Method and device for constructing nasopharyngeal carcinoma risk scoring model

Abstract

The embodiment of the invention relates to a nasopharyngeal carcinoma risk scoring model construction method and device. A nasopharyngeal carcinoma patient and an EBV antibody serological detection positive population are taken as target samples, and plasma samples are collected for data processing; furthermore, by performing nuclear magnetic resonance metabonomics detection on plasma specimens of a nasopharynx cancer queue and an EB virus infected control queue, multiple different components and lipid quantitative data of a hemorrhagic lipoprotein group are screened, and plasma metabolites and blood lipoprotein components are quantitatively detected; characteristic metabolites and blood and blood lipoprotein components closely related to nasopharyngeal carcinoma are screened through single-factor Logistic regression analysis, a nasopharyngeal carcinoma risk scoring model is constructed based on the characteristic metabolites and the blood and blood lipoprotein components, the nasopharyngeal carcinoma risk scoring model is combined with EBV antibody serological detection, the positive predicted value of a nasopharyngeal carcinoma patient can be remarkably increased, individualized risk prediction and early warning can be achieved for high-risk groups, and the risk assessment of the nasopharyngeal carcinoma patient is improved. The incidence rate and the death rate of nasopharynx cancer are reduced, and good hygienic economic benefits are achieved.

Description

Technical Field

[0001] The present invention relates to the field of medical testing technology, and in particular to a method and device for constructing a nasopharyngeal carcinoma risk scoring model. Background Art

[0002] The main causes of NPC include genetic susceptibility, dietary factors and Epstein-Barr virus (EBV) infection. The research on its pathogenesis includes gene mutation, metabolomics, proteomics, genomics and other aspects. Despite this, it is still difficult to fully understand the occurrence and development of NPC, which brings difficulties to its treatment. In an observational trial involving 20,174 middle-aged men aged 40-62, the role of plasma EBV DNA detection in NPC screening was prospectively evaluated. The positive predictive value (PPV) of NPC in high-risk populations with two consecutive positive EBV DNA tests was only 11%. In another large cluster randomized trial (NCT-00941538), the Logistic regression model for predicting NPC using the combined score of viral capsid antigen and Epstein-Barr virus nuclear antigen 1 (VCA / EBNA1) immunoglobulin A antibody showed a specificity of 97% and a positive predictive value of 4.4%. In a report, it was pointed out that the symptoms of early nasopharyngeal carcinoma patients are hidden and atypical, which is extremely difficult to detect. Most patients are already in the local middle and late stages when diagnosed. At present, there is a lack of specific and cost-effective screening tools for early screening of nasopharyngeal carcinoma, and it is impossible to effectively screen out early patients. The current difficulties in nasopharyngeal carcinoma screening: 1. The accuracy of screening methods is poor: Epstein-Barr virus-related antibodies or EBV DNA detection are not effective for screening early patients; nasopharyngeal endoscopy and nuclear magnetic resonance imaging are difficult to detect early mucosal lesions, which can easily lead to missed diagnosis; 2. The screening standards are not unified: each unit and each center lacks unified standards in terms of detection machines, reagents, methods and other technical aspects, resulting in differences and even inaccuracies in the data of each center; 3. The content of the screening plan is rough: some areas only screen according to age stratification, which can easily lead to missed diagnosis and excessive examination. Due to the huge population mobility in various provinces and cities in China, a large number of high-risk people are spread all over the country rather than limited to high-incidence areas, and there is a lack of screening for this part of the high-risk population; 4. The screening form is inefficient and single: mainly setting up on-site screening points, limited population coverage, huge cost, low efficiency, and poor results. Failure to establish a suitable high-risk disease warning model has not only led to very limited population coverage, but also some non-high-risk groups may have been over-screened; 5. Difficulty in updating screening data: The degree of digitization is low, and big data processing is impossible, making it difficult to efficiently achieve data sharing and real-time updates, and long-term follow-up data is also difficult to update in real time. Summary of the invention

[0003] In view of the above-mentioned defects, an embodiment of the present invention discloses a method and device for constructing a nasopharyngeal carcinoma risk scoring model, which can improve the positive predictive value and diagnostic accuracy of nasopharyngeal carcinoma screening.

