System for analyzing influence of oral cavity weakness on sarcopenia

Abstract

The invention relates to the technical field of data analysis, and discloses a system for analyzing the influence of oral cavity weakness on sarcopenia, which comprises a data acquisition module, an influence analysis module, a data visualization module and an interactive feedback module, the influence analysis module comprises a correlation analysis module, a causal relationship analysis module, a longitudinal change analysis module and an action mechanism analysis module; the causal relationship analysis module is used for analyzing the causal relationship between the genetic tool variable and the onset risk of oral cavity weakness and sarcopenia; and the longitudinal change analysis module is used for analyzing the change trends of oral cavity weakness and sarcopenia of the old people at different time points. According to the method, the influence of the oral cavity weakness on the sarcopenia can be deeply researched, the depth and the breadth of research on the relationship between the oral cavity weakness and the sarcopenia are improved, and a scientific, accurate and personalized solution is provided for health management of old people.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data analysis, in particular to an analysis system for the influence of oral frailty on sarcopenia. BACKGROUND

[0002] With the intensification of global population aging, the health problems of the elderly population have attracted increasing attention. Oral health and muscle health, as important components of the overall health of the elderly, are often interrelated and have a profound impact on the quality of life, independence and life expectancy of the elderly. Oral frailty mainly refers to the degradation of oral function in the elderly, which is usually manifested as tooth loss, difficulty in chewing, swallowing disorders, etc. These problems not only affect the eating ability of the elderly, but also may lead to inadequate nutrient intake, thereby affecting overall health. Sarcopenia refers to the gradual decline in muscle mass and strength with age, which is often closely related to the motor ability, metabolic health and quality of life of the elderly. Since the occurrence and development of sarcopenia involve multiple physiological and pathological factors, its research is not limited to muscle itself, but also needs to be considered comprehensively in combination with nutritional status, inflammation level, nervous system function and other health-related factors.

[0003] In recent years, more and more research has focused on the relationship between oral frailty and sarcopenia. Some scholars have suggested that oral frailty may affect food intake and change nutritional status, thereby affecting muscle mass and function. At the same time, the decline in oral function may also affect the social interaction and mental health of the elderly, such as increasing the risk of depression and reducing cognitive function, which may indirectly affect the development of sarcopenia. In addition, some studies have shown that chronic inflammation may play a key role in this process. Elderly people may develop chronic inflammation such as periodontitis due to poor oral health, and chronic inflammation is closely related to muscle atrophy, which may accelerate the progression of sarcopenia. However, although existing research has revealed that there may be some interaction and influence between oral frailty and sarcopenia, the specific mechanisms have not been fully revealed, and many key issues still lack systematic exploration and verification.

[0004] Traditional research on the relationship between oral frailty and sarcopenia mainly relies on single statistical analysis methods such as regression analysis and correlation analysis. Although these methods can reveal the relationship between variables to some extent, they often ignore the complex causal relationship between the two and the potential mediating effect. For example, traditional research usually uses cross-sectional data for analysis, which lacks observation of long-term trends, limiting the causal inference ability of the research conclusions. In addition, since the health problems of the elderly often involve multiple interrelated biological, psychological and social factors, single statistical analysis methods are difficult to fully reveal the complex interaction mechanisms, thereby limiting the depth and breadth of the research results.

[0005] Currently, no effective solutions have been proposed for the problems in related technologies. Summary of the Invention

[0006] In response to the problems in the related art, the present invention proposes an analysis system for the impact of oral weakness on sarcopenia to overcome the above-mentioned technical problems existing in the existing related art.

[0007] To this end, the specific technical solutions adopted in the present invention are as follows: An analysis system for the impact of oral frailty on sarcopenia, comprising a data acquisition module, an impact analysis module, a data visualization module, and an interactive feedback module connected in sequence; Data collection module, used to collect data related to oral frailty and sarcopenia; Data visualization module, used to generate data visualization charts; Interactive feedback module, used to interact with users and provide personalized feedback; Among them, the impact analysis module includes a correlation analysis module, a causal relationship analysis module, a longitudinal change analysis module and an action mechanism analysis module connected in sequence; Correlation analysis module, used to analyze the relationship between oral frailty indicators and sarcopenia indicators through statistical methods; Causality analysis module, used to analyze the causal relationship between genetic instrumental variables and the risk of oral frailty and sarcopenia using two-sample Mendelian randomization combined with genome-wide association study data; The longitudinal change analysis module is used to analyze the changing trends of oral frailty and sarcopenia in the elderly population at different time points using a deep learning model combined with time-varying confounding factor adversarial learning technology; The mechanism analysis module is used to analyze the specific mechanism by which oral frailty affects sarcopenia using structural equation models.

[0008] Furthermore, the data acquisition module includes an oral frailty index acquisition module, a sarcopenia index acquisition module, a health index acquisition module and a data preprocessing module which are connected in sequence; Among them, the oral frailty index collection module is used to collect oral frailty indicators of the elderly population, and oral frailty indicators include the number of teeth, chewing function and swallowing function; The sarcopenia index collection module is used to collect sarcopenia indicators of the elderly population, including grip strength, gait speed and muscle mass; The health indicator collection module is used to collect other relevant health data of the elderly population, including inflammation indicators, cognitive function, depression status, nutritional status and activity ability; The data preprocessing module is configured to fill in missing values, detect outliers, and standardize the collected data to ensure the quality and consistency of the data.

[0009] Further, the correlation analysis module includes the following when analyzing the relationship between the oral frailty indicators and the sarcopenia indicators by statistical methods: calculating the basic statistics of each indicator variable and drawing a frequency distribution graph, a histogram, and a box plot of the data according to the calculation results to check the distribution of the data; evaluating the individual distribution of the oral frailty indicators and the sarcopenia indicators, and selecting a statistical method according to the evaluation results, wherein a parametric statistical method is selected when the evaluation results conform to a normal distribution, and a non-parametric statistical method is selected when the evaluation results do not conform to a normal distribution; analyzing the linear or nonlinear relationship between the oral frailty and the sarcopenia indicators by using the selected statistical method, and analyzing the linear effect of the oral frailty on the sarcopenia by using a regression analysis method.

[0010] Further, the causal relationship analysis module includes a genetic instrumental variable selection module, a Mendelian randomization analysis module, an analysis result verification module, and a causal relationship output module connected in sequence; The genetic instrumental variable selection module is configured to obtain the association data of oral frailty and sarcopenia from two independent whole-genome association study data sets, and extract single nucleotide polymorphism sites related to oral frailty and sarcopenia as genetic instrumental variables, respectively. The Mendelian randomization analysis module is configured to evaluate the effect of the genetic instrumental variables on oral frailty and sarcopenia, and infer the causal relationship between oral frailty and sarcopenia by using multiple Mendelian randomization methods; The analysis result verification module is configured to evaluate whether the hypothesis of the Mendelian randomization analysis is established to ensure the reliability of the causal inference; The causal relationship output module is configured to explain the analysis results and output the causal relationship between oral frailty and sarcopenia.

