阅读说明：本技术 用于预测或评估健康个体认知功能的生物标志物组合及其应用 (Biomarker combination for predicting or evaluating cognitive function of healthy individual and application thereof ) 是由 朱振华 惠李 贾秋放 于 2021-08-12 设计创作，主要内容包括：本发明公开了一种用于预测或评估健康个体认知功能的生物标志物组合及其应用,涉及分子诊断技术领域。本发明公开的生物标志物组包括：肌酐、肌酸、鸟氨酸、苏氨酸、缬氨酸和谷氨酸；采用该生物标志物组预测或评估健康个体的认知功能,具有较高的准确率,通过对认知功能的预测有利于快速筛选出有中枢神经疾病的高危人群；本发明为预测或评估健康个体的认识功能提供了一种客观的技术手段,避免采用RBANS测评带来的受试者疲劳问题。(The invention discloses a biomarker combination for predicting or evaluating cognitive function of healthy individuals and application thereof, and relates to the technical field of molecular diagnosis. The biomarker panel disclosed by the invention comprises: creatinine, creatine, ornithine, threonine, valine, and glutamic acid; the biomarker group is adopted to predict or evaluate the cognitive function of healthy individuals, so that the accuracy is high, and the high risk group with central nervous system diseases can be quickly screened out by predicting the cognitive function; the invention provides an objective technical means for predicting or evaluating the cognitive function of a healthy individual, and avoids the problem of fatigue of a subject caused by RBANS evaluation.)

1. A biomarker panel for predicting or assessing cognitive function in a healthy individual, characterized in that it comprises a first marker panel comprising the following compounds: hypoxanthine, creatinine, ornithine, valine and arginine.

2. The biomarker combination according to claim 1, wherein when the cognitive function is predicted or assessed by the immediate memory dimension, the biomarker combination further comprises a second marker panel comprising the following compounds: creatinine, creatine, ornithine, threonine, valine, and glutamic acid.

3. The biomarker combination according to claim 1 or 2, characterized in that when the cognitive function is predicted or assessed by the attention dimension, the biomarker combination further comprises a third marker panel comprising the following compounds: creatine, ornithine, valine, phenylalanine and glutamic acid;

preferably, when the cognitive function is predicted or assessed by the visual breadth dimension, the biomarker combination further comprises a fourth marker panel comprising the following compounds: creatinine, creatine, isoleucine, lysine, citrulline, and glutamic acid;

preferably, when said cognitive function is predicted or assessed by language functional dimensions, said biomarker panel further comprises a fifth marker panel, hypoxanthine, creatinine, threonine, valine and tryptophan;

preferably, when the cognitive function is predicted or assessed by delayed memory dimensions, the biomarker panel further comprises a sixth marker panel comprising the following compounds: hypoxanthine, creatinine, creatine, ornithine, isoleucine and phenylalanine.

4. Use of a biomarker combination according to any of claims 1 to 3 in the manufacture of a kit for predicting or assessing cognitive function in a healthy individual.

5. Use of an agent for detecting a biomarker combination according to any of claims 1 to 3 in the manufacture of a kit for predicting or assessing cognitive function in a healthy individual.

6. A kit for predicting or assessing cognitive function in a healthy individual, comprising: reagents and/or consumables for detecting the concentration of each compound in a biomarker combination according to any of claims 1 to 3.

7. The kit according to claim 6, wherein the detection sample of the kit is plasma or serum.

8. An apparatus for predicting or assessing cognitive function in a healthy individual, comprising:

an information acquisition module for acquiring evaluation information including concentration information of the biomarker combination according to any one of claims 1 to 3 in a sample to be tested from a target healthy individual, and age information and education level information of the target healthy individual;

and the evaluation module is used for processing the information to be evaluated by using an evaluation model to obtain an evaluation result.

