CN117594228A - Health star-level intelligent evaluation method, system and storage medium - Google Patents

Abstract

The invention belongs to the technical field of health management, and specifically relates to a health star intelligent evaluation method, system and storage medium. The method of the present invention includes the following steps: step 1, collect and input the indicators of the subject; step 2, use the HE algorithm model to calculate the indicators to obtain the health gap index of the subject. The HE algorithm model constructed by the present invention can eliminate the influence of subjective factors in health assessment, take into account the more comprehensive and objective results, and has good application prospects.

Description

A health star intelligent assessment method, system and storage medium

Technical field

The invention belongs to the technical field of health management, and specifically relates to a health star intelligent evaluation method, system and storage medium.

Background technique

Health assessment is the process of systematically collecting and analyzing a patient's health information to clarify their health status, existing health problems and their possible causes, and then assist doctors in obtaining follow-up care diagnosis results. Health assessment is the basis and guarantee for the implementation of overall nursing care, and is an essential ability for medical staff. The key is to comprehensively, systematically, accurately and dynamically collect, analyze and organize patients' health-related information to determine existing or potential problems to facilitate subsequent diagnosis.

Several currently popular health assessment methods include:

(1)BMI (Body Mass Index) is a standard commonly used in the world to measure the body's fatness and thinness and whether it is healthy. The calculation formula is: BMI = weight ÷ height ² . (Unit of weight: kilogram; unit of height: meter.).

(2) HRA health risk assessment: Using bioelectric induction technology, combined with human body electrical impedance measurement technology, and applying chronoamperostatistic analysis method, 3D reconstruction of human tissues and organs can be intuitively seen to change trends of body organs, and early diseases can be judged , thereby assessing human health status.

(3) There are relatively many mechanisms for questionnaire design and evaluation at this stage, such as the SF-36 scale, physical activity screening using PAR-Q, dietary survey review, and psychological methods such as SCL-90 for corresponding screening. For multidimensional assessment of client health.

According to the definition of WHO (World Health Organization), health is a state of good physical, mental and social well-being. The mere absence of disease in the body cannot be judged as healthy. Therefore, existing health assessment technologies mainly have the following two shortcomings:

(1) Entry into many special industries has very high requirements on personal physical condition and requires a comprehensive health status assessment method. The existing methods are single, sporadic and one-sided simple comparisons of measured indicators and standard indicators. This method only reflects problems in a certain aspect of health, and the evaluation results obtained are undoubtedly one-sided and inaccurate. However, because there are so many health indicators, there is a lack of methods to evaluate a person's overall health status.

(2) It is difficult to make an intuitive comparison of a single set of indicators in a high-dimensional space. Therefore, weights determined by humans and experts are generally used to summarize indicators to give a comprehensive score. However, this method is highly arbitrary and Uncertainty.

Based on the above problems, how to take into account the comprehensiveness and objectivity of the evaluation results in individual health assessment is an urgent problem that needs to be solved in this field.

Contents of the invention

Based on the problems of the existing technology, the present invention provides a health star intelligent assessment method, system and storage medium. The purpose is to provide a health assessment method that has comprehensive indicators and does not require artificial determination of indicators and their weights to achieve comprehensive and objective Personal health assessment.

A health star intelligent assessment method includes the following steps:

Step 1, collect and enter subjects’ indicators;

Step 2: Use the HE algorithm model to calculate the indicator to obtain the health gap index of the subject;

The calculation process of the HE algorithm model includes the following steps:

Step a: Perform inversion and normalization preprocessing on the negative indicators among the indicators;

Step b: Perform the following preprocessing on the interval indicators among the indicators:

Assume that when an indicator performs normally, the condition satisfies x∈[a,b], and the indicator is transformed as follows:

Among them, x′ _i ′ is the index after transformation, and x _i is the index before transformation;

Step c, assuming that each health indicator is calculated according to its optimal weight, the ideal value is calculated, and the calculation formula is:

Among them, maxV _j is the ideal value, y _ij is the i-th indicator of the j-th physical examination person, w _i is the weight of the i-th indicator, w _i+a is the weight of the i+a-th indicator, w _r is the weight of the r-th indicator, m is the number of indicators, n is the number of people undergoing physical examination, b _a is a customizable constant between 0 and 1, used when a certain weight needs to be customized , you can add a priori knowledge about the weight w _i -w _i+a ≥ b _a , st represents the constraint condition;

In step d, find the V(w) closest to the ideal value V ^* (w). The calculation formula is:

Among them, min is the minimum value operation, D ₂ is the 2-normal distance function, V (w) is the point closest to the ideal point V ^* (w), W is the set of all weights, and V _j ^* is the maxV obtained in step c _j .

Preferably, the negative indicator is inverted by taking its reciprocal and replacing the original indicator data with the reciprocal.

Preferably, the method for normalizing the negative indicators is:

Among them, x′ is the index after transformation, x is the index before transformation, μ is the average value of the indicator, x _max is the maximum value of the indicator, and x _min is the minimum value of the indicator.

Preferably, when the weight of the indicator needs to be manually adjusted, a priori knowledge about the weight w _i -w _i+a ≥ b _a is added to the constraint conditions.

Preferably, the types of indicators include at least one of the following indicators: indicators used to evaluate genes, physiology, psychology, exercise, nutrition, and environment.

Preferably, the indicators used to evaluate physiology include at least one of the following indicators: hemoglobin, red blood cells, white blood cells, platelet count, fasting blood glucose, serum total cholesterol, serum triacylglycerol, blood uric acid, troponin I, creatinine , alanine aminotransferase, serum total bilirubin, serum conjugated bilirubin, serum unconjugated bilirubin, carcinoembryonic antigen, alpha-fetoprotein, rheumatoid factor, reactive protein, procalcitonin;

The indicators used to evaluate psychology include at least one of the following indicators: intelligence scale, other rating scale, anxiety self-rating scale, and depression self-rating scale;

The indicators used to evaluate nutrition include at least one of the following indicators: sodium, potassium, magnesium, calcium, phosphorus, zinc, iodine, selenium, and manganese;

The indicators used to evaluate exercise include at least one of the following indicators: body mass index, pulse, total blood testosterone in the morning at rest, measurement, plasma cortisol, CD4/CD8, IgA, IgM, IgG, blood ammonia, urine proportion;

The indicators used to evaluate the environment include at least one of the following indicators: pesticide pollution, air pollution, and soil pollution.

Preferably, the health disparity index is used for classification according to the following thresholds:

Five-star rating: indicator gap index ≥90;

Four stars: 90>Indicator gap index ≥75;

Three stars: 75>Indicator gap index ≥60;

Two stars: 60> indicator gap index ≥ 45;

One star: 45>Indicator Gap Index.

The invention also provides a health star intelligent evaluation system, including:

Input module, used to input subjects’ indicators;

The calculation module is used to calculate according to the above-mentioned health star intelligent evaluation method;

The output module is used to output the calculation results of the calculation module.

The present invention also provides a computer-readable storage medium on which is stored a computer program for implementing the above health assessment method.

In the present invention, "negative index" is an index that indicates that the smaller the index value is, the better the health status of the subject is; "interval index" means that the measurement standard of the index is that when an index meets a certain range, the performance as normal or abnormal.

It should be noted that the final result obtained by the health assessment method of the present invention is an objective health assessment score (health gap index), which can only be used as an auxiliary parameter for reference by users and doctors, and cannot be used independently and directly. As a basis for diagnosis of a specific disease.

The present invention provides a HE algorithm for personal health assessment. This algorithm divides the calculation of the comprehensive evaluation model into two stages: the first stage uses the data envelopment analysis method to obtain the optimal score of each person being evaluated for physical examination; the second stage In the second stage, the optimal score obtained in the first stage is used as the ideal point, and regression analysis is performed in the feasible region of the weight to obtain the same weight that maximizes the interests of all evaluated medical examinees, thus solving the problems existing in the above-mentioned data envelopment analysis model. . Through the above calculation, multiple indicators can be comprehensively scored without introducing the steps of manually determining weights, avoiding the influence of human subjective factors and obtaining a more comprehensive and objective health assessment score. Therefore, the present invention has good application prospects.