[0011] Further, the genetic instrumental variable selection module is further configured to evaluate the strength of the genetic instrumental variables to avoid bias and inefficiency caused by weak instrumental variables, enhance the statistical power and robustness of the causal inference, and improve the reliability of the research results; The formula for evaluating the strength of the genetic instrumental variables is as follows: In the formula, F represents the strength of the genetic instrumental variables, R 2 represents the explanatory variance of the genetic instrumental variables on the exposure variables, n represents the number of samples, kThe number of genetic instrumental variables.

[0012] Further, the Mendelian randomization analysis module includes the following when assessing the effect of genetic instrumental variables on oral frailty and sarcopenia and inferring the causal relationship between oral frailty and sarcopenia using various Mendelian randomization methods: obtaining the effect estimate, standard error and statistical significance of the genetic instrumental variables related to oral frailty from the whole genome association study data, and aligning the effect allele direction to ensure consistency of the effect direction of the genetic instrumental variables in different data sets; obtaining the effect estimate, standard error and statistical significance of the genetic instrumental variables related to sarcopenia from the whole genome association study data of sarcopenia, and checking the effect allele direction consistency to ensure the effect allele direction consistency in the oral frailty and sarcopenia data; calculating the weighted average effect value of the genetic instrumental variables using the inverse variance weighting method, and analyzing the preliminary causal effect of oral frailty on sarcopenia according to the weighted average effect value; detecting horizontal pleiotropic bias by regression adjustment method to check whether the genetic instrumental variables affect other pathways, and using weighted median method to improve robustness when some genetic instrumental variables are invalid; identifying and removing abnormal genetic instrumental variables that have a greater impact on the results based on the outlier adjustment method, and performing sensitivity analysis on the removed genetic instrumental variables, summarizing the causal effect, and obtaining the causal relationship between oral frailty and sarcopenia.

[0013] Further, the analysis result test module includes a correlation hypothesis evaluation module, an independence hypothesis evaluation module and an exclusivity hypothesis evaluation module connected in sequence. The correlation hypothesis evaluation module is configured to evaluate whether the genetic instrumental variables are valid according to the effect estimate, strength and significance test of the genetic instrumental variables. The independence hypothesis evaluation module is configured to test whether the genetic instrumental variables are related to confounding factors and whether the independence hypothesis is met. The exclusivity hypothesis evaluation module is configured to evaluate whether the genetic instrumental variables only affect the outcome variable through the exposure variable.

[0014] Further, the longitudinal change analysis module includes the following when analyzing the change trend of oral frailty and sarcopenia of the elderly population at different time points using a deep learning model combined with time-varying confounding factor adversarial learning technology: obtaining longitudinal data of the elderly population in the database at multiple time points about oral frailty and sarcopenia, and preprocessing to select key features affecting the change of oral frailty and sarcopenia, wherein the longitudinal data includes exposure variables, outcome variables and time-varying confounding factors; The time-varying confounders are input into the generator to generate exposure variables without the influence of the confounders. The discriminator receives the generated exposure variables and the original confounder data, evaluates the impact of the time-varying confounders on the exposure variables, and alternately trains the generator and the discriminator in an adversarial manner. The exposure variables, outcome variables, and confounding factors at each time point are used as input data, and the predicted values ​​of the exposure variables and outcome variables at future time points are used as output to construct a long-short-term memory network time series model. An adversarial regularization term is added to the loss function of the long-short-term memory network, and the adversarial regularization is used to train the long-short-term memory network time series model. The trained long short-term memory network temporal model was used to output the predicted values ​​of exposure variables and outcome variables at future time points, and the changing trends of oral frailty and sarcopenia in the elderly population at different time points were analyzed based on the predicted results.

[0015] Furthermore, the loss function of the long short-term memory network is expressed as: Where, L represents the loss function, E Indicates the expectation symbol, y t Indicates the t The actual target value at a time point, Indicates the t The predicted value of the model at a time point, 、 They are used to adjust the hyperparameters of the adversarial loss and regularization term on the total loss, D represents the discriminator model, represents the output prediction of the long short-term memory network timing model, f represents the long short-term memory network temporal model, Indicates the t Exposure variables at time points, Indicates the t Time-varying confounders at time points, represents the L2 regularization term.

[0016] Furthermore, the mechanism analysis module includes the following when using the structural equation model to analyze the specific mechanism by which oral frailty affects sarcopenia: Identify latent variables and determine corresponding observed variables based on the latent variables, including oral frailty, sarcopenia, inflammatory factors, cognitive function, depression, and nutritional status; define the causal relationship and path between the latent variables and the observed variables; The matching degree between the latent variables and the observation variables is verified by using a factor analysis method to ensure the reliability and effectiveness of the measurement results, the causal paths between the latent variables are set, and path analysis is performed to estimate the path coefficients to obtain a structural equation model; The fitting degree of the structural equation model is detected, the path coefficients are interpreted, and the influence mechanism of oral frailty on sarcopenia is analyzed according to the path coefficients.

[0017] The beneficial effects of the present application are: 1) The present application provides a systematic analysis framework by comprehensively using data collection, influence analysis, longitudinal change analysis, causal inference and structural equation modeling techniques, so that the influence of oral frailty on sarcopenia can be deeply studied, not only improving the depth and breadth of the relationship between oral frailty and sarcopenia, but also providing a scientific, accurate and personalized solution for the health management of the elderly population.

[0018] 2) The present application uses statistical methods and Mendelian randomization methods to deeply analyze the relationship between oral frailty and sarcopenia, and reveals the potential causal relationship between the two. Especially in the selection of genetic instrumental variables and the support of multi-sample data, the accuracy and scientificity of causal inference are improved, and the interference of confounding factors is avoided.

[0019] 3) The present application uses the long short-term memory network (LSTM) model in deep learning, combined with time-varying confounding factor countermeasure learning technology, to realize the change trend prediction of oral frailty and sarcopenia at different time points. This technology effectively reduces the influence of time-varying confounding factors, and provides a more accurate basis for future prediction and intervention.

[0020] 4) The present application deeply analyzes how oral frailty affects sarcopenia through multiple paths by using structural equation modeling (SEM). By using factor analysis and path analysis, the direct effect of oral frailty on sarcopenia and its potential mediating mechanism are revealed, which can better understand the influence of oral frailty on sarcopenia. BRIEF DESCRIPTION OF DRAWINGS

[0021] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed in the embodiments. Obviously, the drawings described below are only some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor on the basis of these drawings.

[0022] Figure 1 is a structural diagram of an oral frailty on sarcopenia analysis system according to an embodiment of the present application.

[0023] In the figure: 1. Data collection module; 2. Impact analysis module; 3. Data visualization module; 4. Interactive feedback module. DETAILED DESCRIPTION

[0024] To further illustrate each embodiment, the present invention provides drawings, which are part of the disclosure of the present invention. They are mainly used to illustrate the embodiments and can be used in conjunction with the relevant descriptions in the specification to explain the operating principles of the embodiments. By referring to these contents, ordinary technicians in this field should be able to understand other possible implementation methods and advantages of the present invention. The components in the figures are not drawn to scale, and similar component symbols are generally used to represent similar components.

