Hearing loss illness risk prediction method, system, equipment and medium

Abstract

The invention belongs to the technical field of disease risk assessment, and discloses a hearing loss disease risk prediction method, system and device and a medium, and the method comprises the steps: obtaining hearing test data, serum creatinine detection data and genetic typing data of a tester; calculating estimated glomerular filtration rate data based on the serum creatinine detection data; screening SNP (Single Nucleotide Polymorphism) sites related to hearing loss in the genetic typing data, and calculating multi-gene risk scoring data on the basis of the SNP sites; based on a statistical method, combining the hearing test data, the estimated glomerular filtration rate data, the polygene risk scoring data and the basic feature data of the test personnel to construct a hearing loss disease risk prediction model; and predicting the hearing loss of the person to be predicted and the illness risk probability of the severity of the hearing loss based on the constructed hearing loss illness risk prediction model. According to the technical scheme, genetic and non-genetic factors are comprehensively considered, and the hearing loss disease risk can be evaluated more comprehensively and accurately.

Description

Technical Field

[0001] The present invention belongs to the technical field of disease risk assessment, and in particular, relates to a method, system, device and medium for predicting the risk of hearing loss. Background Art

[0002] Hearing loss (HL), as the most common sensory impairment in humans, is the third leading cause of disability in the global disease burden, with a prevalence of 20.3% in 2019, which means that one in five people suffers from hearing loss. Hearing loss is not only associated with a variety of disease outcomes, bringing adverse physiological and psychological effects, but also leads to a huge socioeconomic burden. It is estimated that more than 50% of adult hearing loss can be prevented through primary prevention measures or interventions implemented before health impacts occur.

[0003] Due to some similarities between the kidney and the inner ear in structure and function, and the fact that water and electrolyte disorders can affect the inner ear microcirculation, a decrease in the estimated glomerular filtration rate (eGFR), which reflects kidney function, may be associated with hearing loss. In addition, hearing loss is a complex polygenic disease, and an individual's genetic predisposition can explain 36% of the risk of hearing loss. The polygenic risk score (PRS) can integrate the effects of multiple genetic variants, quantify the risk of common variants for a specific disease or trait, and is calculated by adding the effects of various single nucleotide polymorphisms (SNPs). Genetic scores for hearing loss have been applied, but most of these genetic scores are based on European populations. Due to the differences in the frequency of variant sites and linkage disequilibrium patterns between different populations, PRSs constructed based on populations of European descent may not be well applied in the Chinese population. At the same time, due to differences in lifestyles and living environments among different populations and gene-environment interactions, the predictive effectiveness of these PRSs in other ethnic populations will also decrease. Summary of the invention

[0004] The purpose of the present invention is to provide a method, system, device and medium for predicting the risk of hearing loss to solve the problems existing in the above-mentioned prior art.

[0005] To achieve the above object, the present invention provides a method for predicting the risk of hearing loss, comprising:

[0006] Obtain the tester's hearing test data, serum creatinine test data and genotyping data;

[0007] Calculating estimated glomerular filtration rate data based on the serum creatinine test data;

[0008] Screening the genotyping data for SNP site data associated with hearing loss, and calculating polygenic risk score data based on the SNP site data;

[0009] Based on statistical methods, a hearing loss risk prediction model is constructed by combining the hearing test data, estimated glomerular filtration rate data, polygenic risk score data and basic characteristic data of the test person; wherein the basic characteristic data includes gender characteristic data and age data;

[0010] Based on the constructed hearing loss risk prediction model, the risk probability of hearing loss and its severity of the person to be predicted is predicted.

[0011] Optionally, the estimated glomerular filtration rate data is calculated based on the serum creatinine test data, specifically including:

[0012] Estimated glomerular filtration rate data were calculated based on the Chronic Kidney Disease Epidemiology Collaboration equation from the test subjects' serum creatinine test data.

[0013] Optionally, the SNP site data includes rs11521650, rs3926510, rs72474224, rs10869435, rs352500, rs78025198, rs4723908, rs748922, rs2923000, rs4114383 and rs6462956.

