing glioma provided in an embodiment of the present invention; Figure 2 This is a graph showing the differential expression of LOXL1, LOXL2, LOXL3, and LOXL4 in tumor tissue and adjacent normal tissue of patients with glioblastoma. Figure 3 The graph shows that serum LOXL1 levels were significantly higher in patients with low-grade glioma (LGG) and high-grade glioma (GBM) than in healthy individuals. Figure 4 The results show that serum LOXL2 levels in patients with low-grade glioma (LGG) are significantly higher than those in healthy individuals, and serum LOXL2 levels in patients with high-grade glioma (GBM) are significantly higher than those in LGG patients and healthy individuals. Figure 5The results show that serum LOXL3 levels in patients with high-grade glioma (GBM) are significantly higher than in healthy individuals. The results also show that there is no statistically significant difference in serum LOXL3 levels between patients with low-grade glioma (LGG) and healthy individuals. Figure 6 The graph shows that there was no statistically significant difference in serum LOXL4 levels between patients with high-grade glioma (GBM) and those with low-grade glioma (LGG) and healthy individuals. Figure 7 This is the ROC curve for LOXL1 in diagnosing glioma; Figure 8 This is the ROC curve for LOXL2 in diagnosing glioma; Figure 9 The ROC curve for predicting glioma patients using serum LOXL1 in a screening set; Figure 10 This is the ROC curve of the validation set serum LOXL1 for predicting glioma patients; Figure 11 The ROC curve for predicting glioma patients using serum LOXL2 in a screening set; Figure 12 This is the ROC curve of the validation set serum LOXL2 for predicting glioma patients; Figure 13 The ROC curves for screening serum LOXL1 and LOXL2 combined to predict glioma patients are shown. Figure 14 This is the ROC curve of the combined prediction of LOXL1 and LOXL2 in serum for glioma patients; Figure 15 This is a schematic flowchart of a method for diagnosing glioma provided in an embodiment of the present invention; Figure 16 The ROC curves show the diagnostic effects of high LOXL1 and high LOXL2 expression on gliomas in the population. Figure 17 This is a schematic flowchart of a method for predicting the prognosis of glioma provided in an embodiment of the present invention; Figure 18 This is the ROC curve for predicting the survival time of GBM patients after treatment using serum LOXL1. Figure 19 This is the ROC curve for predicting the survival time of GBM patients after treatment using serum LOXL2. Figure 20 This is the ROC curve for predicting the survival time of GBM patients after treatment using serum LOXL3. Figure 21 This is the ROC curve for predicting the survival time of GBM patients after treatment using serum LOXL4. Figure 22 It is a ROC curve for screening serum LOXL1 to predict the survival time of GBM patients after treatment; Figure 23 This is the ROC curve for validating serum LOXL1 levels to predict the survival time of GBM patients after treatment; Figure 24 It is a ROC curve for screening serum LOXL2 to predict the survival time of GBM patients after treatment; Figure 25 This is the ROC curve for validating serum LOXL2 levels to predict the survival time of GBM patients after treatment; Figure 26 It is a ROC curve for screening serum LOXL1 combined with LOXL2 to predict the survival time of GBM patients after treatment; Figure 27 This is a validation ROC curve of serum LOXL1 combined with LOXL2 to predict the survival time of GBM patients after treatment; Figure 28 This is a schematic flowchart of a method for predicting the prognosis of glioma provided in this application; Figure 29 The ROC curves show how high LOXL1 and high LOXL2 expression predict the prognosis of GBM patients after treatment. Figure 30 This is a schematic diagram of a system for diagnosing gliomas provided in an embodiment of the present invention; Figure 31 This is a schematic diagram of a system for diagnosing glioma provided in an embodiment of the present invention; Figure 32 This is a schematic diagram of a system for predicting the prognosis of glioma provided in an embodiment of the present invention; Figure 33 This is a system for predicting the prognosis of glioma provided in an embodiment of the present invention; Figure 34 This is a schematic diagram of a computer device / program product provided in an embodiment of the present invention. Detailed Implementation

[0060] To enable those skilled in the art to better understand the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings.