[0025] According to an embodiment of the present invention, a system for analyzing the effect of oral weakness on sarcopenia is provided.

[0026] The present invention will now be further described with reference to the accompanying drawings and specific embodiments. Figure 1 As shown, the analysis system for the impact of oral frailty on sarcopenia according to an embodiment of the present invention includes a data acquisition module 1, an impact analysis module 2, a data visualization module 3 and an interactive feedback module 4 connected in sequence; Data collection module 1, used to collect data related to oral frailty and sarcopenia; Specifically, the data acquisition module 1 includes an oral frailty index acquisition module, a sarcopenia index acquisition module, a health index acquisition module and a data preprocessing module which are connected in sequence; Among them, the oral frailty index collection module is used to collect oral frailty indicators of the elderly population (for example, obtained through questionnaires, oral examinations, etc.), and oral frailty indicators include the number of teeth, chewing function, and swallowing function; The sarcopenia index collection module is used to collect sarcopenia indicators (such as grip strength, gait test, muscle mass measurement, etc.) from the elderly population. The sarcopenia indicators include grip strength, gait speed and muscle mass; The health indicator collection module is used to collect other relevant health data of the elderly population, including inflammation indicators, cognitive function, depression status, nutritional status and activity ability; The data preprocessing module is used to fill missing values, detect outliers and perform data standardization on the collected data to ensure the quality and consistency of the data.

[0027] Handling missing data: You can use methods such as mean filling and interpolation to fill missing values, or choose to delete or fill according to the degree of missingness.

[0028] Outlier detection: Use box plots, standard deviation methods, etc. to detect and handle outliers.

[0029] Data Standardization: For variables with different dimensions, standardization or normalization can be performed to ensure comparability between data.

[0030] The impact analysis module 2 includes a correlation analysis module, a causal relationship analysis module, a longitudinal change analysis module, and a mechanism analysis module connected in sequence. The correlation analysis module is used to analyze the relationship between the oral frailty indicators and the sarcopenia indicators by statistical methods. The correlation analysis module includes the following steps when analyzing the relationship between the oral frailty indicators and the sarcopenia indicators by statistical methods: 1) Calculate the basic statistics of each indicator variable, and draw the frequency distribution chart, histogram, and boxplot of the data according to the calculation results to check the distribution of the data. 11) Calculate descriptive statistics Calculate the mean (Mean), median (Median), standard deviation (Standard Deviation, SD), skewness (Skewness), and kurtosis (Kurtosis).

[0031] Skewness (Skewness): Measures whether the data is symmetric. If the skewness is far from 0 (such as >1 or <-1), it indicates that the data is highly skewed.

[0032] Kurtosis (Kurtosis): Measures the steepness of the data distribution. Higher kurtosis means that the data may have a heavy-tailed distribution (long tail effect).

[0033] 12) Draw the data distribution chart Histogram (Histogram): Can visually observe the distribution of the data, whether it is normally distributed, or whether there is a significant skew.

[0034] Q-Q plot (Quantile-Quantile Plot): If the data points on the Q-Q plot are basically straight lines, it means that the data is close to normal distribution; if the data points deviate from the straight line, it means that the data may have skewness or heavy tail.

[0035] Boxplot (Boxplot): Can help identify outliers, and if there are too many outliers, it may affect the analysis.

[0036] 2) Evaluate the individual distribution of oral frailty indicators and sarcopenia indicators to determine whether further conversion (such as log conversion) is needed to meet the assumptions of subsequent analysis, and select statistical methods according to the evaluation results; Methods to check normality of data: Shapiro-Wilk test (for small samples of less than 50, if p<0.05, it indicates that the data deviates from normal distribution), Kolmogorov-Smirnov test (for large samples, but sensitive to outliers) are used to check if the data is normally distributed. Parametric statistical methods (such as Pearson correlation) are chosen when the results of the evaluation are normally distributed, and non-parametric statistical methods (such as Spearman rank correlation) are chosen when the results of the evaluation are not normally distributed. Log Transformation: Applicable situation: Data distribution is right-skewed (positively skewed), i.e. most data is small, and a few data is large.

[0037] Method: Take the logarithm of the data.

[0038] Applicable scenarios: Number of missing teeth (if some people have no teeth at all); muscle mass (if the distribution span is large); 3) Use the selected statistical method to analyze the linear or nonlinear relationship between oral frailty and sarcopenia indicators, and use regression analysis to analyze the linear effect of oral frailty on sarcopenia.

[0039] 31) Correlation analysis: used to analyze the linear or nonlinear relationship between oral frailty and sarcopenia indicators; Pearson correlation coefficient: used to assess the linear correlation between oral frailty and sarcopenia indicators, provided that the data is normally distributed.

[0040] Spearman rank correlation coefficient: used to assess the monotonic relationship between oral frailty and sarcopenia, suitable for non-normal distribution data.

[0041] Output: Correlation coefficient and its significance level (p-value), if the p-value is less than the significance level (such as 0.05), it is considered that there is a significant statistical relationship between the two.

[0042] 32) Regression analysis: used to further analyze the relationship between oral frailty and sarcopenia indicators, and to assess the predictive effect of oral frailty on sarcopenia.

[0043] Linear regression analysis: used to assess the linear effect of oral frailty (as independent variable) on sarcopenia indicators (as dependent variable). If the data is continuous variable, simple linear regression or multiple linear regression can be performed. The predictive effect of oral frailty on sarcopenia is assessed by the regression coefficient.

[0044] Multiple regression analysis: If there are multiple potential confounding factors (such as age, gender, chronic diseases, etc.), multiple regression analysis can be performed to control these confounding factors in the model and further explore the independent relationship between oral frailty and sarcopenia.

[0045] Output: Regression coefficients (β values), R² values, significance levels (p-values), etc. If the p-value is less than the significance level (0.05), it is considered that oral frailty has a significant impact on sarcopenia.

[0046] 33) Model testing and diagnosis: used to evaluate the effectiveness of the regression model and hypothesis testing.

[0047] Residual analysis: check whether the residuals of the regression model meet the normal distribution, and whether the assumptions of independence and homoscedasticity are met.

[0048] VIF test: check whether there is multicollinearity between the independent variables. If the VIF value is too high (generally greater than 10), some variables may need to be removed.

[0049] Causal relationship analysis module, for analyzing the causal relationship between genetic instrumental variables and the risk of oral frailty and sarcopenia by two-sample Mendelian randomization method combined with whole genome association study data; Specifically, the causal relationship analysis module includes a genetic instrumental variable selection module, a Mendelian randomization analysis module, an analysis result testing module, and a causal relationship output module connected in turn; Among them, the genetic instrumental variable selection module is used to obtain the association data of oral frailty and sarcopenia from two independent whole genome association study data sets, and to extract single nucleotide polymorphism sites related to oral frailty and sarcopenia as genetic instrumental variables, respectively; Specifically, the single nucleotide polymorphisms related to oral frailty are as follows: Select single nucleotide polymorphisms related to oral frailty: select single nucleotide polymorphisms related to oral frailty from GWAS data sets. These single nucleotide polymorphisms usually have significant statistical associations (p-value less than a pre-set significance threshold, such as ). These single nucleotide polymorphisms should have a strong effect on the genes or pathways related to oral frailty.

