Human immune system physiological age prediction method based on machine learning

Abstract

The invention discloses a human immune system physiological age prediction method based on machine learning, and the method comprises the steps: specifically collecting blood detection health data, carrying out the modeling of a computer model for male and female data according to the blood detection health data, building a computer model of a male and female human immune system, and carrying out the prediction of the physiological age of the human immune system. The method is used for evaluating physiological ages of human immune systems of individuals with different genders. According to the human body immune system physiological age prediction method based on machine learning, core data only uses blood detection indexes related to the human body immune system for modeling, the physiological age model of the human body immune system is constructed, and the human body immune physiological age can be evaluated in a targeted mode; moreover, according to the human immune system physiological age prediction method based on machine learning, modeling is carried out for men and women based on the data difference of the genders of the men and the women, so that the prediction result is more accurate.

Description

Technical Field

[0001] The present invention relates to the technical field of machine learning, and in particular to a method for predicting the physiological age of a human immune system based on machine learning with good prediction effect. Background Art

[0002] The immune system is like a loyal guard of the human body, guarding our health. It is composed of immune organs (such as bone marrow, thymus, etc.), immune cells (lymphocytes, phagocytes, etc.) and immune molecules (antibodies, complement, etc.). It has the functions of immune defense to block the invasion of pathogens; immune surveillance to detect and remove abnormal cells in time; immune self-stabilization to maintain the balance of the internal environment. When pathogens attack, the innate immunity responds quickly, and then the adaptive immunity attacks accurately to produce memory cells. The efficient operation of the immune system ensures the normal functioning of the body and enables us to resist the invasion of various diseases. The immune system age is a quantitative indicator based on the relevant indicators of blood tests to evaluate the degree of physiological aging of an individual's immune system. It reflects the actual functional age of the immune system and can be used to measure the health status and aging rate of the immune system. Referring to the article published in Nature Medicine by Tian, ​​YE et al. in 2023, the physiological age of various organs and systems of patients with diseases is significantly higher than that of healthy individuals, and bad living habits (smoking, drinking, staying up late, etc.) are also associated with higher physiological age.

[0003] Most of the existing technical solutions use all physical examination indicators to construct machine learning models, without dividing the relevant indicators into corresponding organs or systems. However, different organs or systems of an individual often have different physiological age indexes. A single overall model is difficult to reflect the physiological age of different organs or systems of an individual. In addition, some indicators differ greatly between men and women, so a single model is difficult to reflect the physiological age of individuals of different genders and different organs or systems.

[0004] Therefore, it is necessary to propose an improvement to overcome the defects of the prior art. Summary of the invention

[0005] The purpose of the present invention is to solve the problems in the prior art and provide a method for predicting the physiological age of the human immune system based on machine learning with good prediction effect.

[0006] The technical solution of the present invention is:

[0007] A method for predicting the physiological age of the human immune system based on machine learning comprises the following steps: S1, collecting health examination data: the health examination data comprises blood test data, gender and calendar age; S2, dividing the health examination data into a male data set and a female data set according to gender; S3, dividing the male data set and the female data set into a training set and a test set in proportion; S4, homogenizing the data of the training set: homogenizing the data in the training set in the male data set and the female data set respectively; S5, modeling and training the model: modeling the data in the training set in the male data set and the female data set respectively, and Cross-validation hyperparameter optimization was performed by grid search; S6. Prediction and correction of the physiological age of the human immune system: age correction was performed according to the regression center effect, calendar age was used as the independent variable, and the original age difference was used as the dependent variable for linear regression; original age difference = slope * calendar age + intercept + residual; where the residual is the age difference after correction, that is, the part related to calendar age is removed from the original age difference, and the formula is: corrected age difference = original age difference – (slope * original age difference + intercept); where the slope for males is -0.9039, the intercept for males is 48.17, and the slope for females is -0.9108, and the intercept for females is 47.41.

[0008] As a preferred technical solution, the following steps are also included after step S6: S7, calculating the aging index: the aging index calculation formula is: aging index = corrected age difference / calendar age.

[0009] As a preferred technical solution, the blood test data in step S1 includes one or more of the following blood indicators: red blood cell count, white blood cell count, platelet count, hemoglobin content and C-reactive protein concentration.

[0010] As a preferred technical solution, the ratio of the data volume of the training set to that of the test set in step S3 is 3:1.

