Prediction method, system and equipment for senile function decline and total cause death risk based on blood biomarkers and storage medium

Abstract

The invention discloses a blood biomarker-based old people function decline and total cause death risk prediction method, system and device, and a storage medium. The method comprises the following steps: establishing a healthy old people blood biomarker database; acquiring blood detection indexes of the participants from the database; screening out key blood indexes related to senile function decline by using an LASSO regression method; aiming at the screened key blood indexes, constructing a column diagram and obtaining a score prediction model; calculating a score value by using a score prediction model, and judging an individual risk level; key blood indexes are input into an input module in the prediction system, and a score prediction model calculates a score value according to an input numerical value; and the prediction result output module predicts the function decline and all-cause death risk level of the individual according to the obtained score value. According to the method, individual risks are effectively evaluated in a layered mode through the quantitative result of scoring, personalized intervention measures can be made conveniently, and the method has the advantages of being low in cost, easy to obtain, high in prediction accuracy and wide in application range.

Description

Technical Field

[0001] The present invention belongs to the field of medical testing and health risk assessment, and specifically relates to a prediction method, system, device and storage medium for the risk of functional decline and all-cause death in the elderly based on blood biomarkers, so as to realize quantitative assessment of functional decline and survival risk in the elderly. Background Art

[0002] With the acceleration of global population aging, frailty has become one of the most common and challenging health problems among the elderly. Frailty is a syndrome involving functional decline of multiple physiological systems, which is mainly manifested by increased susceptibility to environmental stress, significantly increasing the individual's risk of disability, hospitalization rate and mortality. Early identification and assessment of frailty can not only provide a scientific basis for clinical intervention, but also help improve the overall quality of life of the elderly and reduce the medical burden. However, current frailty assessment methods mostly rely on subjective scales and physical tests, such as the Frailty Phenotype and the Frailty Index. Although they have certain clinical reference value, they rely on patient self-report or functional measurement, and have the problems of strong subjectivity, complex operation and poor repeatability. In addition, these methods often fail to accurately capture the complex biological mechanisms behind frailty syndrome.

[0003] In recent years, the rapid development of biomarker research has provided new ideas for the prediction of frailty and all-cause mortality risk. Blood biomarkers have gradually become the focus of research due to their easy access, non-invasiveness, and quantification. However, a single blood marker has limited effect in predicting complex syndromes (such as frailty) and cannot fully reflect the multidimensional physiological changes associated with frailty. Summary of the invention

[0004] In order to solve the above-mentioned technical problems and accurately assess the risk of frailty and all-cause mortality in the elderly population, the present invention provides a prediction method, system, device and storage medium for the risk of functional decline and all-cause mortality in the elderly based on blood biomarkers. The method, system, device and storage medium have the advantages of low cost, easy acquisition and high prediction accuracy, and have a wide range of applications. It provides a scientific tool for the early identification and intervention of functional decline and all-cause mortality risk in the elderly.

[0005] The technical solutions adopted are as follows:

[0006] On the one hand, the present invention provides a method for predicting the risk of functional decline and all-cause mortality in the elderly based on blood biomarkers, and the specific method is: establishing a blood biomarker database of healthy elderly people;

[0007] Obtain participants’ blood test indicators from the database;

[0008] The LASSO regression method was used to screen out key blood indicators related to functional decline in the elderly;

[0009] For the screened key blood indicators, a nomogram was constructed and a scoring prediction model was obtained;

[0010] Obtain key blood indicators of the individual to be tested, use the scoring prediction model to calculate the score value, and determine the risk level of the individual to be tested.

[0011] Preferably, the blood biomarkers screened out by the LSSSO regression method include at least two blood indicators: cystatin C and creatinine.

[0012] Furthermore, the score prediction model constructed based on the two blood indicators of cystatin C and creatinine is:

[0013] CCCS=35.71429*cystatinC-22.27273*creatinine+46.31169

[0014] Among them: CCCS is the cystatin C-creatinine score;

[0015] cystatinC is the cystatin C value in the measured blood;

[0016] Creatinine is the creatinine value measured in the blood.

[0017] Preferably, the risk levels include: low risk group, low-medium risk group, medium-high risk group and high risk group;

[0018] Further preferably, when the CCCS score is lower than 58.5, it is determined to be a low-risk group;

[0019] When the CCCS score was between 58.5 and 73.0, it was determined to be in the low-medium risk group;

[0020] When the CCCS score is between 73.0 and 87.5, it is judged as the medium-high risk group;

[0021] When the CCCS score was higher than 87.5, it was judged as a high-risk group.