[0004] The first aspect of the embodiment of the present invention discloses a method for constructing a nasopharyngeal carcinoma risk scoring model, comprising:

[0005] Selecting several nasopharyngeal carcinoma patients and several people with positive EBV antibody serological test results as target samples, and recording sample information of the target samples;

[0006] Collecting plasma samples of each target sample separately, and processing the plasma samples to obtain quantitative data of various metabolites and different components of lipoproteins in the plasma samples;

[0007] Based on each component of the lipoprotein, a regression analysis is performed to obtain correlation values ​​between different elements in the component and nasopharyngeal carcinoma, and the elements whose correlation values ​​meet preset conditions are used as modeling data;

[0008] A risk scoring model is constructed based on the modeling data to calculate the nasopharyngeal carcinoma risk score of each target sample, and at least one risk threshold is set to delineate different risk areas.

[0009] As an optional embodiment, in the first aspect of the embodiments of the present invention, in the processing of the plasma sample to obtain multiple different components of the lipoprotein group and lipid quantitative data in the plasma sample, the plasma sample is processed based on the partial least squares regression model prediction algorithm to obtain multiple different components of the lipoprotein group and lipid quantitative data.

[0010] As an optional implementation, in the first aspect of the embodiment of the present invention, in the regression analysis performed on each component of the lipoprotein to obtain the correlation value between different elements in the component and nasopharyngeal carcinoma, the correlation value is obtained by Logistic regression analysis.

[0011] As an optional implementation, in the first aspect of the embodiments of the present invention, the correlation value includes a risk odds ratio and P, and the preset condition is that the risk odds ratio is greater than 1 and P is less than 0.05.

[0012] As an optional implementation, in the first aspect of the embodiment of the present invention, the expression of the risk scoring model is: where x is the expression value of each risk factor, β is the risk odds ratio, and β is the correlation value.

[0013] As an optional implementation manner, in the first aspect of the embodiment of the present invention, setting at least one risk threshold to delineate different risk areas includes:

[0014] Three risk thresholds are set at 25%, 50%, and 75%;

[0015] The risk threshold is divided into four quantile ranges of 0%-25%, 25%-50%, 50%-75%, and 75%-100%. Four risk levels are set according to the four ranges, namely high-risk group, medium-high risk group, medium-low risk group, and low risk group.

[0016] A second aspect of an embodiment of the present invention discloses a device for constructing a nasopharyngeal carcinoma risk scoring model, characterized in that it includes:

[0017] Sample collection module: used to select several nasopharyngeal carcinoma patients and several people with positive EBV antibody serological test as target samples, and record the sample information of the target samples;

[0018] Data acquisition module: used to collect plasma samples of each target sample respectively, and process the plasma samples to obtain quantitative data of various metabolites and different components of lipoproteins in the plasma samples;

[0019] Data modeling module: used for performing regression analysis based on each component of the lipoprotein to obtain correlation values ​​between different elements in the component and nasopharyngeal carcinoma, and taking the elements whose correlation values ​​meet preset conditions as modeling data;

[0020] Risk classification module: used to construct a risk scoring model based on the modeling data to calculate the nasopharyngeal carcinoma risk score of each target sample, and set at least one risk threshold to delineate different risk areas.

[0021] As an optional embodiment, in the second aspect of the embodiments of the present invention, in the processing of the plasma sample to obtain the various different components of the lipoprotein group and lipid quantitative data in the plasma sample, the plasma sample is processed based on the partial least squares regression model prediction algorithm to obtain the various different components of the lipoprotein group and lipid quantitative data.

[0022] As an optional implementation, in the second aspect of the embodiment of the present invention, in the regression analysis performed on each component of the lipoprotein to obtain the correlation value between different elements in the component and nasopharyngeal carcinoma, the correlation value is obtained by Logistic regression analysis.