9. The device according to claim 8, characterized in that it comprises: the evaluation model is obtained by training an initial model by an evaluation information standard sample;

the assessment information standard samples comprise concentration information samples, age information, education level information and corresponding RBANS scale score samples of the biomarker combinations of a plurality of healthy individuals.

10. The apparatus of claim 9, wherein the initial model is a support vector machine, a K-nearest neighbor algorithm, or a neural network algorithm;

preferably, the concentration information is detected by an amino acid automatic analyzer, a capillary electrophoresis method, a gas chromatography-mass spectrometry tandem method, a liquid chromatography or a liquid chromatography-mass spectrometry tandem method;

the concentration information is represented by concentration values, ion abundance values or ratios of the concentration values and the ion abundance values of the compounds in the biomarker combination to an internal standard;

preferably, the sample to be tested is plasma or serum.

Technical Field

The invention relates to the technical field of molecular diagnosis, in particular to a biomarker combination for predicting or evaluating cognitive functions of healthy individuals and application thereof.

Background

Cognitive dysfunction is widely present in senile dementia, schizophrenia, bipolar disorder, depression or other central nervous disorders. Potential individuals with central nervous system diseases such as senile dementia, schizophrenia, bipolar disorder, depression and the like may exist in healthy people. The prediction of the cognitive function is beneficial to quickly screening high-risk people suffering from the central nervous system diseases from healthy people. At present, the cognitive function is mainly evaluated by adopting a repetitive complete set of psychoneural state evaluation tools.

The repetitive set of neuropsychological state assessment tools (RBANS) have been widely used in psychiatric clinical research at home and abroad since the completion of Randolph's compilation in 1998. The tool is a simple and single-person operated test, the whole test takes no more than 30 minutes, the cooperation of patients can be obtained to the maximum extent, and the influence of fatigue on the test result is reduced as much as possible. The test difficulty is moderate, the applicable population range is from normal people to moderate dementia, the test comprises 12 tests, and the test can be summarized into 5 groups of neuropsychological states: instant memory (immediate memory), attention (attention), visual breadth (visual/relational), language function (language function), and delayed memory (delayed memory). Immediate memory: subject's ability to remember in the short term after exposure to information, scores derived from story recall and vocabulary learning tests; attention is paid to: the subjects' ability to remember and to present information in short-term memory controlled by their vision and mouth, the scores coming from code and numerical breadth tests; language function: subjects demonstrated their ability to respond in language by recalling or naming known materials, with scores derived from language fluency and picture naming tests; visual breadth: the ability of the subject to be examined to perceive space and to construct a spatial copy of a particular drawing, the scores resulting from line location and graphic replication tests; and (3) delayed memory: the subjects were examined for antegrade memory, scores from recall of words, recognitions of words, memory of stories and recall of graphics.

However, when the RBANS is adopted to evaluate the cognitive function of the testee, the testee still can be fatigued, and at present, no objective technical means is provided for evaluating the cognitive function of healthy people.

In view of this, the invention is particularly proposed.

Disclosure of Invention

The invention aims to provide a biomarker combination for predicting or evaluating cognitive functions of healthy individuals and application thereof. The biomarker group provided by the invention can reflect the cognitive function damage state of a healthy individual, the cognitive function of the healthy individual is predicted or evaluated by adopting the biomarker group, the accuracy is higher, and the prediction of the cognitive function is favorable for quickly screening high-risk groups suffering from senile dementia, schizophrenia, bipolar affective disorder, depression or other central nervous diseases and the like; meanwhile, a relatively objective technical means is provided for evaluating the cognitive function of a healthy individual, and the problem of fatigue of a test subject caused by RBANS evaluation is effectively avoided.

The invention is realized by the following steps:

in one aspect, the present invention provides a biomarker panel for predicting or assessing cognitive function in a healthy individual, comprising a first marker panel comprising the following compounds: hypoxanthine, creatinine, ornithine, valine and arginine.