[0061] In some of the processes described in the specification, claims, and accompanying drawings of this invention, multiple operations appearing in a specific order are included. However, it should be clearly understood that these operations may not be executed in the order they appear herein, or may be executed in parallel. The operation numbers, such as 101, 102, etc., are merely used to distinguish different operations and do not represent any execution order. Furthermore, these processes may include more or fewer operations, and these operations may be executed sequentially or in parallel. It should be noted that the terms "first," "second," etc., used herein are used to distinguish different messages, devices, modules, etc., and do not represent a sequential order, nor do they limit "first" and "second" to different types.

[0062] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0063] Figure 1 This is a schematic flowchart of a method for diagnosing glioma provided in an embodiment of the present invention, specifically including: S101: Obtain expression level data of LOXL1 and LOXL2 from the subjects.

[0064] In some implementations, LOXL1 and LOXL2 were identified by screening using the GEPIA3 online tool (https: / / gepia3.bioinfoliu.com / ) to analyze the differential expression of LOXL1, LOXL2, LOXL3, and LOXL4 in tumor tissues and adjacent normal tissues of glioblastoma patients. The results are as follows... Figure 2 As shown, in patients with glioblastoma, the expression levels of LOXL1, LOXL2, LOXL3, and LOXL4 in tumor tissue were significantly higher than those in adjacent normal tissue.

[0065] In some implementation schemes, to further analyze whether LOXL1, LOXL2, LOXL3, and LOXL4 can be used to diagnose gliomas, peripheral blood was collected from 12 patients with low-grade gliomas, 20 patients with high-grade gliomas, and 12 healthy controls at the First Affiliated Hospital of Fujian Medical University. Serum was separated, and the levels of LOXL1, LOXL2, LOXL3, and LOXL4 were detected using ELISA kits (LOXL1: ELISA kit, ml105618; LOXL2: R&D system, DY2639-05; LOXL3: ELISA kit, ml105483; LOXL4: ABclonal kit, RK06801). The results are as follows: Figures 3-6 As shown. Figure 3 The results showed that serum LOXL1 levels were significantly higher in patients with low-grade glioma (LGG) and high-grade glioma (GBM) than in healthy individuals. Figure 4 The results showed that serum LOXL2 levels in patients with low-grade glioma (LGG) were significantly higher than those in healthy individuals, while serum LOXL2 levels in patients with high-grade glioma (GBM) were significantly higher than those in both LGG patients and healthy individuals. Figure 5 The results showed that serum LOXL3 levels were significantly higher in patients with high-grade glioma (GBM) than in healthy individuals. There was no statistically significant difference in serum LOXL3 levels between patients with low-grade glioma (LGG) and healthy individuals, nor was there a statistically significant difference in serum LOXL3 levels between patients with high-grade glioma (GBM) and those with low-grade glioma (LGG). Figure 6 The results showed no statistically significant difference in serum LOXL4 levels between patients with high-grade glioma (GBM) and those with low-grade glioma (LGG) or healthy individuals.

[0066] Based on the above results, LOXL1 and LOXL2 were selected as biomarkers for diagnosing glioma.

[0067] To further validate the diagnostic efficacy of LOXL1 and LOXL2 in diagnosing gliomas, the collected patient data were processed using SPSS, ROC curves were plotted, and AUC values ​​were calculated. The results are as follows: Figures 7-8 As shown, Figure 7 The AUC value of serum LOXL1 in predicting glioma patients was 0.841; Figure 8 The AUC value of serum LOXL2 in predicting glioma patients was 0.926.

[0068] In some implementation schemes, peripheral blood was collected from patients with low-grade glioma (14 cases), patients with high-grade glioma (40 cases), and healthy controls (14 cases) at the First Affiliated Hospital of Fujian Medical University. Serum was separated, and the levels of LOXL1 and LOXL2 were detected using an ELISA kit. The patients were then randomly divided into two groups, one as a screening set and the other as a validation set.

[0069] In the screening set, there were 7 patients with low-grade gliomas, 20 patients with high-grade gliomas, and 7 healthy controls; in the validation set, there were 7 patients with low-grade gliomas, 20 patients with high-grade gliomas, and 7 healthy controls. ROC plotting was performed, and the results are as follows: Figures 9-12 As shown. Figure 9 The AUC value of serum LOXL1 in the screening set for predicting glioma patients was 0.868, with an optimal threshold of 2.377. Figure 10 The validation set showed that the AUC value of serum LOXL1 in predicting glioma patients was 0.841. Figure 11 The AUC value of serum LOXL2 in the screening set for predicting glioma patients was 0.878, with an optimal threshold of 2.303. Figure 12 The validation set showed that the AUC value of serum LOXL2 in predicting glioma patients was 0.894. Figure 13 The AUC value of serum LOXL1 and LOXL2 combined in the screening set for predicting glioma patients was 0.958, with an optimal threshold of 4.01. Figure 14 The validation set showed that the combined AUC value of serum LOXL1 and LOXL2 in predicting glioma patients was 0.931. This demonstrates that the diagnostic efficacy of combining LOXL1 and LOXL2 was significantly improved.