Obviously, according to the above content of the present invention, according to the common technical knowledge and common means in the field, without departing from the above basic technical idea of the present invention, various other forms of modifications, replacements or changes can also be made.

The above contents of the present invention will be further described in detail below through specific implementation methods in the form of examples. However, this should not be understood to mean that the scope of the above subject matter of the present invention is limited to the following examples. All technologies implemented based on the above contents of the present invention belong to the scope of the present invention.

Description of drawings

Figure 1 is a schematic flow chart of Embodiment 1 of the present invention.

Detailed ways

It should be noted that the algorithms for data collection, transmission, storage, and processing steps that are not specified in the embodiments, as well as the hardware structures, circuit connections, etc. that are not specified in the embodiments can all be implemented by what has been disclosed in the prior art.

Embodiment 1 Health Assessment Method and System

The system provided by this embodiment includes:

Input module, used to input subjects’ indicators;

Calculation module, used to calculate subject health assessment results;

The output module is used to output the calculation results of the calculation module.

Based on the above system, the method flow of subject health assessment in this embodiment is shown in Figure 1, specifically:

Relevant indicators of subjects were collected, including six major aspects: genes, physiology, psychology, exercise, nutrition, and environment, including 1,525 indicators and 5,946 indicators of multiple single-gene genetic diseases. It should be noted that the indicator collection here aims to be comprehensive and systematic.

The specific subject health assessment index system is shown in the following table:

Table 1 Health Assessment Index System

First level indicator Secondary indicators Level three indicators physiological 4 81 891 Gene 4 - 5946 psychology 5 50 309 Nutrition 2 9 95 sports 4 15 85 environment 3 twenty three 144 total twenty two 178 7470

In order to scientifically evaluate the impact and role of indicators on health conditions in dynamic changes, the HE model is used. Compared with the traditional method of artificially determining weights, the advantage of this method is that the weights are calculated through the model and a score is given, which is a completely data-driven method. The specific steps for this algorithm model to obtain the subject's health disparity index include:

Step one: data preprocessing

① First, take the reciprocal of the negative indicator (that is, the smaller the better indicator) and replace the original indicator data. The same inverse operation is performed on various indicators of the environment. Finally, the average normalization method is used to eliminate the dimensions. This method is to re-adjust the data range and scale it to the range of [0,1]. The general formula is as follows:

Among them, x′ is the index after transformation, x is the index before transformation, μ is the average value of the indicator, x _max is the maximum value of the indicator, and x _min is the minimum value of the indicator.

Finally, the processed data is obtained to facilitate input into the algorithm model for calculation.

② In addition, due to the special nature of physical examination indicator data, most of them are interval indicators, that is, the measurement standard of indicators is usually when an indicator meets a certain range, it is normal or abnormal. Therefore, the following special preprocessing is performed for the physical examination index data.

Assume that when an indicator performs normally, the condition satisfies x∈[a,b], and the indicator is transformed as follows:

Among them, xi _′ is the index after transformation, and _xi is the index before transformation;

Step 2: Calculate the ideal value maxV _j assuming each indicator is calculated according to its optimal weight.

Among them, maxV _j is the ideal value, y _ij is the i-th indicator of the j-th physical examination person, w _i is the weight of the i-th indicator, w _i+a is the weight of the i+a-th indicator, w _r is the weight of the r-th indicator, m is the number of indicators, n is the number of people undergoing physical examination, b _a is a customizable constant between 0 and 1, used when a certain weight needs to be customized , you can add a priori knowledge about the weight w _i -w _i+a ≥ b _a , st represents the constraint condition;

third step:

The second step causes each indicator to be calculated according to its own optimal weight, then is the ideal value of the jth evaluation unit, then the vector/> is the ideal point. Since this situation is difficult to achieve in practical applications, our goal is to find the V(w) closest to the ideal point V ^* (w). To do this, we need a distance function that measures the distance between them. When this distance is minimum, the V(w) closest to the ideal value can be obtained. Therefore, we can calculate the weight w=(w ₁ ,...,w _m ) by minimizing the distance function.

s.t.w∈W

Among them, min is the minimum value operation, D ₂ is the 2-normal distance function, V (w) is the point closest to the ideal point V ^* (w), and W is the set of all weights. is the maxV _j obtained in step c. Other variable definitions are the same as in the previous steps.