[0022] Furthermore, the prediction method also includes the steps of verifying and evaluating the score prediction model, specifically including:

[0023] ROC curve and TimeROC were used to evaluate the predictive ability of the score value for the risk of frailty and all-cause mortality;

[0024] Logistic and Cox regression models were used to verify the correlation between the score value and the risk of frailty and all-cause mortality;

[0025] The predictive effect of the score was validated in different subgroups and the applicability was evaluated.

[0026] On the other hand, the present invention also provides a prediction system for the risk of functional decline and all-cause mortality in the elderly based on blood biomarkers, which is placed in a mobile terminal, and the prediction system includes:

[0027] An input module, used to input key blood indicators;

[0028] A rating prediction model, which calculates a rating value based on the input value of the input module;

[0029] The prediction result output module predicts the individual's functional decline and all-cause mortality risk level based on the obtained score value.

[0030] Preferably, the key blood indices used in the scoring prediction model include at least two blood indices, cystatin C and creatinine. The calculation formula for calculating the score value based on cystatin C and creatinine is:

[0031] CCCS=35.71429*cystatinC-22.27273*creatinine+46.31169

[0032] Among them: CCCS is the cystatin C-creatinine score;

[0033] cystatinC is the cystatin C value in the measured blood;

[0034] Creatinine is the creatinine value measured in the blood.

[0035] The present invention also provides an electronic device, comprising:

[0036] processor;

[0037] a memory for storing processor-executable instructions;

[0038] Wherein, the processor is configured to implement the method for predicting the risk of functional decline and all-cause mortality in the elderly based on blood biomarkers as described in any one of claims 1-5 when executing the instructions.

[0039] The present invention also provides a computer-readable storage medium, comprising instructions, wherein the instructions instruct a device to execute the method for predicting the risk of functional decline and all-cause mortality in the elderly based on blood biomarkers as described in any one of claims 1-6.

[0040] The technical solution of the present invention has the following advantages:

[0041] A. In the method of the present invention, LASSO regression is used to screen out the key blood indicators with the most predictive value for frailty, a nomogram is used to establish a scoring prediction model, and a scoring value is obtained through the screened key blood indicators. Then, the risk level associated with frailty and all-cause mortality in the elderly is obtained according to the scoring value; the present invention integrates two key blood indicators, cystatin C and creatinine, through CCCS scoring, which can simultaneously reflect renal function and muscle metabolic status, evaluate frailty risk and its related biological mechanisms from multiple dimensions, and can effectively evaluate individual risks in a stratified manner through the quantitative results of the scoring, which is convenient for formulating personalized intervention measures. It has the advantages of low cost, easy acquisition, high prediction accuracy, and a wide range of applications, and provides a scientific tool for health management and intervention in the elderly.

[0042] B. The cystatin C and creatinine indicators used in the present invention are routine clinical detection items, which are easy to obtain and suitable for promotion and use in various medical institutions. The LASSO regression method is used to optimize the variable screening process, ensuring the scientific nature and predictive performance of the scoring.

[0043] C. The CCCS score in the method of the present invention has a stable predictive effect in different subgroups of the population (such as gender, age, and chronic disease status).

[0044] D. The prediction system provided by the present invention can quickly calculate the user's risk level through the input of key blood test indicators, further improving the practicality of CCCS scoring and providing efficient support for clinical decision-making and health management. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the specific embodiments of the present invention, the drawings required for use in the specific embodiments will be briefly introduced below. Obviously, the drawings described below are 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.

[0046] Figure 1 It is a flow chart provided by the present invention;

[0047] Figure 2 The process of screening out key blood indicators using LASSO regression method;

[0048] Figure 3 A frailty prediction nomogram constructed based on key blood indices screened by LASSO regression;

[0049] The nomogram was used to demonstrate the contribution of two key blood indices to frailty risk and could be used to calculate an individualized frailty risk score;

[0050] Figure 4Comparison of ROC curves between CCCS score and epigenetic clock model in frailty prediction;

[0051] Figure 5 Comparison of time-dependent ROC curves between CCCS score and epigenetic clock model in predicting all-cause mortality risk;

[0052] Figure 6 The CCCS score showed predictive effects in different subgroups;

[0053] (A) Participants aged <70 years; (B) Participants aged ≥70 years; (C) Male participants

[0054] (D) Female participants; (E) Unmarried participants; (F) Married participants; (G) Participants with lower education levels

[0055] (H) Participants with higher education level; (I) Participants without hypertension; (J) Participants with hypertension

[0056] (K) Participants without diabetes; (L) Participants with diabetes; (M) Participants without cancer

[0057] (N) Participants with cancer. DETAILED DESCRIPTION

[0058] The technical solution of the present invention will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0059] like Figure 1 As shown, the present invention provides a method for predicting the risk of functional decline and all-cause mortality in the elderly based on blood biomarkers, and the specific method is:

[0060]

S01

[0061] Participants with complete cystatin C and creatinine values ​​in the established database.