Research in the embodiments of the present invention shows that the biomarker combinations may reflect the cognitive function of healthy individuals as a whole; by detecting the biomarkers, the overall prediction or evaluation of the cognitive function of healthy individuals can be realized, and the accuracy is high. The prediction of the cognitive function of healthy people is beneficial to quickly screening out high-risk people suffering from senile dementia, schizophrenia, bipolar affective disorder, depression or other central nervous diseases and the like; the invention provides a relatively objective technical means for evaluating the cognitive function of a healthy individual, and effectively avoids the problem of fatigue of a test subject caused by RBANS evaluation.

In the invention, the healthy individual refers to a healthy body, a person who does not take lipid-lowering and sugar-lowering drugs and immunomodulators within 3 months, and a subject who does not have serious body or central system diseases or family history of mental diseases.

Alternatively, in some embodiments of the invention, the cognitive function may be predicted or assessed by the immediate memory, attention, visual breadth, language function, or delayed memory dimensions.

Optionally, in some embodiments of the invention, when the cognitive function is predicted or assessed by the immediate memory dimension, the biomarker combination further comprises a second marker panel comprising the following compounds: creatinine, creatine, ornithine, threonine, valine, and glutamic acid.

Optionally, in some embodiments of the invention, when the cognitive function is predicted or assessed by the attention dimension, the biomarker combination further comprises a third marker panel comprising the following compounds: creatine, ornithine, valine, phenylalanine and glutamic acid.

Optionally, in some embodiments of the invention, when the cognitive function is predicted or assessed by the visual breadth dimension, the biomarker combination further comprises a fourth marker panel comprising the following compounds: creatinine, creatine, isoleucine, lysine, citrulline, and glutamic acid.

Optionally, in some embodiments of the invention, when the cognitive function is predicted or assessed by a language functional dimension, the biomarker panel further comprises a fifth marker panel comprising the following compounds: hypoxanthine, creatinine, threonine, valine, and tryptophan.

Optionally, in some embodiments of the invention, when the cognitive function is predicted or assessed by delayed memory dimensions, the biomarker panel further comprises a sixth marker panel comprising the following compounds: hypoxanthine, creatinine, creatine, ornithine, isoleucine and phenylalanine.

Those skilled in the art know that dimensions reflecting cognitive function include: immediate memory, attention, visual breadth, language functionality, and delayed memory. In this regard, the present invention also provides a biomarker set correspondingly used in predicting or evaluating cognitive function for different dimensions, for example, when it is required to predict or evaluate cognitive function of a healthy individual from the memory dimensions, a second marker set may be used; when it is desired to predict or assess cognitive function in a healthy individual from the attention dimension, a third marker panel may be employed; when it is desired to predict or assess cognitive function of a healthy individual from the visual breadth dimension, a fourth marker panel may be employed; when the cognitive function of a healthy individual needs to be predicted or evaluated from language function dimensions, a fifth marker set may be employed; a sixth marker panel may be used when it is desired to predict or assess cognitive function in a healthy individual from the delayed memory dimension. Each marker group has higher accuracy, and the invention also provides a more objective technical means for evaluating different dimensions of the cognitive function of healthy individuals.

In another aspect, the invention provides the use of a biomarker combination as defined in any one of the above in the manufacture of a kit for predicting or assessing cognitive function in a healthy individual.

In another aspect, the invention provides the use of an agent for detecting a biomarker combination as defined in any of the above in the manufacture of a kit for predicting or assessing cognitive function in a healthy individual.

It is to be noted that the above-mentioned compounds are all well-known compounds in the art, and those skilled in the art will readily appreciate that the detection can be achieved by means of conventional techniques in the art, and that the reagents and/or consumables for detecting the biomarker combinations are also well-known in the art.

In another aspect, the present invention provides a kit for predicting or assessing cognitive function in a healthy individual, comprising: reagents and/or consumables for detecting each compound in the biomarker combination concentration as described in any one of the above.