[0011] As a preferred technical solution, in step S4, the data of the training set is normalized, and the normalization method can be any one of Z-score normalization, maximum and minimum normalization, median normalization, and maximum absolute value normalization.

[0012] As a further preferred technical solution, in step S4, the data of the training set is normalized by maximum absolute value normalization, and the formula is x′=x / max(|x|), where x is the original value and max(|x|) is the maximum absolute value of all values ​​in the data set.

[0013] As a preferred technical solution, in step S5, the data of the training sets in the male data set and the female data set are modeled respectively, and cross-validation hyperparameter optimization is performed in a grid search manner, which specifically includes the following steps: S51, creating a model; S52, defining a parameter grid: the parameter grid includes a penalty coefficient, a kernel function, a kernel function coefficient and a slack variable; S53, performing hyperparameter tuning; S54, obtaining the optimal parameters and model.

[0014] As a further preferred technical solution, step S54 further includes the following steps: S55, performing evaluation and verification on the training set and / or the test set.

[0015] As a further preferred technical solution, the modeling in step S5 adopts one or more of the following regression machine learning models: SVR, LightGBM, CatBoost, Ridge, RandomForest, XGBoost, Huber, Lasso, ElasticNet.

[0016] As a preferred technical solution, the slope and intercept in step S6 are calculated, and the specific calculation method is: linearly fit the original age difference and calendar age in the training sets of the male data set and the female data set respectively to obtain the slope and intercept in the male data set and the female data set.

[0017] The present invention discloses a method for predicting the physiological age of the human immune system based on machine learning. By collecting blood test health data in a targeted manner, computer models are built for male and female data based on the blood test health data to construct computer models of the male and female human immune systems, so as to evaluate the physiological age of the human immune system of individuals of different genders. The present invention discloses a method for predicting the physiological age of the human immune system based on machine learning. Since the core data only uses blood test indicators related to the human immune system for modeling, a physiological age model of the human immune system is constructed, which can evaluate the physiological age of human immunity in a targeted manner; and the present invention discloses a method for predicting the physiological age of the human immune system based on machine learning to model male and female data based on the data differences between male and female, so that the prediction results are more accurate. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 This is a flowchart of a specific implementation method of a method for predicting the physiological age of the human immune system based on machine learning in the present invention;

[0019] Figure 2 The prediction performance of the male human immune physiological age model in the test set in a human immune system physiological age prediction method based on machine learning in this embodiment;

[0020] Figure 3 The prediction performance of the female human immune physiological age model in the test set in a human immune system physiological age prediction method based on machine learning in this embodiment;

[0021] Figure 4 The age difference of males after the age prediction correction in step S6 in the human immune system physiological age prediction method based on machine learning in this embodiment;

[0022] Figure 5 It is the age difference of women after the age prediction correction is performed in step S6 in the human immune system physiological age prediction method based on machine learning in this embodiment. DETAILED DESCRIPTION

[0023] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are 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.

[0024] The terms used in the embodiments of the present invention are only for the purpose of describing specific embodiments, and are not intended to limit the present invention. The singular forms "a", "said" and "the" used in the embodiments of the present invention and the appended claims are also intended to include plural forms, unless the context clearly indicates other meanings, and "multiple" generally includes at least two, but does not exclude the inclusion of at least one.

[0025] It should be understood that the term "and / or" used in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship.

[0026] As used herein, the words "if" and "if" may be interpreted as "at the time of" or "when" or "in response to determining" or "in response to detecting", depending on the context. Similarly, the phrases "if it is determined" or "if (stated condition or event) is detected" may be interpreted as "when it is determined" or "in response to determining" or "when detecting (stated condition or event)" or "in response to detecting (stated condition or event)", depending on the context.

[0027] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a product or system including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such a product or system. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the product or system including the elements.