[0070] In some implementations, the diagnostic efficacy of combining LOXL1 with LOXL2 and LOXL3 and / or LOXL4 is lower than that of combining LOXL1 with LOXL2.

[0071] S102: Input the expression level data of LOXL1 and LOXL2 into the constructed classification model; output the classification result based on the classification model to obtain the classification result of whether the subject has glioma or not.

[0072] In this invention, the classification model refers to a mathematical model that associates the expression levels of LOXL1 and LOXL2 with the risk or likelihood of developing glioma. Any suitable mathematical method described in the prior art can be used to associate the combination of LOXL1 and LOXL2 with the disease. Those skilled in the art will have no problem selecting appropriate statistical methods to evaluate the biomarker combination of this invention and thereby obtaining a suitable mathematical algorithm.

[0073] In some implementations, the classification model is constructed using machine learning methods.

[0074] In some implementations, the machine learning method includes an algorithmic model developed using development tools.

[0075] In some implementations, the development tools include, but are not limited to, TensorFlow, Scikit-Learn, PyTorch, OpenNN, RapidMiner, Azure Machine Learning, Apache Mahout, Shogun, KNIME, VertexAI, H2Oai, Anaconda, Keras, Tableau, Fast.ai, Catalyst, Amazon ML, MLJAR, and Spell.

[0076] In some implementations, the algorithm model includes, but is not limited to, linear regression model, logistic regression model, Lasso regression model, Ridge regression model, linear discriminant analysis model, nearest neighbor model, decision tree model, perceptron model, neural network model, support vector machine model, Naive Bayes model, AdaBoost model, GBDT model, XGBoost model, LightGBM model, CatBoost model, and random forest model.

[0077] In some implementations, the algorithm model is selected from logistic regression models.

[0078] S103: Output the classification results.

[0079] Figure 15 This is a schematic flowchart of a method for diagnosing glioma provided by an embodiment of the present invention. The method is computer-based and specifically includes: S201: Obtain expression level data of LOXL1 and LOXL2 from the subjects.

[0080] In some implementations, LOXL1 and LOXL2 were identified by screening using the GEPIA3 online tool (https: / / gepia3.bioinfoliu.com / ) to analyze the differential expression of LOXL1, LOXL2, LOXL3, and LOXL4 in tumor tissues and adjacent normal tissues of glioblastoma patients. The results are as follows... Figure 2 As shown, in patients with glioblastoma, the expression levels of LOXL1, LOXL2, LOXL3, and LOXL4 in tumor tissue were significantly higher than those in adjacent normal tissue.

[0081] In some implementation schemes, peripheral blood was collected from patients with low-grade glioma (12 cases), high-grade glioma (20 cases), and healthy controls (12 cases) at the First Affiliated Hospital of Fujian Medical University. Serum was separated, and the levels of LOXL1 and LOXL2 were detected using ELISA kits (LOXL1: enzyme-linked biotechnology, ml105618; LOXL2: R&D system, DY2639-05). After SPSS processing, ROC curves were plotted, AUC values ​​were calculated, and optimal thresholds were determined. Based on the cutoff values ​​of serum LOXL1 and LOXL2, patients were divided into groups with high serum LOXL1 (>1.852 ng / ml) and low LOXL1 (≤1.852 ng / ml), and groups with high serum LOXL2 (>2.201 ng / ml) and low LOXL2 (≤2.201 ng / ml). Combining high LOXL1 and high LOXL2 expression to predict glioma diagnosis in the population yielded an AUC value of 0.971, higher than that of a single indicator. The results are as follows: Figure 16 As shown.

[0082] Therefore, LOXL1 and LOXL2 were selected as biomarkers for the diagnosis of glioma.

[0083] S202: Based on the expression level data of LOXL1 and LOXL2 of the subject, perform classification prediction to obtain the classification result of the subject having glioma or the classification result of the subject not having glioma; if the expression level data of LOXL1 and LOXL2 are both higher than their respective thresholds, the classification result of the subject having glioma is obtained; otherwise, the classification result of the subject not having glioma is obtained.