Optionally, in some embodiments of the invention, the reagents and/or consumables are suitable for use in LC-MS/MS techniques for detecting the concentration of each compound in the biomarker combination.

Alternatively, in some embodiments of the invention, the test sample of the kit is plasma or serum.

In another aspect, the present invention provides an apparatus for predicting or assessing cognitive function in a healthy individual, comprising:

an information acquisition module for acquiring evaluation information including concentration information of the biomarker combination according to any one of claims 1 to 3 in a sample to be tested from a target healthy individual, and age information and education level information of the target healthy individual;

and the evaluation module is used for processing the information to be evaluated by using an evaluation model to obtain an evaluation result.

It should be noted that, those skilled in the art may select the corresponding biomarker concentration information according to the index to be evaluated.

Alternatively, in some embodiments of the invention, it comprises: the evaluation model is obtained by training an initial model by an evaluation information standard sample;

the assessment information standard samples comprise concentration information samples, age information, education level information and corresponding RBANS scale score samples of the biomarker combinations of a plurality of healthy individuals.

It should be noted that the specific number of the "multiple cases" can be reasonably selected by those skilled in the art according to actual situations, but is at least 3 cases, and of course, the larger the number, the more reliable the obtained prediction result, preferably more than 10 cases. Optionally, in some embodiments of the invention, 241 cases are selected for the invention.

The education level information refers to the education period of the target healthy individual.

Optionally, in some embodiments of the invention, the initial model is a support vector machine, a K-nearest neighbor algorithm, or a neural network algorithm.

Alternatively, in some embodiments of the invention, the concentration information is detected by an amino acid autoanalyzer, capillary electrophoresis, gas chromatography-mass spectrometry tandem, liquid chromatography, or liquid chromatography-mass spectrometry tandem.

Knowing the specific compound, those skilled in the art can readily detect its concentration by means of techniques that are conventional in the art, such as amino acid autoanalyzer, capillary electrophoresis, gas chromatography, tandem gas chromatography-mass spectrometry, liquid chromatography, or tandem liquid chromatography-mass spectrometry.

Optionally, in some embodiments of the invention, the concentration information is detected using LC-MS/MS techniques.

Alternatively, in some embodiments of the invention, the concentration information is represented by concentration values, ion abundance values, or ratios thereof to an internal standard for each compound in the biomarker combination.

Alternatively, in some embodiments of the invention, the chromatographic conditions for detecting said concentration information using LC-MS/MS techniques are as follows:

a chromatographic column: BEH Amide,1.7 μm,2.1 mm. times.100 mm, mobile phase A: water (containing 5mM ammonium formate, 5mM ammonium acetate and 0.2% formic acid), mobile phase B: acetonitrile (containing 1mM ammonium formate, 1mM ammonium acetate and 0.2% formic acid), flow rate 0.4mL/min, column temperature 35 ℃, gradient elution conditions:

0-3min,95％B；3-11min,95％B-70％B；11-13min,70％B-55％B；13-14min,55％B。

the numerical parameters of the above conditions may fluctuate within a range of ± 3%.

Alternatively, in some embodiments of the invention, the mass spectrometric conditions for detecting said concentration information using LC-MS/MS techniques are as follows:

the instrument comprises the following steps: triple quadrupole mass spectrometry, scan mode: multiple reaction monitoring (MRM/SRM), electrospray voltage: positive ion 3.5kv, negative ion 3kv, sheath gas: 35Arb, assist gas: 10Arb, ion source temperature: at 350 ℃.

The numerical parameters of the above conditions may fluctuate within a range of ± 3%.

Alternatively, in some embodiments of the invention, the sample to be tested is plasma or serum.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the embodiments will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present invention and therefore should not be considered as limiting the scope, and for those skilled in the art, other related drawings can be obtained according to the drawings without inventive efforts.