[0028] like Figure 1 The figure shows a specific implementation of a method for predicting the physiological age of the human immune system based on machine learning of the present invention. The method for predicting the physiological age of the human immune system based on machine learning of this embodiment comprises the following steps:

[0029] S1. Collecting health examination data: The health examination data includes blood test data, gender and calendar age;

[0030] S2, divide the health examination data into male data set and female data set according to gender;

[0031] S3, divide the male data set and the female data set into training set and test set respectively according to the proportion;

[0032] S4. Perform homogenization on the data of the training set: perform homogenization on the data in the training set of the male data set and the female data set respectively;

[0033] S5. Modeling and model training: Model the training set data in the male and female datasets respectively, and perform cross-validation hyperparameter optimization by grid search;

[0034] S6. Perform prediction and correction of the physiological age of the human immune system: perform age correction based on the regression center effect, use calendar age as the independent variable and the original age difference as the dependent variable for linear regression;

[0035] Original age difference = slope * calendar age + intercept + residual;

[0036] Among them, the residual is the age difference after correction, that is, the part related to calendar age is removed from the original age difference. The formula is:

[0037] Corrected age difference = original age difference – (slope * original age difference + intercept);

[0038] Among them, the slope for males is -0.9039 and the intercept for males is 48.17, while the slope for females is -0.9108 and the intercept for females is 47.41.

[0039] In this embodiment, the following steps are also included after step S6:

[0040] S7. Calculate the aging index: The calculation formula for the aging index is:

[0041] Aging index = adjusted age difference / calendar age.

[0042] Specifically, this implementation of a method for predicting the physiological age of the human immune system based on machine learning has 953 male individual samples with an age distribution range of 15-90 years old; the number of female individual samples is 884 with an age distribution range of 15-87 years old.

[0043] In this embodiment, a method for predicting the physiological age of the human immune system based on machine learning is used. The blood test data in step S1 includes one or more of the following blood indicators: red blood cell count, white blood cell count, platelet count, hemoglobin content, and C-reactive protein concentration. The above blood test data are closely related to the human immune system. Red blood cells are mainly responsible for transporting oxygen, providing necessary energy support for the activities of immune cells, and ensuring the normal operation of the immune system; white blood cells are key components of the immune system, such as neutrophils that can phagocytize bacteria, and lymphocytes that participate in specific immune responses to resist pathogen invasion; in addition to coagulation, platelets can also release immune-related mediators and participate in inflammatory responses; C-reactive protein rises rapidly when inflammation occurs, and is an important sign of immune system activation; hemoglobin maintains the normal metabolism and function of immune cells by transporting oxygen, indirectly ensuring the effectiveness of the immune system.

[0044] In the method for predicting the physiological age of the human immune system based on machine learning in this embodiment, the ratio of the data volume of the training set and the test set in step S3 is 3:1. After division: among male individuals, the number of training set samples is 715, and the number of test set samples is 238; among female individuals, the number of training set samples is 663, and the number of test set samples is 221. .

[0045] In the present invention, a method for predicting the physiological age of the human immune system based on machine learning is used. In a specific application process, in step S4, the data of the training set is normalized, and the normalization method can be any one of Z-score normalization, maximum and minimum normalization, median normalization, and maximum absolute value normalization. In the present embodiment, a method for predicting the physiological age of the human immune system based on machine learning is used. In step S4, the data of the training set is normalized, and the normalization method is maximum absolute value normalization. The formula is x′=x / max(|x|), where x is the original value, and max(|x|) is the maximum absolute value of all values ​​in the data set.

[0046] In the present embodiment, a method for predicting the physiological age of the human immune system based on machine learning is provided. In step S5, the data of the training set in the male data set and the female data set are modeled respectively, and cross-validation hyperparameter optimization is performed by means of grid search. Specifically, the following steps are included:

[0047] S51, create modeling;

[0048] S52, defining a parameter grid: the parameter grid includes a penalty coefficient, a kernel function, a kernel function coefficient, and a slack variable;

[0049] S53, perform hyperparameter tuning;

[0050] S54. Obtain optimal parameters and models.

[0051] The method for predicting the physiological age of the human immune system based on machine learning of the present invention, in actual application, further comprises the following steps after step S54: S55, performing evaluation and verification on the training set and / or the test set. The method for predicting the physiological age of the human immune system based on machine learning of this embodiment, step S55 is to perform evaluation and verification on both the training set and the test set.