[0062]

S02

[0063]

S03

[0064] Figure 2 The LASSO regression method was used to screen key blood indicators associated with functional decline in the elderly from the participant database.

[0065] Firstly, more than 40 blood biomarkers were extracted from the 2016 VBS dataset, and LASSO regression was applied with "frailty" as the outcome variable.

[0066] Figure 2 (A): Coefficient path analysis: The horizontal axis is the logarithmic lambda value, and the vertical axis is the biomarker coefficient. When lambda is large, most coefficients are 0, among which the changes in the coefficients of cystatin C and creatinine indicate their importance.

[0067] Figure 2 (B): Cross-validation to select lambda: Use cross-validation to evaluate the model under different lambdas with prediction error. The lowest point of the curve corresponds to the optimal lambda, and the final model is determined, which clearly identifies cystatin C and creatinine as key markers.

[0068] This method effectively avoids the problem of multicollinearity of variables by introducing L1 regularization constraints, while optimizing the complexity of the model. The LASSO regression results show that cystatin C and creatinine are the most significant predictors, and they are stable in different subset data validations. [S04] For the selected key blood indicators, a nomogram was constructed and a score prediction model was obtained.

[0069] like Figure 3 As shown, the scoring prediction model constructed based on the two blood indicators of cystatin C and creatinine in the present invention is the CCCS index formula, and its calculation method is as follows:

[0070] CCCS=35.71429*cystatinC-22.27273*creatinine+46.31169.

[0071] Among them: CCCS is the cystatin C-creatinine score;

[0072] cystatinC is the cystatin C value in the measured blood;

[0073] Creatinine is the creatinine value measured in the blood.

[0074]

S05

[0075] The blood indexes of the individual to be tested are detected, and the obtained cystatin C data (cystatinC) and creatinine data (creatinine) are input into the above-mentioned score prediction model to obtain the CCCS score.

[0076] The present invention sets four risk levels, namely a low risk group, a low-medium risk group, a medium-high risk group and a high risk group.

[0077] According to the calculation formula of the above-constructed score prediction model, the cystatin C value is multiplied by a coefficient of 35.71429, the creatinine value is multiplied by a coefficient of -22.27273, and the sum of the above two values ​​plus 46.31169 can obtain the CCCS score.

[0078] According to the CCCS score values, the subjects were divided into four groups: low risk, low-medium risk, medium-high risk, and high risk to reflect different levels of frailty and death risk.

[0079] Among them: when the CCCS score is lower than 58.5, it is judged as a low-risk group;

[0080] When the CCCS score was between 58.5 and 73.0, it was determined to be in the low-medium risk group;

[0081] When the CCCS score was between 73.0 and 87.5, it was determined to be in the medium-high risk group;

[0082] When the CCCS score was higher than 87.5, it was judged as a high-risk group.

[0083]

S06

[0084]

S061

[0085] Figure 4 The ROC curve showed that CCCS showed a good prediction effect for frailty, with an AUC value of 0.687, which outperformed multiple epigenetic clocks.

[0086] Figure 5 The TimeROC curve showed that CCCS showed a good prediction effect on the risk of all-cause mortality, with an average AUC value of 0.733.

[0087]

S062

[0088] The CCCS scores obtained in the present invention are significantly correlated with the risk of physical decline and mortality in the elderly population, wherein the risk of decline and mortality in the high-scoring group is significantly higher than that in the low-scoring group. As shown in Table 1 below:

[0089] Table 1. Association between cystatin C-creatinine scores (categorized into low, low-intermediate, intermediate-high, and high risk categories according to frailty risk) and frailty and all-cause mortality

[0090]

[0091]

[0092] Among them: Model 1: uncorrected (no adjustment for relevant factors);

[0093] Model 2: adjusted for age, sex, marital status, and education level;

[0094] Model 3: adjusted for age, sex, marital status, education level, smoking status, alcohol consumption, albumin, bilirubin, alanine aminotransferase, and alkaline phosphatase.