[0023] As an optional implementation, in the second aspect of the embodiment of the present invention, the correlation value includes a risk odds ratio and P, and the preset condition is that the risk odds ratio is greater than 1 and P is less than 0.05, and P is a statistical indicator.

[0024] As an optional implementation, in the second aspect of the embodiment of the present invention, the expression of the risk scoring model is: where x is the expression value of each risk factor, β is the risk odds ratio, and β is the correlation value.

[0025] As an optional implementation, in the second aspect of the embodiment of the present invention, setting at least one risk threshold to delineate different risk areas includes:

[0026] Three risk thresholds are set at 25%, 50%, and 75%;

[0027] The risk threshold is divided into four quantile ranges of 0%-25%, 25%-50%, 50%-75%, and 75%-100%. Four risk levels are set according to the four ranges, namely high-risk group, medium-high risk group, medium-low risk group, and low risk group.

[0028] A third aspect of an embodiment of the present invention discloses an electronic device, comprising: a memory storing executable program code; a processor coupled to the memory; the processor calls the executable program code stored in the memory to execute the nasopharyngeal carcinoma risk scoring model construction method disclosed in the first aspect of the embodiment of the present invention.

[0029] A fourth aspect of an embodiment of the present invention discloses a computer-readable storage medium storing a computer program, wherein the computer program enables a computer to execute the method for constructing a nasopharyngeal carcinoma risk scoring model disclosed in the first aspect of an embodiment of the present invention.

[0030] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0031] In the embodiment of the present invention, plasma samples are collected from both nasopharyngeal carcinoma patients and people who are positive for EBV antibody serological test as target samples for data processing, and then nuclear magnetic resonance metabolomics detection can be performed on the plasma specimens of the nasopharyngeal carcinoma cohort and the EBV-infected control cohort to screen out various different components of the blood lipoprotein group and lipid quantitative data, and the plasma metabolites and blood lipoprotein components are quantitatively detected. The characteristic metabolites and blood lipoprotein components closely related to the occurrence of nasopharyngeal carcinoma are screened out through univariate Logistics regression analysis. Based on this, a nasopharyngeal carcinoma risk scoring model is constructed. Combined with EBV antibody serological detection, it can not only significantly improve the positive predictive value of nasopharyngeal carcinoma patients, but also realize individualized risk prediction and early warning for high-risk populations, reduce the incidence and mortality of nasopharyngeal carcinoma, and have good health economic benefits. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0033] Figure 1 It is a flowchart of a method for constructing a nasopharyngeal carcinoma risk scoring model disclosed in an embodiment of the present invention;

[0034] Figure 2 It is a schematic diagram of the structure of a device for constructing a nasopharyngeal carcinoma risk scoring model provided by an embodiment of the present invention;

[0035] Figure 3 This is a schematic diagram of the first nasopharyngeal carcinoma risk score analysis result provided by an embodiment of the present invention;

[0036] Figure 4 is a schematic diagram of a second nasopharyngeal carcinoma risk score analysis result provided by an embodiment of the present invention;

[0037] Figure 5 It is a Logistics regression analysis table of the lipoprotein group provided in an embodiment of the present invention;

[0038] Figure 6 It is a structural schematic diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0039] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0040] It should be noted that the terms "first", "second", "third", "fourth", etc. in the specification and claims of the present invention are used to distinguish different objects rather than to describe a specific order. The terms "including" and "having" in the embodiments of the present invention and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device including a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0041] The embodiment of the present invention discloses a method, device, electronic device and storage medium for constructing a nasopharyngeal carcinoma risk scoring model. In the embodiment, plasma samples are collected from nasopharyngeal carcinoma patients and EBV antibody serological test positive populations as target samples for data processing, and then nuclear magnetic resonance metabolomics detection can be performed on plasma specimens of the nasopharyngeal carcinoma cohort and the EBV-infected control cohort to screen out a variety of different components of the hemolipid group and lipid quantitative data, and plasma metabolites and hemolipid components are quantitatively detected. Characteristic metabolites and hemolipid components closely related to the occurrence of nasopharyngeal carcinoma are screened out through univariate Logistics regression analysis. Based on this, a nasopharyngeal carcinoma risk scoring model is constructed. Combined with EBV antibody serological detection, it can not only significantly improve the positive predictive value of nasopharyngeal carcinoma patients, but also realize individualized risk prediction and early warning for high-risk populations, reduce the incidence and mortality of nasopharyngeal carcinoma, and have good health economic benefits.