FIG. 1 is a graph of the results of prediction of immediate memory using biomarker combinations 1 and the regression model of example 1 in an example of the present invention; a: comparing the predicted score with the score of the scale, wherein the abscissa represents the serial number of the subject, the ordinate represents the predicted score, the dot represents the score of the scale, the triangle represents the predicted score, the light gray line represents the difference from the score of the scale is +/-10, and the dark gray line represents the difference from the score of the scale is +/-20; b: the difference value between the predicted score and the scale score is within 10 and is more than 10; fig. 2-a and B in fig. 6 are illustrated as such.

FIG. 2 is a graph showing the prediction results of the prediction of attention using the biomarker combinations 2 in the examples of the present invention and the regression model in example 2.

Fig. 3 is a result of predicting visual span using biomarker combinations 3 and the regression model of example 3 in an example of the present invention.

FIG. 4 is a graph of the results of a regression model prediction language function using biomarker combinations 4 and example 4 in an example of the present invention.

FIG. 5 is a graph of the prediction of delayed memory using biomarker combinations 5 in an example of the present invention and the regression model of example 5.

Fig. 6 is a prediction result of predicting cognitive function using the biomarker combinations 6 and the regression model of example 6 in the example of the present invention.

Detailed Description

In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below. The examples, in which specific conditions are not specified, were conducted under conventional conditions or conditions recommended by the manufacturer. The reagents or instruments used are not indicated by the manufacturer, and are all conventional products available commercially.

The features and properties of the present invention are described in further detail below with reference to examples.

Example 1

Screening and modeling of biomarkers for predicting or assessing instant memory in healthy individuals.

(1) Collecting peripheral venous blood of 241 healthy individual training set samples (information is shown in table 1), extracting blood plasma, and storing at-80 ℃ for later use; healthy subjects are healthy, who do not take lipid-lowering, sugar-lowering and immunomodulators within 3 months, and who have no serious physical or central system diseases, no family history of mental diseases or mental diseases. Taking 90 microliters of plasma, adding 10 microliters of internal standard solution, adding 400 microliters of extraction solution, performing vortex for 30 seconds, performing ultrasonic treatment for 5 minutes, performing centrifugation for 10 minutes at 13000rpm, and taking supernatant to obtain a sample to be detected;

preparation of internal standard solution: accurately weighing L-2-chlorophenylalanine, dissolving in methanol, and preparing into 10 μ g/mL solution;

preparation of extraction solution: mixing chromatographic grade dichloromethane and methanol according to the volume ratio of 2: 1;

quality control of plasma: taking the anticoagulation blood of a healthy individual, centrifuging at normal temperature for 10 minutes, transferring the supernatant into a new centrifuge tube, and mixing the anticoagulation blood with the plasma: the internal standard solution was added at a ratio of 9:1 and mixed.

(2) Carrying out qualitative and quantitative analysis on main metabolites in plasma by a metabonomics method and an LC-MS/MS technology;

injecting the metabolic extract into a chromatographic column by an automatic sample injector for separation, wherein the specific chromatographic conditions are as follows:

packing and specification of chromatographic column: BEH Amide,1.7 μm,2.1mm X100 mm, mobile phase a: water (containing 5mM ammonium formate, 5mM ammonium acetate and 0.2% formic acid), mobile phase B: acetonitrile (containing 1mM ammonium formate, 1mM ammonium acetate and 0.2% formic acid), flow rate 0.4mL/min, column temperature 35 ℃, gradient elution conditions as follows:

0-3min,95％B；3-11min,95％B-70％B；11-13min,70％B-55％B；13-14min,55％B。

the metabolites after chromatographic separation are injected into a mass spectrum for detection, and the specific mass spectrum conditions are as follows: the instrument comprises the following steps: triple quadrupole mass spectrometry, scan mode: multiple reaction monitoring (MRM/SRM), electrospray voltage: positive ion 3.5kv, negative ion 3kv, sheath gas: 35Arb, assist gas: 10Arb, ion source temperature: at 350 ℃.