[0052] The present invention is a method for predicting the physiological age of the human immune system based on machine learning. In specific applications, the modeling in step S5 can adopt one or more of the following regression machine learning models: SVR, LightGBM, CatBoost, Ridge, RandomForest, XGBoost, Huber, Lasso, ElasticNet. In this embodiment, a method for predicting the physiological age of the human immune system based on machine learning, step S5 adopts SVR to build a regression machine learning model. The specific implementation of step S5 of the method for predicting the physiological age of the human immune system based on machine learning in python with male data, maximum absolute value normalization, and SVR modeling is as follows:

[0053] Import the required Python packages:

[0054] import numpy as np

[0055] import pandas as pd

[0056] from sklearn.model_selection import train_test_split,GridSearchCV

[0057] from sklearn.pipeline import Pipeline

[0058] from sklearn.preprocessing import StandardScaler

[0059] from sklearn.svm import SVR

[0060] from sklearn.metrics import mean_squared_error

[0061] from scipy.stats import pearsonr

[0062] #Read in modeling data:

[0063] data = pd.read_csv ("Immune System Modeling Data.csv", header = 0, index_col = 0) # Get male data

[0064] maledf=data[data["gender"]=="male"]

[0065] #Get the detection value matrix of 6 indicators:

[0066] X=maledf[["Cprotein","WhiteCell","RedCell","PLT","HGB"]]

[0067] #Get age

[0068] y = maledf["age"]

[0069] # Split into training set and test set

[0070] X_train,X_test,y_train,y_test=train_test_split(X,y,test_size=0.25,random_state=0)

[0071] #Create modeling Pipeline:

[0072] pipeline=Pipeline([

[0073] ("scaler",MaxAbsScaler()),

[0074] ("svm",SVR()) ])

[0076] #Define the parameter grid:

[0077] param_grid = {

[0078] "svm__C":[0.1,1,10,100],#penalty coefficient

[0079] "svm__kernel":["linear","rbf"],#kernel function

[0080] "svm__gamma":["scale","auto"],#kernel function coefficient

[0081] "svm__epsilon":[0.1,0.2,0.5],#slack variables

[0082] }

[0083] #Use GridSearchCV for hyperparameter tuning:

[0084] grid_search = GridSearchCV(

[0085] estimator = pipeline,

[0086] param_grid = param_grid,

[0087] cv=5,

[0088] scoring="neg_mean_squared_error",

[0089] verbose=1 )

[0091] grid_search.fit(X_train,y_train)

[0092] #Get the best parameters and model:

[0093] best_params=grid_search.best_params_

[0094] best_model=grid_search.best_estimator_

[0095] #Evaluate on the training set

[0096] y_train_pred=best_model.predict(X_train)

[0097] train_mae=mean_squared_error(y_train,y_train_pred)train_r,train_p=pearsonr(y_train,y_train_pred)

[0098] #Evaluate on the test set:

[0099] y_test_pred=best_model.predict(X_test)

[0100] test_mae=mean_squared_error(y_test,y_test_pred)

[0101] test_r,test_p=pearsonr(y_test,y_test_pred)

[0102] In the method for predicting physiological age of human immune system based on machine learning in this embodiment, the slope and intercept in step S6 are calculated, and the specific calculation method is: linearly fit the original age difference and calendar age in the training set of male data set and female data set respectively to obtain the slope and intercept in male data set and female data set. In this embodiment, according to the above calculation method, the slope of male is -0.9039, the intercept of male is 48.17, the slope of female is -0.9108, and the intercept of female is 47.41.

[0103] The human immune system physiological age prediction method based on machine learning in this embodiment has achieved good performance in practical applications. In the test set, Figure 2 As shown in Figure 2, the Pearson correlation coefficient between the physiological age predicted by the male human immune physiological age model and the calendar age reached 0.35. Figure 3 As shown in Figure 1, the correlation coefficient between the female biological age predicted by the female human immune physiological age model and the calendar age reached 0.24, both showing relatively good modeling results. However, both models showed the "regression trend effect" in statistics, that is, for young individuals, the predicted biological age is often higher than the calendar age; for older individuals, the predicted biological age is often lower than the calendar age. After the age prediction correction in step S6, the corrected age difference between males and females is as follows: Figure 4 and Figure 5 As shown, the MAE (mean absolute error) after correction for males is 2.60, and the MAE after correction for females is 2.34, achieving very good results.

[0104] The method for predicting the physiological age of the human immune system based on machine learning in this embodiment also defines and calculates the aging index through step S7. The aging index represents the age of an individual each year and can evaluate the aging rate. If the aging index is greater than 1, it means that the individual is aging faster than the normal speed, and the aging speed is faster; if the aging index is less than 1, it means that the individual is aging slower than the normal speed, and the aging speed is slower.