[0050] Effect size and direction: collect the effect size (such as β value) and effect direction (increasing or decreasing effect of alleles) of these single nucleotide polymorphisms.

[0051] The single nucleotide polymorphisms related to sarcopenia are as follows: Selection of single nucleotide polymorphisms associated with sarcopenia: Significant single nucleotide polymorphisms are screened from the GWAS dataset of sarcopenia. These single nucleotide polymorphisms should play an important role in the pathogenesis of sarcopenia, and have a higher statistical significance.

[0052] Effect size and direction: Similarly, the effect size and direction of the single nucleotide polymorphisms associated with sarcopenia are collected, and they are ensured to meet the significance criteria when selected.

[0053] The genetic instrumental variables are determined as follows: Independence test: To avoid correlation between instrumental variables (such as chain effect), linkage disequilibrium (LD) test is performed on the single nucleotide polymorphisms of oral frailty and sarcopenia, and independent single nucleotide polymorphisms are selected. The degree of LD can be evaluated using the r² value, and a threshold of r² < 0.1 is usually set.

[0054] Effect consistency: Ensure that the selected single nucleotide polymorphisms have consistent effect directions in different datasets. If the effect directions of certain single nucleotide polymorphisms are opposite in different datasets, these single nucleotide polymorphisms should be selected with caution.

[0055] The genetic instrumental variable selection module is also used to evaluate the strength of the genetic instrumental variables to avoid bias and low efficiency caused by weak instrumental variables, enhance the statistical power and robustness of causal inference, and improve the credibility of research results; wherein the evaluation formula of the strength of the genetic instrumental variables is: wherein, F represents the strength of the genetic instrumental variables, R 2 represents the explanatory variance of the genetic instrumental variables on the exposure variable, n represents the number of samples, k represents the number of genetic instrumental variables.

[0056] The Mendelian randomization analysis module is used to evaluate the effects of genetic instrumental variables on oral frailty and sarcopenia, and infer the causal relationship between oral frailty and sarcopenia using multiple Mendelian randomization methods; Specifically, the Mendelian randomization analysis module includes the following when evaluating the effects of genetic instrumental variables on oral frailty and sarcopenia, and inferring the causal relationship between oral frailty and sarcopenia using multiple Mendelian randomization methods: 1) Obtain the effect estimate, standard error, and statistical significance (p-value) of the genetic instrumental variables associated with oral frailty from the whole genome association study data, and perform effect allele direction alignment to ensure that the effect directions of the genetic instrumental variables are consistent in different datasets; Specifically, the effect allele direction alignment includes: 11) Understanding Effect Alleles: Effect Allele: Refers to the variant allele associated with the trait (in this case, oral frailty). Each SNP typically has two alleles (reference and variant), and the effect size is calculated based on the variant allele.

[0057] 12) Checking Effect Alleles Across Different Datasets: Issue of Inconsistent Effect Allele Direction: In different GWAS datasets (e.g., from different populations or studies), the effect allele for the same SNP can have different directions (i.e., positive or negative effect). Therefore, it is necessary to ensure that the direction of the effect allele for the selected SNPs is consistent across all datasets in the analysis.

[0058] 13) Aligning Allele Directions: Finding Allele Directions: For each SNP, check the effect allele in different datasets. Each GWAS dataset provides the reference and effect alleles for the SNP.

[0059] Aligning Effect Alleles: If the direction of the effect allele is inconsistent, it needs to be aligned using the following methods: Checking Effect Sizes: For different datasets, if the direction of the effect allele is inconsistent, the effect direction can usually be reversed to align. For example, if a SNP shows a positive effect in one dataset and a negative effect in another, the effect size of the latter can be reversed to make the effect directions consistent.

[0060] Aligning Effect Sizes: Once the directions are aligned, ensure the effects are consistent by calculating the SNP effect estimates (β values). Assuming the directions of the effect alleles are consistent, the effect values can be directly compared.

[0061] 14) Aligning Genotypes: When aligning the directions of effect alleles, it is necessary to ensure that the alignment of the genotype data used is consistent. Typically, genotype alignment can be done through the following steps: Checking Reference Genomes for Alleles: Ensure that the reference genome versions (e.g., GRCh37 or GRCh38) for all SNPs are consistent.

[0062] Aligning the Same Alleles: For different study datasets, ensure that the meanings of the selected reference and variant alleles are consistent across all datasets.

[0063] 2) Obtain the effect estimates, standard errors, and statistical significance of genetic instrumental variables associated with sarcopenia from the genome-wide association study data of sarcopenia, and check the consistency of effect allele directions to ensure the consistency of effect allele directions in oral frailty and sarcopenia data; 3) Calculate the weighted average effect of genetic instrumental variables using inverse-variance weighting method, and analyze the preliminary causal effect of oral frailty on sarcopenia according to the weighted average effect; detect horizontal pleiotropic bias by regression adjustment method to check whether the genetic instrumental variables affect other pathways, and use weighted median method to improve robustness when some genetic instrumental variables are invalid; 31) Inverse-variance weighting method: Calculate the weighted average effect of instrumental variables: use inverse-variance weighting method to weight the effect of each instrumental variable according to its standard error, and calculate the overall effect of oral frailty on sarcopenia.

[0064] Significance test: by calculating the p-value, if the p-value is less than 0.05, it is considered that oral frailty may have a significant causal effect on sarcopenia.

[0065] Result analysis: this method usually provides the most accurate effect estimate, but its premise is that all instrumental variables meet the Mendelian randomization assumption. If the p-value of this method is less than 0.05, it means that oral frailty may have a causal effect on sarcopenia, but further verification is needed.

[0066] 32) Regression adjustment method (used to detect horizontal pleiotropic bias): Regression intercept test: based on the inverse-variance weighting method, further use the regression adjustment method to test whether the instrumental variables have horizontal pleiotropy; use a regression model to adjust the instrumental variables, and test whether the intercept of the regression model is significantly not 0. If the intercept is not 0 and the p-value is less than 0.05, it means that there is horizontal pleiotropic bias, which may affect the accuracy of causal inference.

[0067] Result analysis: if there is no horizontal pleiotropy (intercept ≈ 0, p> 0.05), the result of inverse-variance weighting method is more reliable, and sensitivity analysis can be continued. If there is horizontal pleiotropy (intercept is significantly ≠ 0, p< 0.05), more robust methods need to be used to further estimate the causal effect to reduce the influence of bias.

[0068] Further adjustment: if pleiotropic bias is found, the analysis can be re-conducted by removing or adjusting specific instrumental variables.

[0069] 33) Weighted median method (applied to the case where some instrumental variables are invalid): Median regression: If some instrumental variable effects are invalid, use weighted median regression to estimate the causal effect using the median of the instrumental variable effects, enhancing the model's robustness to invalid instrumental variables.

