WO2021097731A1

WO2021097731A1 - Human mental stress test method and system

Info

Publication number: WO2021097731A1
Application number: PCT/CN2019/119800
Authority: WO
Inventors: 林千寻
Original assignee: 林千寻
Priority date: 2019-11-20
Filing date: 2019-11-20
Publication date: 2021-05-27
Also published as: CN113194829B; CN113194829A

Abstract

A human mental stress test method and system. The method comprises: providing at least one detection sensor on a certain movable catharsis object to detect the behavior of a sample population towards the catharsis object, so as to obtain a sample test data set (step 1); presetting a mental stress assessment algorithm, wherein the mental stress assessment algorithm has at least one assessment parameter, carrying out a mental stress assessment on samples in the sample test data set by means of the mental stress assessment algorithm, and training and adjusting the assessment parameter, such that an assessment error of the mental stress assessment algorithm having a finally determined assessment parameter statistically satisfies an error range that is medically recognized (step 2); and stopping training and adjusting the assessment parameter, and testing and assessing a test sample by means of the finally determined assessment parameter, so as to obtain a test stress score that can be medically recognized (step 3). The human mental stress test method and system solve the problems in the prior art of only very few professionals being able to carry out a mental stress assessment on people, the cost being high and the application area being small.

Description

Human mental stress testing method and system

Technical field

The invention relates to the field of grassroots public health, in particular to a method and system for testing human mental stress.

Background technique

The pace of society is getting faster and faster, and people’s pressure on work and study is also increasing. However, with the exception of psychological counseling and psychologists, there is no special method and equipment for testing mental stress, and it is impossible to pay attention to the pressure of the crowd at all times, which leads to social problems and serious family problems. There is no real-time monitoring and intervention as soon as possible. Intervention solutions.

Summary of the invention

The purpose of the present invention is to provide a method and system for testing human mental stress, which aims to solve the problems in the prior art that only professional and very few personnel can make mental stress assessments for people, and the cost is high and the application range is small.

The present invention provides a method for testing human mental stress, including:

Step 1: Set up at least one detection sensor on a certain emotional catharsis object to detect the behavior of the sample crowd towards the emotional catharsis object to obtain a sample detection data set;

Step 2: Preset a mental stress evaluation algorithm, the mental stress evaluation algorithm has at least one evaluation parameter, the mental stress evaluation algorithm is used to evaluate the mental stress of the samples in the sample detection data set, and the evaluation parameters are trained and adjusted so that The evaluation error of the mental stress evaluation algorithm with the final determined evaluation parameters meets the medically recognized error range statistically;

Step 3: Stop the evaluation parameter training and adjustment, and use the final evaluation parameters to test and evaluate the test samples, so as to obtain a test pressure score that can be recognized by medicine.

In the present invention, the human body behavior can be monitored by monitoring the emotional catharsis object first, such as tapping, shouting, kneading and other behaviors. These behaviors are recorded by the detection sensor. For example, the detection sensor is a strength sensor. The force record is detected in the part, and it is converted into test data, of course, including other behaviors. A test data is formed for a human sample, and the mental stress of the sample is evaluated by the preset mental stress evaluation algorithm, and the mental stress of the sample is evaluated after the evaluation. It is necessary to train and adjust the evaluation parameters, recalculate multiple times to determine the evaluation parameters that can be medically recognized, so that no human operation is required, and then use the same determined evaluation parameters to make a test pressure score that can be medically recognized on the sample population. So far, once the final evaluation parameter is confirmed, the result made by using the evaluation parameter is relatively accurate, and a very small number of professionals are no longer needed to evaluate the mental stress, which solves the problems in the prior art.

Using this method to test the human mental stress, the corresponding emotional catharsis objects can be set in the primary health center, so as to replace the doctor to make the test and evaluation.