[0014] Optionally, the polygenic risk score data is calculated based on the SNP site data, and the specific calculation formula is:

[0015] PRS=n×(β1×SNP1+β2×SNP2+…+βn×SNPn) / (β1+β2+…+β11)

[0016] Where PRS is the polygenic risk score data, n is the number of SNPn in the model, βn is the effect value of the nth SNP, and SNPn is the number of effect alleles of the nth SNP carried by the individual.

[0017] Optionally, the process of constructing the hearing loss risk prediction model specifically includes:

[0018] Taking the hearing test data, estimated glomerular filtration rate data, polygenic risk score data and basic characteristic data of the test person as independent variables, taking hearing loss and its severity as dependent variables, and fitting a logistic regression model based on the independent variables and the dependent variables as a hearing loss risk prediction model;

[0019] Verify the discrimination of the hearing loss risk prediction model based on the ROC curve;

[0020] Based on the hearing loss risk prediction model verified by discrimination, the risk probability of hearing loss and its severity of the person to be predicted is predicted.

[0021] A hearing loss risk prediction system, comprising:

[0022] A data acquisition module is used to obtain the hearing test data, serum creatinine test data and genotyping data of the test person; calculate the estimated glomerular filtration rate data based on the serum creatinine test data; screen the SNP site data related to hearing loss in the genotyping data, and calculate the polygenic risk score data based on the SNP site data;

[0023] A model building module, for building a hearing loss risk prediction model based on a statistical method, combining the test person's hearing test data, estimated glomerular filtration rate data, polygenic risk score data and basic characteristic data; wherein the basic characteristic data includes gender characteristic data and age data;

[0024] The risk prediction module is used to predict the risk probability of hearing loss and its severity of the person to be predicted based on the constructed hearing loss risk prediction model.

[0025] An electronic device comprises a memory and a processor, wherein the memory is used to store a computer program, and the processor runs the computer program to enable the electronic device to execute a method for predicting the risk of hearing loss.

[0026] A computer-readable storage medium stores a computer program, which implements the method for predicting the risk of hearing loss when executed by a processor.

[0027] The technical effects of the present invention are:

[0028] Compared with traditional prediction models, the method of the present invention comprehensively considers genetic and non-genetic factors, and more comprehensively and accurately evaluates the risk of hearing loss, which has important public health significance for providing early intervention and preventive measures to achieve primary prevention of hearing loss; the present invention is based on the Chinese population, and the 11 hearing loss-related SNPs screened out through whole genome association studies are more suitable for genetic risk assessment in the Chinese population than other existing studies; the present invention uses a polygenic risk score to assess genetic risk, which can integrate the weak effects of single SNP site variations and more accurately predict the disease risk brought by genetics; the present invention combines the individual's estimated glomerular filtration rate level and the individual's age and gender to predict the risk of hearing loss, which has certain practical guiding significance for preventing and controlling hearing loss in people with abnormal estimated glomerular filtration rate levels. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0030] The drawings constituting a part of the present application are used to provide a further understanding of the present application. The illustrative embodiments and descriptions of the present application are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0031] Figure 1 It is a flowchart of a method for screening the risk of hearing loss based on the polygenic risk score of hearing loss in the Dongfeng Tongji cohort population, and jointly estimating the glomerular filtration rate and basic characteristics (age and gender) to construct a hearing loss risk prediction model in an embodiment of the present invention;

[0032] Figure 2 is an exposure-response relationship diagram of the polygenic risk score and the risk of hearing loss in an embodiment of the present invention;

[0033] Figure 3 is an exposure-response relationship diagram of estimated glomerular filtration rate and risk of hearing loss in an embodiment of the present invention;

[0034] Figure 4 It is a nomogram of a risk prediction model for hearing loss based on polygenic risk score, estimated glomerular filtration rate, gender and age in an embodiment of the present invention;

[0035] Figure 5 It is the ROC curve of the risk prediction model for hearing loss and its severity based on polygenic risk score, estimated glomerular filtration rate, gender and age in the embodiment of the present invention. DETAILED DESCRIPTION

[0036] Various exemplary embodiments of the present invention will now be described in detail. This detailed description should not be considered as limiting the present invention, but should be understood as a more detailed description of certain aspects, features, and embodiments of the present invention.