[0095] Table 1 above divides the participants into four risk groups according to the frailty incidence predicted by the cystatin C-creatinine score in the nomogram: low risk (<58.5, frailty incidence <0.25), low-medium risk (58.5-73, frailty incidence 0.25-0.5), medium-high risk (73-87.5, frailty incidence 0.5-0.75) and high risk (>87.5, frailty incidence>0.75). Compared with the low-risk group, the risk of frailty in the low-medium risk group was 1.12 times higher (OR=1.12, 95% confidence interval: 1.07-1.18, p<0.001), the risk of frailty in the medium-high risk group was 1.31 times higher (OR=1.31, 95% confidence interval: 1.23-1.39, p<0.001), and the risk of frailty in the high-risk group increased by 1.48 times (OR=1.48, 95% confidence interval: 1.35-1.62, p<0.001). In terms of all-cause mortality, participants in the medium-high risk group had a 2.24-fold increased risk of death compared with the low-risk group (HR=2.24, 95% confidence interval: 1.25-4.00, p=0.007), and participants in the high-risk group had a 3.34-fold higher risk of death compared with the low-risk group (HR=3.34, 95% confidence interval: 1.75-6.38, p<0.001).

[0096]

S063

[0097] like Figure 6 As shown in the results, the cystatin C-creatinine score has a consistent predictive performance for frailty and all-cause mortality in different subgroups and is widely applicable. The score performs similarly in participants of different ages, genders, marital status, education levels, and those with or without chronic diseases such as hypertension, diabetes, and cancer.

[0098] In addition, the present invention also provides a prediction system for the risk of functional decline and all-cause death in the elderly based on blood biomarkers, which is placed in a mobile terminal, and the prediction system includes: an input module, a scoring prediction model and a prediction result output module. The input module is used to input key blood indicators; the scoring prediction model calculates the score value according to the input value of the input module; the prediction result output module predicts the individual's functional decline and all-cause death risk level according to the obtained score value.

[0099] The prediction system here can be an online calculation tool developed based on CCCS score. Users can quickly obtain individual frailty and death risk assessment results (Points Calculator) by inputting cystatin C and creatinine levels. The given assessment results can be used in the following aspects:

[0100] Clinicians can use it to screen high-risk patients and develop intervention measures; health managers can use it for dynamic health monitoring of the elderly population; researchers can use the CCCS score for exploratory studies of other elderly-related diseases.

[0101] From the above results, it can be seen that the CCCS score provided by the present invention can not only accurately reflect the renal function and muscle metabolism status of the elderly, but also effectively predict the risk of functional decline and all-cause mortality of individuals. The CCCS score is a comprehensive indicator that can easily quantify the health status of individuals and predict short-term and long-term health outcomes through non-invasive blood testing. It can also provide clinicians with an evaluation tool to assist in decision-making.

[0102] The CCCS score constructed by the present invention can deeply analyze the key factors of health decline in the elderly from multiple dimensions, and provide accurate risk assessment by combining the different physiological index levels and clinical characteristics of individuals. On the one hand, it can directly reflect the physiological state of individuals, and on the other hand, it can also effectively predict the decline process and survival risk, providing an easily accessible, operational, and highly applicable health assessment tool for clinical practice.

[0103] The electronic device provided by the present invention comprises at least a processor and a memory. The memory is used to store processor executable instructions, and the processor is configured to implement a prediction method for elderly functional decline and all-cause mortality risk based on blood biomarkers when executing the instructions.

[0104] Also included is a computer-readable storage medium comprising instructions that instruct a device to execute a method for predicting the risk of functional decline and all-cause mortality in the elderly based on blood biomarkers.

[0105] The program running in the electronic device may be a program that controls the central processing unit CPU and the like to realize the functions of the above-mentioned embodiment involved in one scheme of the present invention. Then, the information processed by these devices is temporarily stored in the random access memory RAM during the processing, and then stored in various ROMs such as read-only memory and hard disk drive HDD, and is read, modified, and written by the CPU as needed.

[0106] It should be noted that a part of the electronic device of the above embodiment can also be implemented by a computer. In this case, a program for implementing the control function can be recorded in a computer-readable recording medium, and the program recorded in the recording medium can be read into a computer system and executed to implement it.

[0107] It should be noted that the "computer system" mentioned here refers to a computer system built into an electronic device, and uses a computer system including hardware such as an OS and peripheral devices. In addition, "computer-readable recording medium" refers to removable media such as floppy disks, magneto-optical disks, ROMs, CD-ROMs, and storage devices such as hard disks built into computer systems.

[0108] Furthermore, "computer-readable recording media" may include: media that dynamically store programs for a short period of time, such as communication lines when sending programs via networks such as the Internet or communication lines such as telephone lines; and media that store programs for a fixed period of time, such as volatile memories inside computer systems that serve as servers or clients in this case. In addition, the above-mentioned program may be a program for realizing a part of the above-mentioned functions, or a program that can realize the above-mentioned functions by combining with a program already recorded in a computer system.