[0070] Robustness verification: Verify the robustness of causal inference by comparing the results of weighted median regression with other methods.

[0071] Result analysis: If the results of weighted median regression are similar to those of inverse variance weighting, it indicates that the causal inference is robust and the horizontal multiple effect is small. If the results of weighted median regression are significantly different from those of inverse variance weighting, it indicates that some instrumental variables are biased, and further removal of abnormal instrumental variables and reanalysis may be needed.

[0072] 4) Identification and removal of abnormal genetic instrumental variables with a greater impact on results based on the abnormal value adjustment method, and sensitivity analysis of the removed genetic instrumental variables, summarizing the causal effect to obtain the causal relationship between oral weakness and sarcopenia.

[0073] 41) Abnormal value detection and adjustment: Abnormal value identification: Identify and remove abnormal genetic instrumental variables that have a greater impact on results to reduce the interference of abnormal values on causal inference.

[0074] Reanalysis: After removing abnormal values, re-analyze causal inference to ensure the robustness of the results.

[0075] Result analysis: If the results remain unchanged after removing abnormal values, it indicates that the causal inference is robust and the causal effect of oral weakness on sarcopenia is reliable. If the results change significantly after removing abnormal values, it indicates that the causal inference is affected by individual genetic instrumental variables, and the results need to be interpreted with caution.

[0076] 42) Sensitivity analysis: Sensitivity analysis of removing each instrumental variable: By removing each genetic instrumental variable one by one, it is tested whether a single genetic instrumental variable has a significant impact on the estimation of causal effect, ensuring the stability of causal inference.

[0077] Multiple method comparison: Use different Mendelian randomization methods (such as inverse variance weighting, regression adjustment, median method, etc.) for causal inference, compare the consistency of results of different methods, and further verify the robustness of causal relationship.

[0078] 43) Result interpretation and causal relationship inference: Summarize causal effects: Summarize the results of different methods (inverse variance weighting, regression adjustment, weighted median method, etc.) to obtain the causal effect estimate of oral weakness on sarcopenia, and report the 95% confidence interval.

[0079] Robustness of causal effect: If all method results are consistent and there is no multiplicity bias, and the p-value is less than 0.05, it can be considered that there is a significant causal effect of oral frailty on sarcopenia.

[0080] An analysis result test module is configured to evaluate whether the Mendelian randomization analysis satisfies the hypothesis to ensure the reliability of the causal inference. Specifically, the analysis result test module comprises a correlation hypothesis evaluation module, an independence hypothesis evaluation module and an exclusivity hypothesis evaluation module connected in sequence. The correlation hypothesis evaluation module is configured to evaluate whether the genetic instrumental variable is effective according to the effect estimate of the genetic instrumental variable, the strength and the significance test. When the strength is greater than or equal to 10, the tool variable is generally considered strong and can effectively predict the exposure variable. When the strength is less than 10, it may indicate that the tool variable is weak, and there may be a weak tool variable bias.

[0081] The independence hypothesis evaluation module is configured to test whether the genetic instrumental variable is related to the confounding factors and whether the independence hypothesis is satisfied. The evaluation steps are as follows: Check the correlation between the tool variable and the confounding factors: correlation test is performed using known confounding factors (such as age, gender, socioeconomic status, lifestyle habits, etc.). Correlation analysis or regression analysis can be used to evaluate whether there is a significant relationship between the tool variable and the confounding factors.

[0082] Evaluate the random allocation characteristics of the tool variable: Since Mendelian randomization is based on the random allocation characteristics of genetic variation, the genetic tool variable should be independent of the confounding factors in theory. The hypothesis can be evaluated by horizontal multiplicity test.

[0083] Use IVW and MR-Egger methods: Use the MR-Egger regression method to test whether the tool variable affects the outcome variable through other paths, thereby detecting potential confounding factors. If the intercept of MR-Egger is significantly not zero (p<0.05), it indicates that the tool variable may have horizontal multiplicity (i.e. the tool variable may affect the outcome variable through other paths, violating the independence hypothesis).

[0084] Result interpretation: If there is no significant relationship between the tool variable and the confounding factors (p>0.05), it indicates that the independence hypothesis is satisfied. If the tool variable is found to be related to the confounding factors through regression analysis, or the MR-Egger method detects a significant intercept, the selection of the tool variable may need to be re-considered, or a robust method may be used for analysis.

[0085] an exclusivity assumption evaluation module configured to evaluate whether the genetic instrumental variable only affects the outcome variable through the exposure variable.

[0086] the evaluation step: checking whether the instrumental variable has pleiotropy: MR-Egger regression: evaluating whether the regression intercept is significantly non-zero. If the intercept term is significant (p<0.05), it indicates that the instrumental variable has pleiotropy, possibly affecting the outcome variable through other mechanisms.

[0087] pleiotropy test: if there are multiple instrumental variables, the Q statistic can be used to test whether the effects of individual instrumental variables are consistent. If p<0.05, it indicates that there is heterogeneity between the instrumental variables, which may affect causal inference.

[0088] using the MR-PRESSO method: the MR-PRESSO (Mendelian Randomization Pleiotropy RESidual Sum and Outlier) method can identify and remove abnormal SNPs with pleiotropy, reducing the impact of pleiotropy bias.

[0089] removing instrumental variables with pleiotropy: instrumental variables with pleiotropy identified by the above methods should be removed or adjusted to ensure that the effects of instrumental variables only affect the outcome variable through the exposure variable.

[0090] result interpretation: if the intercept of MR-Egger regression is not significant (p>0.05) and the instrumental variable does not show pleiotropy (the p value of Q statistic is greater than 0.05), it means that the exclusivity assumption is met, and the instrumental variable only affects the outcome through the exposure variable. If there is a significant intercept (p<0.05), it means that the instrumental variable may affect the outcome variable through other ways, violating the exclusivity assumption, which needs to be further adjusted or using other methods.

[0091] causal relationship output module, for explaining the analysis results and outputting the causal relationship between oral frailty and sarcopenia.

[0092] longitudinal change analysis module, for analyzing the change trend of oral frailty and sarcopenia of the elderly population at different time points by using deep learning model and combining time-varying confounding factor adversarial learning technology; Specifically, the longitudinal change analysis module includes the following steps when analyzing the change trend of oral frailty and sarcopenia of the elderly population at different time points by using deep learning model and combining time-varying confounding factor adversarial learning technology: 1) Obtain longitudinal data of the elderly population in the database at multiple time points about oral frailty and sarcopenia, and preprocess to select key features affecting the changes of oral frailty and sarcopenia, wherein the longitudinal data includes exposure variables, outcome variables and time-varying confounding factors; 2) Input the time-varying confounding factors into the generator to generate exposure variables without the influence of confounding factors, and the discriminator accepts the generated exposure variables and the original confounding factor data to evaluate the influence of the time-varying confounding factors on the exposure variables, and alternately trains the generator and the discriminator in an adversarial manner; 3) Construct a long short-term memory network time series model with the exposure variables, outcome variables and confounding factors at each time point as input data, and the predicted values of the exposure variables and outcome variables at future time points as output, and add an adversarial regularization term to the loss function of the long short-term memory network, and train the long short-term memory network time series model using the adversarial regularization; The expression of the loss function of the long short-term memory network is: In the formula, L represents the loss function, E represents the expected symbol, y t represents the true target value at the t time point, represents the predicted value of the model at the t time point, , are used to adjust the hyperparameters of the adversarial loss and the regularization term to the total loss, respectively, D represents the discriminator model, represents the output prediction of the long short-term memory network time series model, f represents the long short-term memory network time series model, represents the exposure variable at the t time point, represents the time-varying confounding factor at the t time point, represents the L2 regularization term.