To use the HE model, the data must meet the following principles:

①The data must be positive numbers;

②The output data must be as large as possible;

③Indicators must be representative and important.

After calculating the health disparity index (1-100 points), it is graded through the following threshold division:

Five stars: indicator gap index ≥90, all indicators are normal among peers;

Four stars: 90>Indicator gap index ≥75, some indicators are abnormal, but it does not affect normal life and work;

Three stars: 75> Indicator gap index ≥ 60, some indicators are abnormal, affecting work;

Two stars: 60> Indicator gap index ≥ 45, some indicators are abnormal, medication is required, affecting work and life;

One star: 45> indicator gap index, requiring hospitalization from time to time in the hospital.

As an example, this embodiment selected 45 common indicators for experiments, including 19 physiological indicators, 0 genetic indicators, 4 psychological indicators, 9 nutritional indicators, 10 exercise indicators, and 3 environmental indicators. The physical examination indicator information of 50 individuals who underwent physical examination was collected, including 25 males and 25 females. The data format is as shown in the following table:

Table 2 Experimental data format

It should be noted that in the above table, the quantification method of pesticide pollution, air pollution and soil pollution is: if the subject is affected by this pollution, the value is 1, and if the subject is not affected by this pollution, the value is 0.

By inputting the HE algorithm model according to this data format, the weight of each indicator can be calculated directly through data-driven calculation. The process of determining the weight does not require human participation. The resulting weight is as shown in the following table:

Table 3 Weight of each indicator

Using the weights derived from the algorithm, the final score and health grade of each physical examination person are calculated. The final results are as shown in the following table:

Table 4 Experimental results

Physical examination person Score star rating Physical examination person 2 0.999999979 five stars Physical examination person 4 0.987332632 five stars Physical examination person 8 0.985760297 five stars Physical examination person 9 0.983248777 five stars Physical examination person 21 0.978555832 five stars Physical examination person 5 0.977107164 five stars Physical examination person 24 0.973747018 five stars Physical examination person 13 0.973346324 five stars Physical examination person 11 0.973149123 five stars Physical examination person 7 0.961049773 five stars Physical examination person 6 0.957013539 five stars Physical examination person 15 0.943922458 five stars Physical examination person 22 0.930948597 five stars Physical examination person 18 0.92647623 five stars Physical examination person 17 0.921797046 five stars Physical examination person 20 0.916345704 five stars Physical examination person 19 0.916129007 five stars Physical examination person 10 0.908976003 five stars Physical examination person 1 0.90743439 five stars Physical examination person 23 0.882876812 four stars Physical examination person 12 0.875160523 four stars Physical examination person 14 0.86806093 four stars Physical examination person 16 0.852761555 four stars Physical examination person 3 0.846589336 four stars Physical examination person 25 0.749603358 Two stars

Compared with the traditional method of artificially determining weights, the HE model proposed in this embodiment has the advantage that the weights are calculated through the model and scores are given, which is a completely data-driven method.

It can be seen from the above embodiments that the present invention constructs a HE algorithm model, which can eliminate the influence of subjective factors in health assessment, take into account the more comprehensive and objective results, and has good application prospects.