[0042] Embodiment 1

[0043] See also Figure 1 , Figure 1 It is a flow chart of a method for constructing a nasopharyngeal carcinoma risk scoring model disclosed in an embodiment of the present invention. Among them, the execution subject of the method described in the embodiment of the present invention is an execution subject composed of software and / or hardware, and the execution subject can receive relevant information by wired or / and wireless means, and can send certain instructions. Of course, it can also have certain processing functions and storage functions. The execution subject can control multiple devices, such as a remote physical server or cloud server and related software, or it can be a local host or server and related software that performs related operations on a device placed somewhere. In some scenarios, multiple storage devices can also be controlled, and the storage devices can be placed in the same place or different places as the devices. For example Figure 1 As shown, the method for constructing the nasopharyngeal carcinoma risk scoring model includes the following steps:

[0044] 101. Select several nasopharyngeal carcinoma patients and several people with positive EBV antibody serological test results as target samples, and record sample information of the target samples.

[0045] In the embodiment, the samples selected include nasopharyngeal carcinoma patients and people with positive EBV antibody serological test. In a specific clinical trial, 297 nasopharyngeal carcinoma patients and 149 participants with positive EBV serological test were selected. The sample information includes clinical data such as gender, age, tumor stage, family history, smoking history, drinking history, etc. of each target sample.

[0046] 102. Collect plasma samples from each target sample separately, and process the plasma samples to obtain quantitative data of various metabolites and different components of lipoprotein in the plasma samples.

[0047] In order to obtain the data required for modeling, plasma samples need to be collected. The embodiment uses serum collection tubes to collect peripheral blood of all target samples according to the standardized collection process, and uses anticoagulant tubes containing sodium heparin to obtain blood samples. The centrifuge is used to separate plasma within 24 hours at 4 ° C, 2000 rpm, and 10 minutes. Each tube is filled with 1000 μL and stored at ultra-low temperature (-80 ° C) using cryopreservation technology. The plasma in the supernatant is transferred to a clean cryotube and stored at -80 ° C until nuclear magnetic resonance metabolomics analysis. In order to ensure the stability of the metabolites and maintain their original state, dry ice express service is used to maintain low temperature during transportation and transportation.

[0048] In this step, the plasma samples were processed based on the partial least squares regression model prediction algorithm to obtain various different components and lipid quantitative data of the lipoprotein group. The specific operation was first to take the plasma samples out of the refrigerator and thaw them. Then, 400 μL of each plasma sample was mixed with an equal volume of NMR lipid buffer (BrukerBiospin, Etdlingen, Germany) in a ratio of 1:1. The mixture was homogenized by vortexing for 30 seconds and then transferred to a 5mm NMR tube for analysis. The lipoprotein group includes various components, such as the total components HDL, LDL, IDL and VLDL, and 15 subcomponents sorted by number according to increasing density and decreasing size, including 4 HDL subcomponents (HDL-1 to HDL-4), 6 LDL subcomponents (LDL-1 to LDL-6) and 5 VLDL subcomponents (VLDL-1 to VLDL-5). Each subfraction was quantified for parameters such as cholesterol, free cholesterol, phospholipids, triglycerides, Apo A1, A2, B100 fractions, and 9 particle numbers. In addition, ratios such as Chol-LDL / CholHDL (low-density cholesterol / high-density cholesterol) and Apo-B100 / Apo-A1 (human apolipoprotein / recombinant protein) were calculated, resulting in a total of 112 parameters.