Of course, the catharsis objects can even be made into dolls, and robots and other images can be placed in homes, offices, and entertainment venues. They can monitor human behavior anytime and anywhere to complete the detection and evaluation of mental stress; compared to the traditional ones that can only go to medical institutions In the short-term form evaluation at that time, the doctor in charge judged that this method can be monitored in real time and long-term, and can form multiple data based on multiple monitoring data, which is convenient for evaluation, analysis and statistics.

The present invention also provides a human body stress testing system. The system includes: a movable emotional catharsis object; a plurality of detection sensors arranged on the emotional catharsis object, and the detection sensor is used to detect the human body's emotional catharsis object The behavior of sending detection data; a data computing center with preset mental stress evaluation algorithms, the calculation algorithm calculates the detection data through at least one preset evaluation parameter, evaluates the human mental stress, and continuously trains and adjusts the Evaluation parameters, determine the final evaluation parameters, so that the evaluation error of the mental stress evaluation algorithm with the final evaluation parameters meets the medically recognized error range statistically; the data computing center uses the spirit of the final evaluation parameters The stress evaluation algorithm evaluates the human mental stress and obtains the test stress score.

Description of the drawings

Figure 1 is a flowchart of the human mental stress testing method of the present invention;

Figure 2 is a process diagram of the current training adjustment method of the present invention;

Figure 3 is a process diagram of the statistical training adjustment method of the present invention;

Fig. 4 is a schematic diagram of modules of the human mental stress testing system of the present invention.

Detailed ways

The present invention will be further elaborated and illustrated below in conjunction with specific embodiments and the accompanying drawings of the specification:

Referring to Fig. 1, the present invention provides a method for testing human mental stress. The method includes:

Step 1: Set up at least one detection sensor on a movable emotional catharsis object to detect the behavior of the sample crowd towards the emotional catharsis object to obtain a sample detection data set.

Refer to Figure 2. For example, the sample data set in Figure 2 includes multiple sample data α ₁ , α ₂ , α ₃ …, each test sample, that is, each subject of mental stress that needs to be tested, will generate a sample data α _N ; and each sample data includes multiple sample elements, for example, the sample data α ₂ includes sample elements as (X ₂ , Y ₂ …Z ₂ ), and each sample element comes from a detection sensor, and these detection sensor pairs The detection content of the catharsis object includes data such as position, strength and frequency. Specifically, the catharsis object can be divided into different areas, and the sensors are arranged in different areas according to the position sensor, the force sensor, the counting sensor and the combination thereof, and the detection The sensor monitors the venting position, the venting force, and the venting frequency to obtain the sample element, and the sample elements of a plurality of detection sensors constitute the sample data.

In an actual application, the catharsis object is a plastic human head structure. The eyes and nose area of the human head structure are divided into the first area, and the face and mouth are divided into the second area. The hair-covered part and the neck area are divided into the third area; then position sensors and force sensors are set up in these areas; when the detected person slaps the cathartic object, after the twisting action, the sensors in these areas record the slap respectively. The position and the power of the slap; at the same time corresponding to the formation of sample data.

Step 2: Preset the mental stress evaluation algorithm, the mental stress evaluation algorithm has evaluation parameters, the mental stress evaluation algorithm is used to evaluate the mental stress of the samples in the sample detection data set, and the evaluation parameters are trained and adjusted to have the final The evaluation error of the determined evaluation parameter of the mental stress evaluation algorithm meets the medically recognized error range statistically.

Specifically, the mental stress evaluation algorithm can have multiple evaluation parameters, and the number of evaluation parameters is set according to the detection data. For example, in the above application, there should be two evaluation parameters corresponding to the position and strength of the detection data slap, such as : The location data is that the first area, the second area and the third area respectively correspond to an evaluation parameter A ₁ , that is, A ₁ ={0.5, 0.3, 0.2}. In some applications, the evaluation parameter can be a weight value calculated by mathematical statistics. In some other embodiments, the evaluation parameter may also be a Boolean value, a logical operation, and so on. Through the pre-set evaluation parameter A ₁ , the magnitude of the mental pressure shown by the different areas of the human head structure during the action of being slapped is given. From the evaluation parameter A _{1 of} this application, namely A ₁ ={0.5,0.3,0.2}, it can be seen that the weight in the first area is greater than that in the third area. That is to say, when we calculate mental stress, we pre-set the eye and The mental stress of the nose area being slapped is greater than that of the hair-covered area and the neck area.