Claims (9)

1. The intelligent evaluation method for the health star level is characterized by comprising the following steps of:

step 1, collecting and inputting indexes of a subject;

step 2, calculating the index by adopting an HE algorithm model to obtain a health gap index of the subject;

the calculation process of the HE algorithm model comprises the following steps:

step a, carrying out pretreatment of negation and normalization on negative indexes in the indexes;

step b, preprocessing section indexes in the indexes as follows:

assuming that when a certain index shows normal conditions are satisfied with x epsilon a, b, the following transformation is performed on the index:

wherein x' _i As the index after transformation, x _i Is an index before transformation;

step c, assuming that each health index is calculated according to the optimal weight, calculating an ideal value, wherein the calculation formula is as follows:

wherein maxV _j For the ideal value, y _ij Is the ith index of the jth individual inspector, w _i Is the weight of the ith index, w _i+a Is the weight of the (i+a) th index, w _r Is the weight of the r index, m is the number of indexes, n is the number of evaluated people, b _a Is a constant between 0 and 1 which can be customized, and is used for adding prior knowledge w about weight when a certain weight needs to be customized _i -w _i+a ≥b _a S.t. represents a constraint;

step d, finding a value V from the ideal value ^* (w) V (w) nearest to the nearest V (w), the calculation formula is:

wherein min is minimum value operation, D ₂ As a distance function of 2-range, V (w) is the distance from the ideal point V ^* (W) the nearest point, W is the set of ownership weights, V _j ^* Is maxV obtained in step c _j 。

2. The healthy star-level intelligent assessment method according to claim 1, wherein: the negative index is inverted by taking the reciprocal and replacing the original index data with the reciprocal.

3. The healthy star-level intelligent assessment method according to claim 1, wherein: the normalization method of the index comprises the following steps:

wherein x' is the index after transformation, x is the index before transformation, mu is the average value of the index, x _max Is the maximum value of the index, x _min Is the minimum value of the index.

4. According to the weightThe healthy star-level intelligent assessment method as claimed in claim 1, characterized in that: when the weight of the index needs to be manually adjusted, the prior knowledge w about the weight is added into the constraint condition _i -w _i+a ≥b _a 。

5. The healthy star-grade intelligent assessment method according to claim 1, wherein the category of the index includes at least one of the following indexes: is used for evaluating indexes of genes, physiology, psychology, sports, nutrition and environment.

6. The healthy star-grade intelligent assessment method according to claim 5, wherein the index for assessing physiology comprises at least one of the following: hemoglobin, red blood cells, white blood cells, platelet count, fasting blood glucose, serum total cholesterol, serum triacylglycerol, uric acid, troponin I, creatinine, alanine aminotransferase, serum total bilirubin, serum-bound bilirubin, serum-unbound bilirubin, carcinoembryonic antigen, alpha fetoprotein, rheumatoid factor, reactive protein, procalcitonin;

the index for evaluating psychology includes at least one of the following indexes: intellectual, he, anxiety, depression, etc. self-rating;

the index for evaluating nutrition includes at least one of the following indexes: sodium, potassium, magnesium, calcium, phosphorus, zinc, iodine, selenium, manganese;

the index for evaluating exercise includes at least one of the following indexes: body mass index, pulse, total testosterone in early morning resting state, measurement, plasma cortisol, CD4/CD8, igA, igM, igG, blood ammonia, urine specific gravity;

the index for evaluating the environment includes at least one of the following indexes: pesticide pollution, atmospheric pollution and soil pollution.

7. The healthy star rating intelligent assessment method according to claim 1, wherein the healthy gap index is used for rating according to the following thresholds:

five star grade: the index gap index is more than or equal to 90;

four star stages: the index gap index of 90 is more than or equal to 75;

three star stages: 75> index gap index is more than or equal to 60;

two star stages: the index gap index of 60 is more than or equal to 45;

a star grade: 45> index gap index.

8. A health star-level intelligent assessment system, comprising:

the input module is used for inputting the index of the subject;

a calculation module for performing calculations according to the healthy star grade intelligent assessment method of any one of claims 1-7;

and the output module is used for outputting the calculation result of the calculation module.

9. A computer-readable storage medium having stored thereon: computer program for implementing a healthy star grade intelligent assessment as defined in any one of claims 1-7.