[0109] In addition, the electronic device in the above-mentioned embodiment can also be implemented as a collection (device group) composed of multiple devices. Each device constituting the device group can have a part or all of the functions or functional blocks of the electronic device in the above-mentioned embodiment. As a device group, it is sufficient to have all the functions or functional blocks of the electronic device.

[0110] Anything not described in the present invention is applicable to the prior art.

[0111] Obviously, the above embodiments are merely examples for the purpose of clear explanation, and are not intended to limit the implementation methods. For those skilled in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the implementation methods here. However, the obvious changes or modifications derived therefrom are still within the scope of protection of the present invention.

Claims

1. A method for predicting the risk of functional decline and all-cause mortality in the elderly based on blood biomarkers, characterized in that: Establish a blood biomarker database for healthy elderly people; Obtain participants’ blood test indicators from the database; The LASSO regression method was used to screen out key blood indicators related to functional decline in the elderly; For the screened key blood indicators, a nomogram was constructed and a scoring prediction model was obtained; Obtain key blood indicators of the individual to be tested, use the scoring prediction model to calculate the score value, and determine the risk level of the individual to be tested.

2. The method for predicting the risk of functional decline and all-cause mortality in the elderly based on blood biomarkers according to claim 1, characterized in that: The blood biomarkers screened out by the LSSSO regression method include at least two blood indicators: cystatin C and creatinine.

3. The method for predicting the risk of functional decline and all-cause mortality in the elderly based on blood biomarkers according to claim 2, characterized in that: The score prediction model constructed based on the two blood indicators of cystatin C and creatinine is: CCCS=35.71429*cystatinC-22.27273*creatinine+46.31169 Among them: CCCS is the cystatin C-creatinine score; cystatinC is the cystatin C value in the measured blood; Creatinine is the creatinine value measured in the blood.

4. The method for predicting the risk of functional decline and all-cause mortality in the elderly based on blood biomarkers according to claim 3, characterized in that: The risk levels include: low risk group, low-medium risk group, medium-high risk group and high risk group.

5. The method for predicting the risk of functional decline and all-cause mortality in the elderly based on blood biomarkers according to claim 4, characterized in that: in: When the CCCS score obtained was lower than 58.5, it was judged as the low-risk group; When the CCCS score was between 58.5 and 73.0, it was determined to be in the low-medium risk group; When the CCCS score was between 73.0 and 87.5, it was determined to be in the medium-high risk group; When the CCCS score was higher than 87.5, it was judged as a high-risk group.

6. The method for predicting the risk of functional decline and all-cause mortality in the elderly based on blood biomarkers according to claim 3, characterized in that: The prediction method also includes the steps of verifying and evaluating the score prediction model, specifically including: ROC curve and TimeROC were used to evaluate the predictive ability of the score value for the risk of frailty and all-cause mortality; Logistic and Cox regression models were used to verify the correlation between the score value and the risk of frailty and all-cause mortality; The predictive effect of the score was validated in different subgroups and the applicability was evaluated.

7. A prediction system for the risk of functional decline and all-cause mortality in the elderly based on blood biomarkers, which is placed on a mobile terminal and is characterized by: The prediction system comprises: An input module, used to input key blood indicators; A rating prediction model, which calculates a rating value based on the input value of the input module; The prediction result output module predicts the individual's functional decline and all-cause mortality risk level based on the obtained score value.

8. The prediction system for elderly functional decline and all-cause mortality risk based on blood biomarkers according to claim 7, characterized in that: The key blood indicators used in the scoring prediction model include at least two blood indicators, cystatin C and creatinine. The calculation formula for calculating the score value based on cystatin C and creatinine is: CCCS=35.71429*cystatinC-22.27273*creatinine+46.31169 Among them: CCCS is the cystatin C-creatinine score; cystatinC is the cystatin C value in the measured blood; Creatinine is the creatinine value measured in the blood.

9. An electronic device, characterized in that: include: processor; a memory for storing processor-executable instructions; Wherein, the processor is configured to implement the method for predicting the risk of functional decline and all-cause mortality in the elderly based on blood biomarkers as described in any one of claims 1-6 when executing the instructions.

10. A computer-readable storage medium, characterized in that: The method comprises instructions for instructing the device to execute the method for predicting the risk of functional decline and all-cause mortality in the elderly based on blood biomarkers as described in any one of claims 1 to 6.

Images