[0084] S203: Output the classification results.

[0085] Figure 17 This is a schematic flowchart of a method for predicting the prognosis of glioma provided in an embodiment of the present invention, specifically including: S301: Obtain expression level data for LOXL1 and LOXL2 in the subjects.

[0086] In some implementation schemes, to determine whether LOXL1, LOXL2, LOXL3, and LOXL4 can be used to predict glioma prognosis, collected patient data (low-grade (12 cases), high-grade glioma patients (20 cases), and healthy controls (12 cases)) were processed using SPSS, ROC curves were plotted, and AUC values ​​were calculated. The results are as follows: Figures 18-21 As shown, Figure 18 The AUC value for serum LOXL1, which predicts the duration of post-treatment survival in GBM patients, was 0.81. Figure 19 The AUC value for serum LOXL2, which predicts the duration of post-treatment survival in GBM patients, was 0.89. Figure 20 The AUC value for serum LOXL3, which predicts the duration of post-treatment survival in GBM patients, was 0.61. Figure 21 The AUC value for serum LOXL4 in predicting post-treatment survival in GBM patients was 0.61. Therefore, it can be determined that LOXL1 and LOXL2 can be used to predict glioma prognosis, while LOXL3 and LOXL4 cannot.

[0087] In some implementation schemes, patients with low-grade gliomas (14 cases), high-grade gliomas (40 cases), and healthy controls (14 cases) were collected from the First Affiliated Hospital of Fujian Medical University. Peripheral blood was collected, serum was separated, and the levels of LOXL1 and LOXL2 were detected using an ELISA kit. The patients were randomly divided into two groups: one as a screening set and the other as a validation set. In the screening set, there were 7 patients with low-grade gliomas, 20 patients with high-grade gliomas, and 7 healthy controls; in the validation set, there were 7 patients with low-grade gliomas, 20 patients with high-grade gliomas, and 7 healthy controls. Results are as follows: Figures 22-27 As shown, Figure 22 The AUC value of serum LOXL1 in the screening set for predicting post-treatment survival in GBM patients was 0.73, with an optimal threshold of 2.85. Figure 23 The validation set showed that the AUC value of serum LOXL1 in predicting post-treatment survival in GBM patients was 0.68. Figure 24 The AUC value of serum LOXL2 in the screening set for predicting post-treatment survival in GBM patients was 0.73, with an optimal threshold of 3.532. Figure 25 The validation set showed that the AUC value of serum LOXL2 in predicting post-treatment survival in GBM patients was 0.72. Figure 26 The AUC value of serum LOXL1 combined with LOXL2 in the screening set for predicting post-treatment survival (long survival: greater than or equal to 15 months, short survival: less than 15 months) in GBM patients was 0.82, with an optimal threshold of 6.186. Figure 27 The validation set showed that the AUC value of serum LOXL1 combined with LOXL2 in predicting post-treatment survival in GBM patients was 0.80. This indicates that the predictive power of combining LOXL1 and LOXL2 was improved.

[0088] In some implementations, the performance of combining LOXL1 with LOXL2 and LOXL3 and / or LOXL4 is lower than that of combining LOXL1 with LOXL2.

[0089] S302: Input the expression level data of LOXL1 and LOXL2 into the constructed classification model; output the classification result based on the classification model to obtain a classification result with a long subject survival time or a classification result with a short subject survival time; S303: Output the classification results.

[0090] Figure 28 This is a flowchart illustrating a method for predicting the prognosis of glioma provided in this application, specifically including: S401: Obtain expression level data for LOXL1 and LOXL2 in the subjects.

[0091] In some implementations, to determine whether LOXL1, LOXL2, LOXL3, and LOXL4 could be used to predict glioma prognosis, the collected patient data (12 low-grade glioma patients, 20 high-grade glioma patients, and 12 healthy controls) were processed using SPSS, ROC curves were plotted, and AUC values ​​were calculated. The results are as follows: Figures 18-21 As shown, Figure 18 The AUC value for serum LOXL1, which predicts the duration of post-treatment survival in GBM patients, was 0.81. Figure 19 The AUC value for serum LOXL2, which predicts the duration of post-treatment survival in GBM patients, was 0.89. Figure 20 The AUC value for serum LOXL3, which predicts the duration of post-treatment survival in GBM patients, was 0.61. Figure 21 The AUC value for serum LOXL4 in predicting post-treatment survival in GBM patients was 0.61. Therefore, it can be determined that LOXL1 and LOXL2 can be used to predict glioma prognosis, while LOXL3 and LOXL4 cannot.