[0037] It should be understood that the terms described in the present invention are only for describing special embodiments and are not intended to limit the present invention. In addition, for the numerical range in the present invention, it should be understood that each intermediate value between the upper and lower limits of the scope is also specifically disclosed. Each smaller range between the intermediate value in any stated value or stated range and any other stated value or intermediate value in the described range is also included in the present invention. The upper and lower limits of these smaller ranges can be independently included or excluded in the scope.

[0038] It will be apparent to those skilled in the art that various modifications and variations may be made to the specific embodiments of the present invention description without departing from the scope or spirit of the present invention. Other embodiments derived from the present invention description will be apparent to those skilled in the art. The present application description and examples are exemplary only.

[0039] The words “include,” “including,” “have,” “contain,” etc. used in this article are open-ended terms, meaning including but not limited to.

[0040] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0041] Embodiment 1

[0042] like Figure 1 - Figure 5 As shown, this embodiment provides a method for predicting the risk of hearing loss, including: obtaining hearing test data, serum creatinine detection data and genotyping data of a test person; calculating estimated glomerular filtration rate data based on the serum creatinine detection data; screening SNP site data related to hearing loss in the genotyping data, and calculating polygenic risk score data based on the SNP site data; based on a statistical method, combining the hearing test data, estimated glomerular filtration rate data, polygenic risk score data and basic characteristic data of the test person to construct a hearing loss risk prediction model; wherein the basic characteristic data includes gender characteristic data and age data; based on the constructed hearing loss risk prediction model, predicting the risk probability of hearing loss and its severity of the person to be predicted.

[0043] This embodiment provides a method for predicting the risk of hearing loss based on a polygenic risk score and an estimated glomerular filtration rate, the method comprising obtaining an estimated glomerular filtration rate and basic characteristics of a population, calculating a polygenic risk score for hearing loss, and constructing a risk prediction model for hearing loss and its severity. The estimated glomerular filtration rate is calculated by serum creatinine, the basic characteristics include gender and age, and the polygenic risk score is constructed by 11 SNP sites screened by a genome-wide association study. This method integrates genetic and non-genetic factors, has good predictive efficacy, and can more accurately screen out people at high risk of hearing loss. It has certain practical guiding significance for preventing and controlling hearing loss in people with abnormal estimated glomerular filtration rate.

[0044] A method for screening the risk of hearing loss and its severity using a risk prediction model for hearing loss and its severity constructed based on polygenic risk scores and jointly estimating glomerular filtration rate, basic characteristics (age and gender) is proposed in this embodiment. The flowchart is shown in Figure 1 , where the method includes the following steps:

[0045] It is determined that the target population of this embodiment is from the participants with complete hearing test data in the Dongfeng Tongji Cohort Survey in 2013, and 13,203 participants without traumatic deafness, without ototoxic drug use, and with genotyping data are determined as the participants of this embodiment.

[0046] Among the above-mentioned target population, hearing is detected by professionals in a professional soundproof room. The determination of hearing loss is based on the 2017 WHO standard, that is, the average pure tone audiometric threshold (PTA) of the better ear at frequencies of 500, 1000, 2000, and 4000 Hz > 25 dB; the severity of hearing loss is divided into normal hearing (PTA ≤ 25 dB), mild hearing loss (25 < PTA ≤ 40 dB), moderate hearing loss (40 < PTA ≤ 60 dB), and severe hearing loss (PTA > 60 dB).

[0047] In the genome-wide association study, related participants in the above-mentioned target population are removed, and single nucleotide polymorphism sites SNP related to hearing loss are selected according to the screening criteria of P < 5×10 -6 , minor allele frequency (MAF) > 0.05, and Hardy-Weinberg equilibrium P > 1×10 -6 , and linkage disequilibrium sites of SNPs are removed according to the standard of r 2 < 0.2.