[0049] 103. Perform regression analysis on each component of the lipoprotein to obtain correlation values ​​between different elements in the component and nasopharyngeal carcinoma, and use the elements whose correlation values ​​meet preset conditions as modeling data.

[0050] Example By analyzing the plasma NMR metabolomics data of NPC cohort and EBV cohort, based on OPLS-DA analysis and t-test, a total of 8 marker metabolites (Glucose, Lactic acid, Pyruvic acid, Glutamine, Methionine, Acetic acid, Phenylalanine, Ornithine) and 2 subcomponents of lipoprotein particles (VLDL-5, LDL-1) were identified, and the screening conditions were: p<0.05 and VIP>1. Logistic regression analysis was performed on these marker metabolites and lipoprotein subcomponents. Correlation values ​​were obtained, which included risk odds ratio and P. Combined Figure 5 The results showed that five metabolites (glucose, lactic acid, pyruvic acid, methionine, ornithine), four components of VLDL-5 particles (VLDL-5 triglycerides, VLDL-5 cholesterol, VLDL-5 free cholesterol, VLDL-5 phospholipids) and three components of LDL-1 particles (LDL-1 triglycerides, LDL-1 phospholipids, LDL-1 apo-B) were closely related to the occurrence of nasopharyngeal carcinoma. Based on these 12 factors, closely related elements were set, that is, the preset conditions satisfied by the correlation value were that the risk odds ratio was greater than 1 and P was less than 0.05. P was the statistical indicator P value, that is, the significance level calculated according to the actual statistic. A t-test was performed on the metabolic profiles of the nasopharyngeal carcinoma patient cohort and the EBV serology-positive cohort, with a significance level of 0.05. Metabolites and lipoprotein subfractions with p values ​​< 0.05 were selected as candidate differential metabolites. OPLS-DA analysis can be used to screen differential metabolites between different groups in metabolomics data analysis. Through OPLS-DA analysis, a VIP value can be obtained for each metabolite. The larger the VIP value, the greater the contribution of the substance to distinguishing the two groups. The VIP threshold set in this study is 1, and the OR threshold is 1. The above-mentioned closely related factors meet the preset conditions and are included in the construction of the NRS risk stratification model for nasopharyngeal carcinoma.

[0051] 104. Construct a risk scoring model based on the modeling data to calculate the nasopharyngeal carcinoma risk score of each target sample, and set at least one risk threshold to define different risk areas.

[0052] In the embodiment, the expression of the risk scoring model is: Where x is the expression value of each risk factor, β is the risk odds ratio, and β is the correlation value. Based on this risk scoring model, the risk score of nasopharyngeal carcinoma can be calculated. Specifically, the calculation method of the risk score of nasopharyngeal carcinoma is: 4.934819197*Glucose-6.214608098*Lactic acid+2.593013391*Pyruvic acid+1.467643907*Methionine-2.603690186*Ornithine+1.380024748*V5TG+1.005033942*

[0053] V5CH+1.365836519*V5FC+1.180806526*V5PL+1.25846099*L1TG+

[0054] 0.548699276*L1PL+0.63180355*L1AB. Among them, V5TG: VLDL-5Triglycerides, V5CH: VLDL-5Cholesterol, V5FC: VLDL-5Free Cholesterol, V5PL: VLDL-5Phospholipids, L1TG: LDL-1Triglycerides, L1PL: LDL-1Phospholipids, L1AB: LDL-1Apo-B. Xi can be understood as a set of characteristic metabolites and lipoprotein subcomponents {Glucose, Lactic acid, Pyruvic acid, Methionine, Ornithine, V5TG, V5CH, V5FC, V5PL, L1TG, L1PL, L1AB}, x1 = Glucose, x2 = Lactic acid... Similarly, βi is the set of OR values ​​of each component in Xi. The OR values ​​corresponding to each characteristic metabolite and lipoprotein subcomponent are shown in Figure 4 ln(βi) is the natural logarithm of the OR value corresponding to each component in Xi. n represents the number of components involved in modeling, n=12, namely Glucose, Lactic acid, Pyruvic acid, Methionine, Ornithine, V5TG, V5CH, V5FC, V5PL, L1TG, L1PL, L1AB, these 12 modeling factors.