(3) And performing partial least squares discriminant regression (PLS) on the obtained data and the immediate memory score, screening by VIP >1, and comparing and identifying with a reference substance to obtain the alternative metabolic marker.

(4) Taking the ion abundance ratio of the alternative metabolic markers and the internal standard of 241 samples and the corresponding immediate memory score (which is scored by a professional physician according to a RBANS scale), age and education age as input data, establishing a regression model capable of predicting the immediate memory score by using a Support Vector Machine (SVM), and resampling method: repeatedcv, number of iterations of resampling: 10 times; finally, a group of biomarker combinations 1 with high prediction accuracy and a regression model for predicting the immediate memory are obtained through machine learning screening, and are used for predicting or evaluating the immediate memory score of a schizophrenia patient, and the biomarker combinations specifically comprise the following compounds: creatinine, creatine, ornithine, threonine, valine, and glutamic acid.

TABLE 1

Note: the score data in the table are expressed as mean ± standard deviation, and the education years are counted from grade 1 of primary school.

In other embodiments, a person skilled in the art can directly detect the concentration of the biomarker panel obtained in the embodiments of the present invention by using multiple healthy individuals (more than 10 cases are recommended) as samples according to the biomarker panel obtained in the embodiments of the present invention, and then establish a regression model that can predict the immediate memory by combining the scoring results of the RBANS scale, such as the immediate memory; under the condition, the screening step of the marker is not needed, the workload is reduced, and the prediction efficiency is improved.

Example 2

Screening and modeling of biomarkers for predicting or assessing attention of healthy individuals.

Referring to the method of example 1, using the ages, educational years, ion abundance ratios of alternative metabolic markers to internal standards and corresponding attention scores (scored by a professional physician according to the RBANS scale) of the above 241 samples of healthy individuals as input data, a set of biomarker combinations 2 with higher prediction accuracy and regression models for predicting attention were obtained, which were used to predict or assess the attention scores of patients with schizophrenia, biomarker combinations 2 specifically comprising the following compounds: creatine, ornithine, valine, phenylalanine and glutamic acid.

Example 3

Screening and modeling of biomarkers for predicting or assessing the visual span of healthy individuals.

Referring to the method of example 1, using the ages, educational years, ion abundance ratios of alternative metabolic markers to internal standards and corresponding visual breadth scores (scored by a professional physician according to the RBANS scale) of the above 241 samples of healthy individuals as input data, a set of biomarker combinations 3 with higher prediction accuracy and regression models for predicting visual breadth were obtained, which were used to predict or assess the visual breadth scores of patients with schizophrenia, biomarker combinations 3 specifically including the following compounds: creatinine, creatine, isoleucine, lysine, citrulline, and glutamic acid.

Example 4

Screening and modeling of biomarkers for predicting or assessing language function in healthy individuals.

Referring to the method of example 1, using the ages, educational years, ion abundance ratios of alternative metabolic markers to internal standards and corresponding language function scores (scored by a professional physician according to the RBANS scale) of the above 241 samples of healthy individuals as input data, a set of biomarker combinations 4 with higher predictive accuracy and regression models for predicting language function were obtained for predicting or assessing the language function scores of patients with schizophrenia, biomarker combinations 4 specifically comprising the following compounds: hypoxanthine, creatinine, threonine, valine, and tryptophan.

Example 5

Screening and modeling of biomarkers for predicting or assessing delayed memory in healthy individuals.

Referring to the method of example 1, using the ages, educational years, ion abundance ratios of alternative metabolic markers to internal standards and corresponding delayed memory scores (scored by a professional physician according to the RBANS scale) of the above 241 samples of healthy individuals as input data, a set of biomarker combinations 5 with higher prediction accuracy and regression models for predicting delayed memory were obtained, which were used to predict or assess delayed memory scores of patients with schizophrenia, biomarker combinations 5 specifically include the following compounds: creatinine, creatine, ornithine, isoleucine and phenylalanine.