[0092] In some implementation schemes, peripheral blood was collected from patients with low-grade glioma (12 cases), high-grade glioma (20 cases), and healthy controls (12 cases) at the First Affiliated Hospital of Fujian Medical University. Serum was separated, and the levels of LOXL1 and LOXL2 were detected using ELISA kits (LOXL1: enzyme-linked biotechnology, ml105618; LOXL2: R&D system, DY2639-05). After SPSS processing, ROC curves were plotted, AUC values ​​were calculated, and optimal thresholds were determined. Based on the cutoff values ​​of serum LOXL1 and LOXL2, GBM patients were divided into high serum LOXL1 (>2.85 ng / ml) and low LOXL1 (≤2.85 ng / ml) groups, and high serum LOXL2 (>3.532 ng / ml) and low LOXL2 (≤3.532 ng / ml) groups. The combined expression of high LOXL1 and high LOXL2 was used to predict the prognosis of GBM patients after treatment, with an AUC value of 0.94, which was higher than that of a single indicator. The results are as follows Figure 29 As shown.

[0093] S402: Based on the expression level data of LOXL1 and LOXL2 of the subjects, perform classification prediction to obtain classification results with long subject survival or classification results with short subject survival; If the expression levels of LOXL1 and LOXL2 are both higher than their respective thresholds, a classification result of short subject survival is obtained; otherwise, a classification result of long subject survival is obtained.

[0094] S403: Output classification results In some implementations, the thresholds for the expression level data of LOXL1 and LOXL2 are obtained by the following method: ROC curves predicting glioma prognosis were plotted using LOXL1 and LOXL2 expression level data, respectively. The optimal AUC value corresponding to each ROC curve was calculated. The LOXL1 expression level data corresponding to the optimal AUC value is the LOXL1 expression level threshold, and the LOXL2 expression level data corresponding to the optimal AUC value is the LOXL2 expression level threshold.

[0095] Figure 30 This is a schematic diagram of a system for diagnosing gliomas provided in an embodiment of the present invention, specifically including: 501: Acquisition Unit: Acquires expression level data of LOXL1 and LOXL2 from subjects; 502: Prediction Unit: Inputs the expression level data of LOXL1 and LOXL2 into the constructed classification model; Based on the classification model, the classification results are output to obtain the classification result of whether the subject has glioma or not. 503: Output Unit: Outputs the classification results.

[0096] Figure 31 This is a schematic diagram of a system for diagnosing gliomas provided in an embodiment of the present invention, specifically including: 601: Acquisition Unit: Acquires expression level data of LOXL1 and LOXL2 from subjects; 602: Prediction Unit: Based on the expression level data of LOXL1 and LOXL2 of the subject, perform classification prediction to obtain the classification result of whether the subject has glioma or not. If the expression levels of LOXL1 and LOXL2 are both higher than their respective thresholds, a classification result indicating that the subject has glioma is obtained; otherwise, a classification result indicating that the subject does not have glioma is obtained. 603: Output Unit: Outputs the classification results.

[0097] Figure 32This is a schematic diagram of a system for predicting the prognosis of glioma provided in an embodiment of the present invention, specifically including: 701: Acquisition Unit: Acquires expression level data of LOXL1 and LOXL2 from subjects; 702: Prediction Unit: Inputs the expression level data of LOXL1 and LOXL2 into the constructed classification model; Based on the classification model, the classification results are output to obtain either a classification result indicating a long survival time for the subject or a classification result indicating a short survival time for the subject. 703: Output Unit: Outputs the classification results.

[0098] Figure 33 The system for predicting the prognosis of glioma provided in this embodiment of the invention specifically includes: 801: Acquisition Unit: Acquires expression level data of LOXL1 and LOXL2 from subjects; 802: Prediction Unit: Based on the expression level data of LOXL1 and LOXL2 of the subjects, perform classification prediction to obtain classification results with long subject survival or classification results with short subject survival; If the expression levels of LOXL1 and LOXL2 are both higher than their respective thresholds, a classification result of short subject survival is obtained; otherwise, a classification result of long subject survival is obtained. 803: Output Unit: Outputs classification results.