[0093] 4) Use the trained long short-term memory network time series model to output the predicted values of the exposure variables and outcome variables at future time points, and analyze the change trend of oral frailty and sarcopenia of the elderly population at different time points according to the prediction results.

[0094] Specifically, use the trained long short-term memory network (LSTM) time series model to output the predicted values of the exposure variables and outcome variables at future time points, and analyze the change trend of oral frailty and sarcopenia of the elderly population at different time points according to the prediction results, which can be performed according to the following steps: 41) Using the trained LSTM model for prediction: After the LSTM model is trained, we can use it to make predictions for future time points. Specifically, the LSTM model has learned the temporal dependencies in the data and can predict future values of the exposure variable (e.g., oral frailty progression) and outcome variable (e.g., sarcopenia changes) based on the current time point and historical information.

[0095] Predicting exposure and outcome variables: Exposure variable (e.g., oral frailty): By inputting the current time point's exposure variable and relevant features (e.g., lifestyle, diet, chronic diseases) into the LSTM model, the model will output predicted values of oral frailty for a future time period.

[0096] Outcome variable (e.g., sarcopenia): Similarly, the LSTM model will output predicted values of sarcopenia based on the current time point's outcome variable and relevant features.

[0097] 42) Analyzing trends in oral frailty and sarcopenia changes: Based on the predicted values of exposure and outcome variables at future time points output by the LSTM model, we can analyze the trends. Specifically, we can do the following: Visualizing prediction trends: Time series plot: Use a time series plot to show the predicted trends of oral frailty and sarcopenia. The horizontal axis represents time, and the vertical axis represents the predicted values. By comparing the predicted results at different time points, we can clearly see the trends of oral frailty and sarcopenia in the future time period.

[0098] If the predicted values of oral frailty continue to rise while the predicted values of sarcopenia decrease, it may indicate that the negative impact of oral frailty on sarcopenia is intensifying.

[0099] If the trends of both show synchronization or reverse changes, further exploration of the interaction or causal relationship between the two can be conducted.

[0100] Rate of change analysis: Rate of change: Calculating the rate of change (e.g., using the difference or percentage change between two time points) of oral frailty and sarcopenia in each time period can help quantify the speed and magnitude of change. This way, we can derive the trends and rates of change between different time points.

[0101] Comparing trends across different groups: According to different group characteristics (e.g., age, gender, chronic diseases), we can analyze the trends of different groups. For example: Compare the differences in oral frailty and sarcopenia changes between healthy and chronic disease groups.

[0102] Analyze the progression of oral frailty and sarcopenia in different age groups to explore the impact of age on the trends of both.

[0103] For example, younger old groups may experience slower frailty and muscle mass decline than older old groups.

[0104] Causal inference based on prediction results: If the model has been causally modeled, further analysis can attempt to explore the causal relationship between oral frailty and sarcopenia. By observing the prediction results at different time points, combined with the time-varying confounding factors removed by counterfactual learning, explore which exposure factors (such as oral health, diet, exercise, etc.) have a significant causal effect on the change of sarcopenia.

[0105] For example, the long-term impact of oral frailty on sarcopenia can be quantified through causal effect estimates (such as ATE, Average Treatment Effect) output by the model.

[0106] Mechanism analysis module for analyzing the specific mechanism of oral frailty affecting sarcopenia using structural equation modeling.

[0107] Specifically, the mechanism analysis module includes the following when analyzing the specific mechanism of oral frailty affecting sarcopenia using structural equation modeling: 1) Determine latent variables and corresponding observed variables based on latent variables, where latent variables include oral frailty, sarcopenia, inflammatory factors, cognitive function, depression status, and nutritional status; define the causal relationship and path between latent variables and observed variables; 11) Determine latent variables: Oral frailty: Oral health status (measured by oral examination scores, number of teeth, masticatory function, etc.).

[0108] Sarcopenia: Muscle mass and function (measured by muscle mass, grip strength, gait speed, etc.).

[0109] Inflammatory factors: such as C-reactive protein (CRP), serum albumin, etc.

[0110] Cognitive function: assessed by cognitive function scales (such as MMSE).

[0111] Depression status: assessed by depression scales (such as PHQ-9).

[0112] Nutritional status: assessed by dietary intake, body weight changes, malnutrition screening tools, etc.

[0113] 12) Determine observed variables: For each potential variable, determine the corresponding observed variable (i.e., actual measured data). For example: Oral frailty indicators: such as number of missing teeth, chewing difficulty score, etc.

[0114] Sarcopenia indicators: such as grip strength measurement, muscle mass (measured by DXA or BIA), etc.

[0115] Inflammatory factors: such as serum CRP, white blood cell count, etc.

[0116] Cognitive function: such as MMSE score.

[0117] Depression status: such as PHQ-9 score.

[0118] Nutritional status: such as body weight, BMI, dietary survey data, etc.

[0119] Confounding factors: age, gender, education level, bone density, chronic disease history, etc. should be included as control variables in the model.

[0120] 13) Determine causal relationships and paths: Define the causal relationships between potential variables and determine the paths of mediating variables and confounding variables.

[0121] Causal relationship assumptions: Oral frailty → Inflammatory factors: Oral health problems may trigger or exacerbate inflammatory responses.

[0122] Oral frailty → Cognitive function: Oral frailty may affect the nutritional intake of older adults, thereby affecting cognitive function.

[0123] Oral frailty → Depression status: Oral health problems may increase the risk of depression.

[0124] Oral frailty → Nutritional status: Oral health problems may affect the ability to eat, leading to malnutrition.

[0125] Mediating paths: Inflammatory factors, cognitive function, depression, and nutritional status may further affect the occurrence of sarcopenia.

[0126] Control paths: Variables such as age, gender, education level, bone density, and chronic disease history will affect all paths, so they need to be included in the model control.

[0127] 2) Use factor analysis to verify the matching degree between potential variables and observed variables to ensure the reliability and effectiveness of the measurement results, set the causal paths between potential variables, and perform path analysis to estimate the path coefficients to obtain the structural equation model. Specifically, before establishing the structural equation model, it is necessary to evaluate the measurement model to ensure that the observed variables of each potential variable can accurately reflect its potential meaning.

[0128] 21) Check data suitability: Data distribution: Check if the data follows a normal distribution.