[0055] In this step, setting at least one risk threshold to delineate different risk areas includes setting three risk thresholds to 25%, 50%, and 75%, respectively; dividing the risk thresholds into four quantile ranges of 0%-25%, 25%-50%, 50%-75%, and 75%-100%, and setting four risk levels corresponding to the four ranges, namely, a high-risk group, a medium-high-risk group, a medium-low-risk group, and a low-risk group.

[0056] Receiver operating characteristic (ROC) curve analysis was performed to evaluate the MRI-based NPC risk score (NRS), which was presented differently in two different validation cohorts, e.g. Figure 3 and Figure 4 As shown, the performance of the NPC diagnostic model under different NRS thresholds was compared, and the evaluation indicators included specificity, sensitivity, positive predictive value (PPV), AUC (area under the ROC curve), Youden index, etc. To evaluate the application value of NRS in NPC screening, NRS was combined with serum EBV VCA- / EBNA1-IgA detection for the differential diagnosis of NPC, and its specificity, sensitivity PPVs, NPVs and Youden index were calculated according to the detection method. NRS with different critical values ​​was combined with EBV serological test results, and high-risk individuals with NRS> cutoff value and positive EBV antibody serological test were classified as high-risk group. In the combined cohort, the AUC result (0.841, 95% CI: 0.811-0.871) tested in the combined cohort was superior to that using the 25% quantile (0.693, 95% CI: 0.664-0.722) or 75% quantile (0.773, 95% CI: 0.739-0.808) of NRS 50% as the cutoff value. The high-risk population for NPC was divided into groups with NRS>14.95, and its specificity and sensitivity were both above 0.80 (PPV: 0.7008; NPV: 0.9271; Youden index: 0.6817).

[0057] Embodiment Combining NMR-based risk score (NRS) and EBV VCA / EBNA1-IgA serological detection, this NPC screening strategy not only has a good application prospect in identifying high-risk individuals in the EBV-infected population, but also greatly reduces the economic burden and unnecessary medical examinations of the participants. The method showed excellent cost-effectiveness and enhanced the effectiveness of secondary risk stratification. In our joint cohort, individuals with positive EBV VCA / EBNA1-IgA serological tests were identified as a potential risk group for NPC. In this risk group, the median cutoff value of NRS (N50) was used for NRS-based risk stratification. The results showed that a total of 24 people were classified as the NRS high-risk group and 178 people were classified as the NRS low-risk group. For the NRS high-risk population, different recommendations are provided for different populations. For example, it is recommended that high-risk populations quickly complete nasopharyngeal endoscopy and magnetic resonance imaging (MRI), increase the frequency of follow-up, and dynamically monitor their clinical symptoms and examination results, for example, re-evaluate the NRS risk stratification results within 3-6 months or conduct additional necessary investigations; for individuals in the NRS low-risk group, additional investigations for further NPC screening are considered unnecessary. Instead, it is recommended to regularly monitor the NRS risk stratification results every 6-12 months. If some individuals are reclassified as NRS high-risk populations during follow-up, further nasopharyngeal endoscopy or magnetic resonance imaging (MRI) examinations are recommended to screen for nasopharyngeal carcinoma (NPC). NRS combined with Epstein-Barr virus antibody serological testing significantly improves the accuracy of NPC screening, has good health economic benefits, and provides important guidance for establishing personalized follow-up strategies for high-risk populations of NPC.