[0099] Figure 34 This is a schematic diagram of a computer device / program product provided in an embodiment of the present invention.

[0100] In some of the processes described in the specification, claims, and accompanying drawings of this invention, multiple operations appearing in a specific order are included. However, it should be clearly understood that these operations may not be executed in the order they appear herein, or may be executed in parallel. The operation numbers, such as 101, 102, etc., are merely used to distinguish different operations and do not represent any execution order. Furthermore, these processes may include more or fewer operations, and these operations may be executed sequentially or in parallel. It should be noted that the descriptions such as "first," "second," etc., in this document are used to distinguish different messages, devices, modules, etc., and do not represent a sequential order, nor do they limit "first" and "second" to different types.

[0101] In the several embodiments provided in this application, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces, devices, or units, and may be electrical, mechanical, or other forms.

[0102] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0103] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0104] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, which may include: read-only memory (ROM), random access memory (RAM), magnetic disk or optical disk, etc.

[0105] Those skilled in the art will understand that all or part of the steps in the methods of the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, such as a read-only memory, a disk, or an optical disk.

[0106] For those skilled in the art, based on the ideas of the embodiments of the present invention, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of the present invention.

[0107] The above description of the embodiments is only for understanding the method and core ideas of the present invention. It should be noted that those skilled in the art can make various improvements and modifications to the present invention without departing from the principles of the invention, and these improvements and modifications will also fall within the protection scope of the claims of the present invention.

Claims

1. A method for diagnosing glioma, characterized in that, The method is computer-based and specifically includes: Obtain expression level data of LOXL1 and LOXL2 from the subjects; The expression level data of LOXL1 and LOXL2 are input into the constructed classification model; Based on the classification model, the classification results are output to determine whether the subject has glioma or not.

2. The method according to claim 1, characterized in that, The classification model is constructed using machine learning methods; Optionally, the machine learning method includes an algorithmic model developed using development tools; Optionally, the development tools include TensorFlow, Scikit-Learn, PyTorch, OpenNN, RapidMiner, Azure Machine Learning, Apache Mahout, Shogun, KNIME, Vertex AI, H2Oai, Anaconda, Keras, Tableau, Fast.ai, Catalyst, Amazon ML, MLJAR, and Spell. Optionally, the algorithm model includes linear regression model, logistic regression model, Lasso regression model, Ridge regression model, linear discriminant analysis model, nearest neighbor model, decision tree model, perceptron model, neural network model, support vector machine model, Naive Bayes model, AdaBoost model, GBDT model, XGBoost model, LightGBM model, CatBoost model, and random forest model; Optionally, the algorithm model is selected from the logistic regression model.

3. A method for diagnosing glioma, characterized in that, The method is computer-based and specifically includes: Obtain expression level data of LOXL1 and LOXL2 from the subjects; Based on the expression level data of LOXL1 and LOXL2 in the subjects, classification prediction is performed to obtain the classification result of whether the subjects have glioma or not. If the expression levels of LOXL1 and LOXL2 are both higher than their respective thresholds, a classification result indicating that the subject has glioma is obtained; otherwise, a classification result indicating that the subject does not have glioma is obtained. Optionally, the thresholds for the expression level data of LOXL1 and LOXL2 are obtained by the following method: ROC curves for diagnosing gliomas were plotted using the expression level data of LOXL1 and LOXL2, respectively. The optimal AUC value corresponding to each ROC curve was calculated. The LOXL1 expression level data corresponding to the optimal AUC value is the expression level threshold of LOXL1, and the LOXL2 expression level data corresponding to the optimal AUC value is the expression level threshold of LOXL2.

4. A method for predicting the prognosis of glioma, characterized in that, The method is computer-based and specifically includes: Obtain expression level data of LOXL1 and LOXL2 from the subjects; The expression level data of LOXL1 and LOXL2 are input into the constructed classification model; Based on the classification model, the classification results are output, either yielding a long-term subject survival result or a short-term subject survival result.