[0048] The 11 SNPs screened out are: rs11521650, rs3926510, rs72474224, rs10869435, rs352500, rs78025198, rs4723908, rs748922, rs2923000, rs4114383, and rs6462956.

[0049] Extract the effect allele, reference allele, and effect value of the above 11 SNPs from the genome-wide association study results, and obtain the number of effect alleles of the 11 SNPs from the genotype typing information of the Dongfeng Tongji Cohort population, and calculate the polygenic risk score for hearing loss:

[0050] PRS = n×(β1×SNP1 + β2×SNP2 + … + β n × SNP n ) / (β1 + β2 + … + β 11 )

[0051] Where n is the SNP in the model n The number of n refers to the effect value of the nth SNP, SNP n It refers to the number of effect alleles of the nth SNP carried by an individual. The higher the polygenic risk score, the higher the risk of hearing loss in the individual. The effect values ​​of each SNP are shown in Table 1:

[0052] Table 1 SNP site combination

[0053]

[0054]

[0055] Obtain the estimated glomerular filtration rate (mL / min / 1.73m 2 ) and basic characteristics (age and sex), the estimated glomerular filtration rate of the target population was calculated from the serum creatinine level tested by the Dongfeng Tongji Cohort Library laboratory (using the Chronic Kidney Disease Epidemiology Collaboration equation, CKD-EPI), and the calculation formula was as follows:

[0056] Estimated glomerular filtration rate = 141 × min (serum creatinine / k,1) α ×max(serum creatinine / k,1) -1.209 ×0.993 年龄 ×(1.018 if female)

[0057] If the patient is female, k is 0.7 and α is -0.329; if the patient is male, k is 0.9 and α is -0.411; min represents the smaller value of serum creatinine / k or 1, and max represents the larger value of serum creatinine / k or 1. The unit of serum creatinine is mg / dL.

[0058] The basic characteristics (age and gender) of the target population were collected through a semi-structured questionnaire during the field survey. The average age (standard deviation) of the target population was 64.8 (8.0) years old.

[0059] For continuous variables, mean ± standard deviation was used for description, and t-test or analysis of variance was used to compare the distribution differences among groups; for discrete variables, the number of people (percentage) was used for description, and χ 2The distribution differences between the groups were compared by test.Multivariate linear regression models and multivariate logistic regression models were fitted to estimate the relationship between the participants' polygenic risk score and estimated glomerular filtration rate and the mean pure tone hearing threshold, the hazard ratio between the risk of hearing loss, and further predict the probability of hearing loss or severity. Based on the multivariate restricted cubic spline function, the exposure-response relationship between the polygenic risk score or estimated glomerular filtration rate and the risk of hearing loss was evaluated. The ROC curve was drawn to evaluate the discrimination of the model. All statistical analyses were performed using SAS (version 9.4) and R software (version 4.4.0), with a two-sided test level of α=0.05.

[0060] Among the 13,203 target participants included in this method, the mean age (standard deviation) was 64.8 (8.0) years, of which 55.2% were female and 44.8% were male.

[0061] Table 2 Characteristics of participants with normal hearing and hearing loss

[0062] variable Total population Normal hearing Hearing loss P-value Sample size 13203 6812 6391 Polygenic risk score 9.53±2.30 9.30±2.27 9.77±2.31 <0.001 <![CDATA[Estimated glomerular filtration rate, mL / min / 1.73m 2 > 83.89±14.60 86.81±14.27 80.78±14.32 <0.001 Age (years), n (%) 64.8±8.0 61.9±7.5 68.0±7.2 <0.001 Gender, n (%) <0.001 female 7292(55.2) 4425(65.0) 2867(44.9) male 5911(44.8) 2387(35.0) 3524(55.1)

[0063] Table 3 Characteristics of participants by hearing loss severity

[0064]

[0065] In this example, a linear regression model is fitted to predict the average pure tone hearing threshold (PTA) of the participants based on the polygenic risk score, estimated glomerular filtration rate, age and gender (female is taken as the reference, and the model coefficient is 0). The specific model coefficients are shown in Table 4.