[0058] Embodiment 2

[0059] See also Figure 5 , Figure 5 Schematic diagram of a device for constructing a nasopharyngeal carcinoma risk scoring model disclosed in an embodiment of the present invention. Figure 5As shown, the nasopharyngeal carcinoma risk scoring model construction device may include: a sample collection module 501, a data acquisition module 502, a data modeling module 503 and a risk classification module 504, wherein the sample collection module 501 is used to select a number of nasopharyngeal carcinoma patients and a number of EBV antibody serological test positive people as target samples, and record the sample information of the target samples; the data acquisition module 502 is used to collect plasma samples of each target sample respectively, and process the plasma samples to obtain quantitative data of multiple metabolites and different components of lipoprotein in the plasma sample; the data modeling module 503 is used to perform regression analysis based on each component of the lipoprotein to obtain the correlation values ​​of different elements in the component with nasopharyngeal carcinoma, and use the elements whose correlation values ​​meet the preset conditions as modeling data; the risk classification module 504 is used to construct a risk scoring model according to the modeling data to calculate the nasopharyngeal carcinoma risk score of each target sample, and set at least one risk threshold to delineate different risk areas. The NRS risk threshold is specifically 14.95, and the model effect of distinguishing high-risk group and low-risk group with reference to the NRS50% quantile identified by the exploration cohort is the best.

[0060] In the data acquisition module 502, the plasma sample is processed to obtain various components of the lipoprotein group and lipid quantitative data in the plasma sample. The plasma sample is processed based on the partial least squares regression model prediction algorithm to obtain various components of the lipoprotein group and lipid quantitative data.

[0061] The correlation values ​​in the data modeling module 503 include the risk odds ratio and P, and the preset condition is that the risk odds ratio is greater than 1 and P is less than 0.05. The specific risk scoring model in the risk classification module 504 is expressed as Where x is the expression value of each risk factor, β is the risk odds ratio, and β is the correlation value. The NPC risk score is calculated based on this risk scoring model.

[0062] Embodiment 3

[0063] See also Figure 6 , Figure 6 Schematic diagram of the structure of an electronic device disclosed in an embodiment of the present invention. The electronic device may be a computer, a server, etc. Of course, in certain circumstances, it may also be a smart device such as a mobile phone, a tablet computer, a monitoring terminal, and an image acquisition device with processing functions. Figure 6 As shown, the electronic device may include:

[0064] A memory 601 storing executable program codes;

[0065] a processor 602 coupled to the memory 601;

[0066] The processor 602 calls the executable program code stored in the memory 601 to execute part or all of the steps in the method for constructing a nasopharyngeal carcinoma risk scoring model in the first embodiment.

[0067] An embodiment of the present invention discloses a computer-readable storage medium storing a computer program, wherein the computer program enables a computer to execute part or all of the steps in the method for constructing a nasopharyngeal carcinoma risk scoring model in the first embodiment.

[0068] The embodiment of the present invention further discloses a computer program product, wherein when the computer program product is run on a computer, the computer executes part or all of the steps in the method for constructing a nasopharyngeal carcinoma risk scoring model in embodiment one.

[0069] An embodiment of the present invention further discloses an application publishing platform, wherein the application publishing platform is used to publish a computer program product, wherein when the computer program product runs on a computer, the computer executes some or all of the steps in the method for constructing a nasopharyngeal carcinoma risk scoring model in embodiment one.

[0070] In various embodiments of the present invention, it should be understood that the size of the serial numbers of the processes does not necessarily mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0071] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, i.e., they may be located in one place or distributed over multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0072] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0073] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-accessible memory. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product, which is stored in a memory and includes several requests for a computer device (which can be a personal computer, a server or a network device, etc., specifically a processor in a computer device) to perform some or all of the steps of the method described in each embodiment of the present invention.

[0074] In the embodiments provided by the present invention, it should be understood that "B corresponding to A" means that B is associated with A, and B can be determined according to A. However, it should also be understood that determining B according to A does not mean determining B only according to A, and B can also be determined according to A and / or other information.

[0075] A person of ordinary skill in the art can understand that some or all of the steps in the various methods of the embodiments can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable storage medium, and the storage medium includes a read-only memory (ROM), a random access memory (RAM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), a one-time programmable read-only memory (OTPROM), an electronically erasable rewritable read-only memory (EEPROM), a compact disc (CD-ROM) or other optical disc storage, magnetic disk storage, magnetic tape storage, or any other computer-readable medium that can be used to carry or store data.