5. The method according to claim 4, characterized in that, The classification model is constructed using machine learning methods; Optionally, the machine learning method includes an algorithmic model developed using development tools; Optionally, the development tools include TensorFlow, Scikit-Learn, PyTorch, OpenNN, RapidMiner, Azure Machine Learning, Apache Mahout, Shogun, KNIME, Vertex AI, H2Oai, Anaconda, Keras, Tableau, Fast.ai, Catalyst, Amazon ML, MLJAR, and Spell. Optionally, the algorithm model includes linear regression model, logistic regression model, Lasso regression model, Ridge regression model, linear discriminant analysis model, nearest neighbor model, decision tree model, perceptron model, neural network model, support vector machine model, Naive Bayes model, AdaBoost model, GBDT model, XGBoost model, LightGBM model, CatBoost model, and random forest model; Optionally, the algorithm model is selected from the logistic regression model.

6. A method for predicting the prognosis of glioma, characterized in that, The method is computer-based and specifically includes: Obtain expression level data of LOXL1 and LOXL2 from the subjects; Classification prediction is performed based on the expression level data of LOXL1 and LOXL2 of the subjects to obtain classification results of subjects with long survival or subjects with short survival. If the expression levels of LOXL1 and LOXL2 are both higher than their respective thresholds, a classification result of short subject survival is obtained; otherwise, a classification result of long subject survival is obtained. Optionally, the thresholds for the expression level data of LOXL1 and LOXL2 are obtained by the following method: ROC curves predicting glioma prognosis were plotted using LOXL1 and LOXL2 expression level data, respectively. The optimal AUC value corresponding to each ROC curve was calculated. The LOXL1 expression level data corresponding to the optimal AUC value is the LOXL1 expression level threshold, and the LOXL2 expression level data corresponding to the optimal AUC value is the LOXL2 expression level threshold.

7. Any of the following systems for diagnosing gliomas: (1) A system for diagnosing glioma, characterized in that, Specifically, it includes: Acquisition Unit: Acquires expression level data of LOXL1 and LOXL2 from subjects; Prediction unit: Inputs the expression level data of LOXL1 and LOXL2 into the constructed classification model; Based on the classification model, the classification results are output to obtain the classification result of whether the subject has glioma or not. Output unit: Outputs the classification results; (2) A system for diagnosing gliomas, characterized in that it specifically comprises: Acquisition Unit: Acquires expression level data of LOXL1 and LOXL2 from subjects; Prediction Unit: Based on the expression level data of LOXL1 and LOXL2 of the subject, classification prediction is performed to obtain the classification result of whether the subject has glioma or not. If the expression levels of LOXL1 and LOXL2 are both higher than their respective thresholds, a classification result indicating that the subject has glioma is obtained; otherwise, a classification result indicating that the subject does not have glioma is obtained. Output unit: Outputs the classification results.

8. Any of the following systems for predicting the prognosis of glioma: (1) A system for predicting the prognosis of glioma, characterized in that, Specifically, it includes: Acquisition Unit: Acquires expression level data of LOXL1 and LOXL2 from subjects; Prediction unit: Inputs the expression level data of LOXL1 and LOXL2 into the constructed classification model; Based on the classification model, the classification results are output to obtain either a classification result indicating a long survival time for the subject or a classification result indicating a short survival time for the subject. Output unit: Outputs the classification results; (2) A system for predicting the prognosis of glioma, characterized in that it specifically includes: Acquisition Unit: Acquires expression level data of LOXL1 and LOXL2 from subjects; Prediction Unit: Based on the expression level data of LOXL1 and LOXL2 of the subjects, classification prediction is performed to obtain classification results of subjects with long survival or subjects with short survival. If the expression levels of LOXL1 and LOXL2 are both higher than their respective thresholds, a classification result of short subject survival is obtained; otherwise, a classification result of long subject survival is obtained. Output unit: Outputs the classification results.

9. A computer device / program product, characterized in that, The device includes a memory and a processor, wherein the memory is used to store program instructions, and the processor is used to invoke the program instructions, which, when executed, are used to perform the method for diagnosing glioma as described in any one of claims 1-3, or the method for predicting the prognosis of glioma as described in any one of claims 4-6.

10. Any of the following applications: (1) Application of reagents for detecting subjects LOXL1 and LOXL2 in the preparation of products for diagnosing glioma; (2) Application of reagents for detecting subjects LOXL1 and LOXL2 in the preparation of products for predicting the prognosis of glioma; Optionally, the reagents include reagents for detecting the expression level of biomarker proteins; Optionally, the reagent for detecting the expression level of the biomarker protein includes antibodies, antibody fragments, high-affinity polymers, peptides, peptide mimics, and nucleic acid aptamers that specifically bind to the full length of the protein encoded by the biomarker or a fragment thereof.

Images