[0066] Table 4 Average pure tone hearing threshold (PTA) model coefficients

[0067]

[0068]

[0069] Average pure tone hearing threshold (PTA) model formula:

[0070] PTA = -16.08 + 0.43 × genetic risk score - 0.35 × estimated glomerular filtration rate / 10 + 0.64 × age + 1.61 × sex

[0071] The estimated glomerular filtration rate is calculated from the serum creatinine value of the target to be tested, and the polygenic risk score is the PRS calculated above. According to the current age of the target to be tested, the current average pure tone hearing threshold can be calculated.

[0072] In this example, a logistic regression model was fitted to predict the risk of hearing loss and its severity among participants based on the polygenic risk score, estimated glomerular filtration rate, age and gender (females were used as a reference, and the model coefficient was 0). The specific model coefficients are shown in Table 5.

[0073] Table 5 Hearing loss risk model coefficients

[0074]

[0075] The predicted probability p of hearing loss and its severity risk is:

[0076]

[0077] Among them, β i is the weight of the ith risk factor, X i is the value of the ith risk factor. Using the polygenic risk score, estimated glomerular filtration rate, age, and gender as predictor variables, the predicted probability of hearing loss or its severity for individuals in different states can be calculated.

[0078] The ROC curve is plotted to evaluate the ability of the prediction model to correctly distinguish between people with high and low risk of disease (discrimination). The ROC curve is a curve with 1-specificity on the horizontal axis and sensitivity on the vertical axis. The area under the ROC curve is AUC. The AUC value ranges from 0 to 1, and the larger the value, the better the classification performance. This embodiment uses a multivariate logistic regression model to construct a risk prediction model for hearing loss and its severity and plots the ROC curve to evaluate the discrimination of the model, such as Figure 5 As shown in the figure, the AUC of the prediction model for hearing loss and its severity constructed based on the polygenic risk score and estimated glomerular filtration rate was above 0.7, and increased from 0.701 for mild hearing loss to 0.831 for severe hearing loss with increasing severity, indicating that this prediction model has a strong ability to distinguish between people with hearing loss and those without the disease.

[0079] This embodiment is based on a large cohort study with a complete and unified quality control process. The hearing loss is tested by professionals, which provides more accurate data and a larger sample size basis for the construction of a hearing loss prediction model.

[0080] Compared with the traditional prediction model, the method of this embodiment comprehensively considers genetic and non-genetic factors, and more comprehensively and accurately assesses the risk of hearing loss, which has important public health significance for providing early intervention and preventive measures to achieve primary prevention of hearing loss.

[0081] This example is based on the Chinese population, and the 11 hearing loss-related SNPs screened out through a genome-wide association study are more suitable for genetic risk assessment in the Chinese population than other existing studies;

[0082] This embodiment uses a polygenic risk score to assess genetic risk, which can integrate the weak effects of single SNP site variations and more accurately predict the disease risk caused by genetics;

[0083] This embodiment predicts the risk of hearing loss based on the individual's estimated glomerular filtration rate level and the individual's age and gender, which has certain practical guiding significance for preventing and controlling hearing loss in people with abnormal estimated glomerular filtration rate levels.

[0084] A hearing loss risk prediction system, comprising:

[0085] A data acquisition module is used to obtain the hearing test data, serum creatinine test data and genotyping data of the test person; calculate the estimated glomerular filtration rate data based on the serum creatinine test data; screen the SNP site data related to hearing loss in the genotyping data, and calculate the polygenic risk score data based on the SNP site data;

[0086] A model building module, for building a hearing loss risk prediction model based on a statistical method, combining the test person's hearing test data, estimated glomerular filtration rate data, polygenic risk score data and basic characteristic data; wherein the basic characteristic data includes gender characteristic data and age data;

[0087] The risk prediction module is used to predict the risk probability of hearing loss and its severity of the person to be predicted based on the constructed hearing loss risk prediction model.