[0105] The present invention discloses a method for predicting the physiological age of the human immune system based on machine learning. By collecting blood test health data in a targeted manner, computer models are built for male and female data based on the blood test health data to construct computer models of the male and female human immune systems, so as to evaluate the physiological age of the human immune system of individuals of different genders. The present invention discloses a method for predicting the physiological age of the human immune system based on machine learning. Since the core data only uses blood test indicators related to the human immune system for modeling, a physiological age model of the human immune system is constructed, which can evaluate the physiological age of human immunity in a targeted manner; and the present invention discloses a method for predicting the physiological age of the human immune system based on machine learning to model male and female data based on the data differences between male and female, so that the prediction results are more accurate.

[0106] The above description is only a preferred embodiment of the present invention and is not intended to limit the scope of implementation of the present invention. That is, all equivalent changes and modifications made according to the content of the patent application scope of the present invention should belong to the technical scope of the present invention.

Claims

1. A method for predicting the physiological age of the human immune system based on machine learning, characterized in that: The following steps are involved: S1. Collecting health examination data: The health examination data includes blood test data, gender and calendar age; S2, divide the health examination data into male data set and female data set according to gender; S3, divide the male data set and the female data set into training set and test set respectively according to the proportion; S4. Perform homogenization on the data of the training set: perform homogenization on the data in the training set of the male data set and the female data set respectively; S5. Modeling and model training: Model the training set data in the male and female datasets respectively, and perform cross-validation hyperparameter optimization by grid search; S6. Perform prediction and correction of the physiological age of the human immune system: perform age correction based on the regression center effect, use calendar age as the independent variable and the original age difference as the dependent variable for linear regression; Original age difference = slope * calendar age + intercept + residual; Among them, the residual is the age difference after correction, that is, the part related to calendar age is removed from the original age difference. The formula is: Corrected age difference = original age difference – (slope * original age difference + intercept); Among them, the slope for males is -0.9039 and the intercept for males is 48.17, while the slope for females is -0.9108 and the intercept for females is 47.

41.

2. The method for predicting the physiological age of the human immune system based on machine learning according to claim 1, characterized in that: After step S6, the following steps are also included: S7. Calculate the aging index: The calculation formula for the aging index is: Aging index = corrected age difference / calendar age.

3. The method for predicting the physiological age of the human immune system based on machine learning according to claim 1, characterized in that: The blood test data in step S1 includes one or more of the following blood indicators: red blood cell count, white blood cell count, platelet count, hemoglobin content and C-reactive protein concentration.

4. The method for predicting the physiological age of the human immune system based on machine learning according to claim 1, characterized in that: In step S3, the ratio of the data volume of the training set to that of the test set is 3:

1.

5. The method for predicting the physiological age of the human immune system based on machine learning according to claim 1, characterized in that: In step S4, the training set data is normalized, and the normalization method can be any one of Z-score normalization, maximum and minimum normalization, median normalization, and maximum absolute value normalization.

6. The method for predicting the physiological age of the human immune system based on machine learning according to claim 5, characterized in that: In step S4, the training set data is normalized by maximum absolute value normalization, and the formula is x′=x / max(|x|), where x is the original value and max(|x|) is the maximum absolute value of all values ​​in the data set.

7. The method for predicting the physiological age of the human immune system based on machine learning according to claim 1, characterized in that: In step S5, the data of the training set in the male data set and the female data set are modeled respectively, and cross-validation hyperparameter optimization is performed by grid search, which specifically includes the following steps: S51, create modeling; S52, defining a parameter grid: the parameter grid includes a penalty coefficient, a kernel function, a kernel function coefficient, and a slack variable; S53, perform hyperparameter tuning; S54. Obtain optimal parameters and models.

8. The method for predicting the physiological age of the human immune system based on machine learning according to claim 7, characterized in that: After step S54, the following steps are also included: S55. Perform evaluation and verification on the training set and / or test set.

9. A method for predicting the physiological age of the human immune system based on machine learning according to any one of claims 1, 7 or 8, characterized in that: In step S5, one or more of the following regression machine learning models are used for modeling: SVR, LightGBM, CatBoost, Ridge, RandomForest, XGBoost, Huber, Lasso, and ElasticNet.

10. The method for predicting the physiological age of the human immune system based on machine learning according to claim 1, characterized in that: The slope and intercept in step S6 are calculated. The specific calculation method is: linearly fit the original age difference and calendar age in the training sets of the male data set and the female data set respectively to obtain the slope and intercept in the male data set and the female data set.

Images