[0076] The above is a detailed introduction to the method, device, electronic device and storage medium for constructing a nasopharyngeal carcinoma risk scoring model disclosed in the embodiments of the present invention. Specific examples are used herein to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scopes. In summary, the content of this specification should not be understood as limiting the present invention.

Claims

1. A method for constructing a nasopharyngeal carcinoma risk scoring model, characterized in that: include: Selecting several nasopharyngeal carcinoma patients and several people with positive EBV antibody serological test results as target samples, and recording sample information of the target samples; Collecting plasma samples of each target sample separately, and processing the plasma samples to obtain quantitative data of various metabolites and different components of lipoproteins in the plasma samples; Based on each component of the lipoprotein, a regression analysis is performed to obtain correlation values ​​between different elements in the component and nasopharyngeal carcinoma, and the elements whose correlation values ​​meet preset conditions are used as modeling data; A risk scoring model is constructed based on the modeling data to calculate the nasopharyngeal carcinoma risk score of each target sample, and at least one risk threshold is set to delineate different risk areas.

2. The method for constructing a nasopharyngeal carcinoma risk scoring model according to claim 1, characterized in that: In the process of processing the plasma sample to obtain the various components of the lipoprotein group and the lipid quantitative data in the plasma sample, the plasma sample is processed based on the partial least squares regression model prediction algorithm to obtain the various components of the lipoprotein group and the lipid quantitative data.

3. The method for constructing a nasopharyngeal carcinoma risk scoring model according to claim 1, characterized in that: In the step of performing regression analysis on each component of the lipoprotein to obtain correlation values ​​between different elements in the component and nasopharyngeal carcinoma, the correlation values ​​are obtained through Logistic regression analysis.

4. The method for constructing a nasopharyngeal carcinoma risk scoring model according to claim 1, characterized in that: The correlation value includes a risk odds ratio and P, and the preset condition is that the risk odds ratio is greater than 1 and P is less than 0.05, and P is a statistical indicator.

5. The method for constructing a nasopharyngeal carcinoma risk scoring model according to claim 1, characterized in that: The expression of the risk scoring model is: where x is the expression value of each risk factor, β is the risk odds ratio, and β is the correlation value.

6. The method for constructing a nasopharyngeal carcinoma risk scoring model according to claim 1, characterized in that: The setting of at least one risk threshold to define different risk areas includes: Three risk thresholds are set at 25%, 50%, and 75%; The risk threshold is divided into four quantile ranges of 0%-25%, 25%-50%, 50%-75%, and 75%-100%. Four risk levels are set according to the four ranges, namely high-risk group, medium-high risk group, medium-low risk group, and low risk group.

7. A device for constructing a nasopharyngeal carcinoma risk scoring model, characterized in that: include: Sample collection module: used to select several nasopharyngeal carcinoma patients and several people with positive EBV antibody serological test as target samples, and record the sample information of the target samples; Data acquisition module: used to collect plasma samples of each target sample respectively, and process the plasma samples to obtain quantitative data of various metabolites and different components of lipoproteins in the plasma samples; Data modeling module: used for performing regression analysis based on each component of the lipoprotein to obtain correlation values ​​between different elements in the component and nasopharyngeal carcinoma, and taking the elements whose correlation values ​​meet preset conditions as modeling data; Risk classification module: used to construct a risk scoring model based on the modeling data to calculate the nasopharyngeal carcinoma risk score of each target sample, and set at least one risk threshold to delineate different risk areas.

8. An electronic device, characterized in that: include: A memory storing executable program code; a processor coupled to the memory; The processor calls the executable program code stored in the memory to execute the method for constructing a nasopharyngeal carcinoma risk scoring model according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, wherein the computer program enables a computer to execute the method for constructing a nasopharyngeal carcinoma risk scoring model according to any one of claims 1 to 6.

Images