[0129] Missing value handling: Impute missing values in the data or use appropriate imputation methods (e.g., multiple imputation).

[0130] Multicollinearity: Ensure there is no severe multicollinearity problem among the independent variables.

[0131] 22) Evaluation of measurement model: Check the relationship between each latent variable and observed variable through Confirmatory Factor Analysis (CFA). Specifically check: Factor loading: Whether each observed variable can significantly reflect the latent variable.

[0132] Reliability and validity: Whether the measurement tool is reliable (e.g., Cronbach's α coefficient) and valid (e.g., construct validity).

[0133] 23) Building structural equation model: After the measurement model is verified, build the structural model, i.e., set the causal relationship (path) between latent variables. Use software such as AMOS, Mplus, LISREL or lavaan package in R for path analysis.

[0134] Path analysis: Direct effect: Analyze the direct impact of oral frailty on sarcopenia (i.e., path coefficient).

[0135] Indirect effect: Analyze the indirect impact of oral frailty on sarcopenia through mediating variables (e.g., nutrition, activity, psychological state).

[0136] Total effect: The total effect is the sum of direct and indirect effects, reflecting the overall impact of oral frailty on sarcopenia.

[0137] 3) Check the fit of the structural equation model, interpret the path coefficients, and analyze the impact mechanism of oral frailty on sarcopenia according to the path coefficients.

[0138] 31) Model fit evaluation: Check the fit of the model, common indicators include: Chi-Square Test: Used to test the fit of the model and the data, the larger the p-value, the better the fit.

[0139] RMSEA (Root Mean Square Error of Approximation): Less than 0.08 indicates a good fit.

[0140] CFI (Comparative Fit Index): Greater than 0.90 indicates a good model fit.

[0141] TLI (Tucker-Lewis Index): Greater than 0.90 indicates a good model fit.

[0142] If the model does not fit well, it can be adjusted through modification indices or by considering modifying paths and relationships.

[0143] 32) Results interpretation and analysis: Interpretation of path coefficients: Path coefficients (β): Used to explain the causal relationship between latent variables. For example, the direct path coefficient of oral frailty on sarcopenia is 0.3, indicating that oral frailty has a positive impact on sarcopenia, and the impact degree is 30%.

[0144] Indirect effects: For example, the indirect path coefficient of oral frailty on sarcopenia through nutritional status indicates that oral health may indirectly affect the occurrence of sarcopenia by affecting the nutritional status of the elderly.

[0145] Data visualization module 3, for generating data visualization charts; Specifically, to generate data visualization charts, various graphics libraries can be used, such as Matplotlib, Seaborn, Plotly, etc. These charts can help clearly show the relationship between oral frailty, sarcopenia, genetic instrumental variables, and other health-related data.

[0146] Interactive feedback module 4, for interacting with users and providing personalized feedback; Specifically, the core function of the interactive feedback module is to enhance user experience, improve the accuracy of analysis and user understanding through dynamic interaction and personalized feedback.

[0147] The goals of the interactive feedback module are as follows: Real-time user interaction: Allows users to input specific requirements, such as adjusting statistical thresholds, selecting specific SNPs, filtering specific population data, etc.

[0148] Personalized feedback: Based on user input data, provide specific analysis results such as SNP statistical information, significance analysis, causal inference results, etc.

[0149] Dynamic adjustment of parameters: users can adjust parameters such as effect size threshold, significance level, tool variable screening criteria, etc., and the system will dynamically update the results.

[0150] Visual feedback: through charts and data tables, help users understand the analysis results, such as the effect size of genetic tool variables, p-value distribution, F-statistics, etc.

[0151] In summary, with the above technical solutions of the present application, by comprehensively using data collection, influence analysis, longitudinal change analysis, causal inference and structural equation modeling, etc. Technical means, a systematic analysis framework is provided, so as to deeply study the influence of oral weakness on sarcopenia, not only improves the depth and breadth of the relationship between oral weakness and sarcopenia, but also provides a scientific, accurate and personalized solution for the health management of the elderly population.

[0152] In addition, the present application uses statistical methods and Mendelian randomization methods to analyze the relationship between oral weakness and sarcopenia, and reveals the potential causal relationship between the two. Especially in the selection of genetic tool variables and the support of multi-sample data, the accuracy and scientificity of causal inference are improved, and the interference of confounding factors is avoided.

[0153] In addition, the present application uses long short-term memory network (LSTM) model in deep learning, combined with time-varying confounding factor countermeasure learning technology, realizes the change trend prediction of oral weakness and sarcopenia at different time points. This technology effectively reduces the influence of time-varying confounding factors, and provides more accurate basis for future prediction and intervention.

[0154] In addition, the present application uses structural equation model (SEM) to analyze how oral weakness affects sarcopenia through multiple paths. Using factor analysis and path analysis, it reveals the direct effect of oral weakness on sarcopenia and its potential mediating mechanism, which can better understand the influence of oral weakness on sarcopenia.

[0155] The above only describes the preferred embodiments of the present application and does not limit the present application. Any modification, equivalent replacement, improvement, etc. within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. An analysis system for the effect of oral frailty on sarcopenia, characterized in that: It includes a data acquisition module, an impact analysis module, a data visualization module and an interactive feedback module connected in sequence; The data acquisition module is used to collect data related to oral frailty and sarcopenia; The data visualization module is used to generate data visualization charts; The interactive feedback module is used to interact with users and provide personalized feedback; The impact analysis module includes a correlation analysis module, a causal relationship analysis module, a longitudinal change analysis module and an action mechanism analysis module connected in sequence; The correlation analysis module is used to analyze the relationship between the oral frailty index and the sarcopenia index through statistical methods; The causal relationship analysis module is used to analyze the causal relationship between genetic instrumental variables and the risk of oral frailty and sarcopenia by combining the two-sample Mendelian randomization method with genome-wide association study data; The longitudinal change analysis module is used to analyze the changing trends of oral frailty and sarcopenia in the elderly population at different time points by using a deep learning model combined with time-varying confounding factor adversarial learning technology; The mechanism analysis module is used to analyze the specific mechanism by which oral frailty affects sarcopenia using a structural equation model.

2. The analysis system for the effect of oral frailty on sarcopenia according to claim 1, characterized in that: The data acquisition module includes an oral frailty index acquisition module, a sarcopenia index acquisition module, a health index acquisition module and a data preprocessing module connected in sequence; The oral frailty index collection module is used to collect oral frailty indicators of the elderly population, and the oral frailty indicators include the number of teeth, chewing function and swallowing function; The sarcopenia index collection module is used to collect sarcopenia indicators of the elderly population, and the sarcopenia indicators include grip strength, gait speed and muscle mass; The health index collection module is used to collect other relevant health data of the elderly population, and other relevant health data include inflammation indicators, cognitive function, depression status, nutritional status and activity ability; The data preprocessing module is used to perform missing value filling, outlier detection and data standardization on the collected data to ensure the quality and consistency of the data.