[0088] An electronic device comprises a memory and a processor, wherein the memory is used to store a computer program, and the processor runs the computer program to enable the electronic device to execute a method for predicting the risk of hearing loss.

[0089] A computer-readable storage medium stores a computer program, which implements the method for predicting the risk of hearing loss when executed by a processor.

[0090] The above is only a preferred specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A method for predicting the risk of hearing loss, characterized in that: include: Obtain the tester's hearing test data, serum creatinine test data and genotyping data; Calculating estimated glomerular filtration rate data based on the serum creatinine test data; Screening the genotyping data for SNP site data associated with hearing loss, and calculating polygenic risk score data based on the SNP site data; Based on statistical methods, a hearing loss risk prediction model is constructed by combining the hearing test data, estimated glomerular filtration rate data, polygenic risk score data and basic characteristic data of the test person; wherein the basic characteristic data includes gender characteristic data and age data; Based on the constructed hearing loss risk prediction model, the risk probability of hearing loss and its severity of the person to be predicted is predicted.

2. A method for predicting the risk of hearing loss according to claim 1, characterized in that: The estimated glomerular filtration rate data is calculated based on the serum creatinine test data, specifically including: Estimated glomerular filtration rate data were calculated based on the Chronic Kidney Disease Epidemiology Collaboration equation from the test subjects' serum creatinine test data.

3. A method for predicting the risk of hearing loss according to claim 1, characterized in that: The SNP site data include rs11521650, rs3926510, rs72474224, rs10869435, rs352500, rs78025198, rs4723908, rs748922, rs2923000, rs4114383 and rs6462956.

4. The method for predicting the risk of hearing loss according to claim 1, characterized in that: The polygenic risk score data is calculated based on the SNP site data, and the specific calculation formula is: PRS=n×(β1×SNP1+β2×SNP2+…+β n ×SNP n ) / (β1+β2+…+β 11 ) In the formula, PRS is the polygenic risk score data, n is the SNP in the model n The number of n is the effect value of the nth SNP, SNP n is the number of effect alleles of the nth SNP carried by the individual.

5. The method for predicting the risk of hearing loss according to claim 1, characterized in that: The process of constructing the hearing loss risk prediction model specifically includes: Taking the hearing test data, estimated glomerular filtration rate data, polygenic risk score data and basic characteristic data of the test person as independent variables, taking hearing loss and its severity as dependent variables, and fitting a logistic regression model based on the independent variables and the dependent variables as a hearing loss risk prediction model; Verify the discrimination of the hearing loss risk prediction model based on the ROC curve; Based on the hearing loss risk prediction model verified by discrimination, the risk probability of hearing loss and its severity of the person to be predicted is predicted.

6. A hearing loss risk prediction system, characterized in that: include: A data collection module is used to obtain the tester's hearing test data, serum creatinine test data and genotyping data; Calculating estimated glomerular filtration rate data based on the serum creatinine test data; Screening the genotyping data for SNP site data associated with hearing loss, and calculating polygenic risk score data based on the SNP site data; A model building module, for building a hearing loss risk prediction model based on a statistical method, combining the test person's hearing test data, estimated glomerular filtration rate data, polygenic risk score data and basic characteristic data; wherein the basic characteristic data includes gender characteristic data and age data; The risk prediction module is used to predict the risk probability of hearing loss and its severity of the person to be predicted based on the constructed hearing loss risk prediction model.

7. An electronic device, characterized in that: It comprises a memory and a processor, the memory is used to store a computer program, and the processor runs the computer program to enable the electronic device to execute a method for predicting the risk of hearing loss according to any one of claims 1-5.

8. A computer-readable storage medium, characterized in that: It stores a computer program, which, when executed by a processor, implements a method for predicting the risk of hearing loss as described in any one of claims 1 to 5.

Images