Example 6

Screening and modeling of biomarkers for predicting or assessing cognitive function in healthy individuals.

Referring to the method of example 1, using the ages, educational years, ion abundance ratios of alternative metabolic markers to internal standards and corresponding overall scores of cognitive function (i.e., the overall scores obtained by a professional according to the RBANS scale) of the above 241 samples of healthy individuals as input data, a set of biomarker combinations 6 with higher prediction accuracy and regression models for predicting cognitive function were obtained for predicting or assessing cognitive function scores of patients with schizophrenia, biomarker combinations 6 specifically including the following compounds: hypoxanthine, creatinine, ornithine, valine and arginine.

Experimental example 1

The evaluation effect of the prediction regression model established in example 1 on the cognitive function and each sub-dimension was evaluated.

Selecting blood randomly obtained from outpatients or inpatients of a certain hospital and blood of enrolled healthy individuals (111 samples of a test set of healthy individuals (information is shown in table 1), wherein all samples have complete scale data and are signed with informed consent), directly detecting ion abundance ratios of biomarkers to internal standards by referring to the methods of examples 1-6, inputting data (age, education degree and ion abundance ratio of a subject (ratio of biomarkers to internal standards of a corresponding biomarker group of the dimension to be predicted)) into the prediction regression models established in examples 1-6 to obtain prediction scores of different dimensions of cognitive function and evaluation of cognitive function, as represented by triangles in fig. 1-6, comparing the prediction scores with the scores of 30 samples diagnosed by a professional ANS through RB scale, see A in the figure; and counting the data, wherein the predicted value is within the range of +/-10 of the RBANS scale value, the predicted result is considered to be correct, and the statistical result is shown as B in the figure.

It can be seen that 78.38% accuracy was achieved in predicting the immediate memory dimension of healthy individuals using the predictive regression model of example 1 (figure 1).

It can be seen that 77.48% accuracy can be achieved in predicting the attention dimension of healthy individuals using the predictive regression model of example 2 (figure 2).

It can be seen that 83.78% accuracy can be achieved in predicting the visual breadth dimension of healthy individuals using the predictive regression model of example 3 (fig. 3).

It can be seen that 83.78% accuracy can be achieved in predicting language function dimensions of healthy individuals using the predictive regression model of example 4 (fig. 4).

It can be seen that 92.79% accuracy was achieved in predicting the delayed memory dimension of healthy individuals using the predictive regression model of example 5 (figure 5).

It can be seen that 90.09% accuracy was achieved in predicting cognitive function in healthy individuals using the predictive regression model of example 6 (fig. 6).

From the above results, it can be seen that the biomarker set provided by the embodiments of the present invention and the regression model established therefrom can predict the cognitive function of healthy individuals and evaluate the dimensional indexes of the cognitive function with high accuracy, including: immediate memory, attention, visual breadth, language function, delayed memory; a relatively objective technical means is provided for evaluating the cognitive function of a healthy individual, the fatigue problem of a test subject caused by RBANS evaluation can be effectively avoided, namely, a technician in the field only needs to inquire age information and education degree information of the test subject and collect blood samples, concentration information of biomarkers can be obtained through instrument detection, and the detected marker concentration result data is input into a pre-established regression model to obtain the cognitive function and the prediction scores of different dimensions of the test subject. Whether cognition is impaired based on the resulting predictive score, for example, when the predictive score for cognitive function is less than 70 points (corresponding to the total RBANS score), is considered cognitive impairment, which indicates that the individual is at a higher risk of senile dementia, schizophrenia, bipolar disorder, depression or other central nervous system disorder, and a further diagnosis is advised if the predictive score for cognitive function is greater than or equal to 70 points is considered normal.

The predicted cognitive function and scores of different dimensions predicted by the regression model are those skilled in the art, and the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and various modifications and variations may be made by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.