3. The analysis system for the effect of oral weakness on sarcopenia according to claim 1, characterized in that: The correlation analysis module includes the following steps when analyzing the relationship between the oral frailty index and the sarcopenia index by statistical methods: Calculate the basic statistics of each indicator variable and draw the frequency distribution diagram, histogram and box plot of the data based on the calculation results to check the distribution of the data; Evaluate the individual distributions of oral frailty indicators and sarcopenia indicators, and select a statistical method based on the evaluation results. Select a parametric statistical method when the evaluation results conform to a normal distribution, and a non-parametric statistical method when the evaluation results do not conform to a normal distribution. Selected statistical methods were used to analyze the linear or nonlinear relationship between oral frailty and sarcopenia indicators, and regression analysis was used to analyze the linear effect of oral frailty on sarcopenia.

4. The analysis system for the effect of oral frailty on sarcopenia according to claim 1, characterized in that: The causal relationship analysis module includes a genetic instrument variable selection module, a Mendelian randomization analysis module, an analysis result verification module and a causal relationship output module connected in sequence; The genetic instrumental variable selection module is used to obtain association data of oral frailty and sarcopenia from two independent genome-wide association study data sets, and to extract single nucleotide polymorphism sites associated with oral frailty and sarcopenia as genetic instrumental variables respectively; The Mendelian randomization analysis module is used to evaluate the effects of genetic instrumental variables on oral frailty and sarcopenia, and to infer the causal relationship between oral frailty and sarcopenia using multiple Mendelian randomization methods; The analysis result verification module is used to evaluate whether the assumptions of Mendelian randomization analysis are valid to ensure the reliability of causal inference; The causal relationship output module is used to interpret the analysis results and output the causal relationship between oral weakness and sarcopenia.

5. The analysis system for the effect of oral weakness on sarcopenia according to claim 4, characterized in that: The genetic instrumental variable selection module is also used to evaluate the robustness of genetic instrumental variables to avoid bias and inefficiency caused by weak instrumental variables, enhance the statistical power and robustness of causal inference, and improve the credibility of research results; Among them, the evaluation formula for the effectiveness of genetic instrumental variables is: Where, F represents the robustness of the genetic instrumental variable, R 2 represents the degree of variation explained by the genetic instrumental variable on the exposure variable, n represents the number of samples, k represents the number of genetic instrumental variables.

6. The analysis system for the effect of oral frailty on sarcopenia according to claim 4, characterized in that: The Mendelian randomization analysis module includes the following when evaluating the effects of genetic instrumental variables on oral frailty and sarcopenia and using multiple Mendelian randomization methods to infer the causal relationship between oral frailty and sarcopenia: Obtain the effect estimates, standard errors, and statistical significance of genetic instrumental variables associated with oral frailty from genome-wide association study data, and align the effect allele directions to ensure that the effect directions of genetic instrumental variables are consistent across different datasets; Obtain effect estimates, standard errors, and statistical significance of genetic instrumental variables associated with sarcopenia from genome-wide association study data of sarcopenia, and check the consistency of effect allele direction to ensure that the effect allele direction in oral frailty and sarcopenia data is consistent; The weighted mean effect size of the genetic instrumental variables was calculated using the inverse variance weighting method, and the preliminary causal impact of oral frailty on sarcopenia was analyzed based on the weighted mean effect size. Horizontal pleiotropy bias was tested using regression adjustment to examine whether the genetic instrumental variables affected other pathways. A weighted median method was used to improve robustness when some genetic instrumental variables were ineffective. Based on the outlier adjustment method, abnormal genetic instrumental variables with a greater impact on the results were identified and removed, and sensitivity analysis was performed on the removed genetic instrumental variables to summarize the causal effects and obtain the causal relationship between oral frailty and sarcopenia.

7. The analysis system for the effect of oral frailty on sarcopenia according to claim 4, characterized in that: The analysis result verification module includes a correlation hypothesis evaluation module, an independence hypothesis evaluation module and an exclusivity hypothesis evaluation module which are connected in sequence; The correlation hypothesis evaluation module is used to evaluate whether the genetic instrument variable is effective based on the effect estimate, robustness and significance test of the genetic instrument variable; The independence hypothesis evaluation module is used to test whether the genetic instrumental variables are related to the confounding factors and whether the independence hypothesis is met; The exclusive hypothesis evaluation module is used to evaluate whether the genetic instrumental variable affects the outcome variable only through the exposure variable.

8. The analysis system for the effect of oral frailty on sarcopenia according to claim 1, characterized in that: The longitudinal change analysis module uses a deep learning model combined with time-varying confounding factor adversarial learning technology to analyze the changing trends of oral frailty and sarcopenia in the elderly population at different time points, including: Obtain longitudinal data on oral frailty and sarcopenia in the elderly population at multiple time points in the database, perform preprocessing, and select key features that influence changes in oral frailty and sarcopenia. Longitudinal data include exposure variables, outcome variables, and time-varying confounders. The time-varying confounders are input into the generator to generate exposure variables without the influence of the confounders. The discriminator receives the generated exposure variables and the original confounder data, evaluates the impact of the time-varying confounders on the exposure variables, and alternately trains the generator and the discriminator in an adversarial manner. The exposure variables, outcome variables, and confounding factors at each time point are used as input data, and the predicted values ​​of the exposure variables and outcome variables at future time points are used as output to construct a long-short-term memory network time series model. An adversarial regularization term is added to the loss function of the long-short-term memory network, and the adversarial regularization is used to train the long-short-term memory network time series model. The trained long short-term memory network temporal model was used to output the predicted values ​​of exposure variables and outcome variables at future time points, and the changing trends of oral frailty and sarcopenia in the elderly population at different time points were analyzed based on the predicted results.

9. The analysis system for the effect of oral frailty on sarcopenia according to claim 8, characterized in that: The loss function of the long short-term memory network is expressed as: Where, L represents the loss function, E Indicates the expectation symbol, y t Indicates the t The actual target value at a time point, Indicates the t The predicted value of the model at a time point, 、 They are used to adjust the hyperparameters of the adversarial loss and regularization term on the total loss, D represents the discriminator model, represents the output prediction of the long short-term memory network timing model, f represents the long short-term memory network temporal model, Indicates the t Exposure variables at time points, Indicates the t Time-varying confounders at time points, represents the L2 regularization term.

10. The analysis system for the effect of oral frailty on sarcopenia according to claim 1, characterized in that: The mechanism analysis module includes the following when using the structural equation model to analyze the specific mechanism by which oral frailty affects sarcopenia: Identify latent variables and determine corresponding observed variables based on the latent variables, including oral frailty, sarcopenia, inflammatory factors, cognitive function, depression, and nutritional status; define the causal relationship and path between the latent variables and the observed variables; Factor analysis was used to verify the matching degree between latent variables and observed variables to ensure the reliability and validity of the measurement results. The causal paths between latent variables were set and path analysis was performed to estimate the coefficients of each path and obtain the structural equation model. The goodness of fit of the structural equation model was tested, the path coefficients were interpreted, and the mechanism of the impact of oral frailty on sarcopenia was analyzed based on the path coefficients.

Images