The mental stress evaluation algorithm includes mathematical calculation formulas. Such mathematical calculation formulas are not the same in different application scenarios and calculations. When calculating, the preset evaluation parameters are added to the calculation formula, and then the detection data is entered. Get the estimated pressure score. However, in the process of using the calculation formula for the first time, because the evaluation parameters and mathematical calculation formula are affected by subjective experience and evaluation, the calculated estimated pressure score will not be accurate, or the correlation will be poor. .

In practical applications, it is necessary to train and adjust the evaluation parameters, and determine the final evaluation parameters. The final evaluation parameters mean that the test pressure scores evaluated by the evaluation parameters are already very close to the actual evaluation scores, which are statistically similar. , Recognized by medicine, can be used as the final evaluation score.

Specifically, the training and adjustment of the evaluation parameters include two methods that are implemented in sequence, that is, the current training adjustment method and the statistical training adjustment method.

Wherein, the current training adjustment method refers to performing multiple calculations on a certain sample data and adjusting the evaluation parameters, so that the calculation results obtained by calculating the current sample data according to the currently determined evaluation parameters can meet the error range of medical requirements. . Simply put, it is to adjust the evaluation parameters many times, and when the value is adjusted to a certain value, the calculated estimated pressure score is considered to be relatively accurate in medicine.

Specifically, the current training adjustment method includes:

Step 211: Use a mental stress evaluation algorithm to calculate a certain sample data in the sample detection data set to obtain an estimated stress score.

Step 212: Perform artificial mental stress detection on a sample corresponding to the sample data to obtain an actual stress score.

Step 213: Evaluate the error level between the estimated pressure score and the actual pressure score.

As shown in Figure 2, the error levels are divided into 1, 2, 3, and 4 levels. Of course, in other applications, the error levels can be more. More error levels mean higher error requirements, higher computational complexity, and longer periods of training and adjustment.

In this embodiment, when the error level is not greater than 2, it can be considered that the error is medically acceptable and within the error range.

Step 214: Adjust the evaluation parameter according to the preset correspondence table, and the mapping relationship between the error level and the adjustment value of the evaluation parameter is pre-stored in the correspondence table.

Step 215: After each adjustment of the evaluation parameters, the mental stress evaluation algorithm is used to calculate the sample data again, and the calculated estimated pressure score and the actual pressure score are evaluated for the error level until the error level meets the preset level So far.

In this step, as shown in Figure 2, in the first calculation and evaluation, the estimated pressure score obtained by the evaluation is 89 points, and the actual pressure score is 45, with an error of 44 points. When evaluating the error level, it is considered It is at the error level 4 and greater than the preset error level 2; you need to adjust the evaluation parameters according to the evaluation adjustment parameter value {0.10,0.05} corresponding to the error level 4; this is the first adjustment.

After that, after the first adjustment, a second evaluation is carried out. The adjusted evaluation parameters are used to calculate and evaluate the same test data again. The evaluation pressure score obtained by the evaluation is 65 points, and the actual pressure score is 45. The error is 20 points. When evaluating the error level, it is considered to be at the error level 3, which is greater than the preset error level 2. The evaluation parameter needs to be adjusted according to the evaluation adjustment parameter value {0.05,0.05} corresponding to the error level 3; This is the second adjustment.

After the second adjustment, the third evaluation is carried out, and the same test data is calculated and evaluated again using the two adjusted evaluation parameters. The evaluation pressure score obtained by the evaluation is 40 points, while the actual pressure score is 45. The error is 5 points. When evaluating the error level, it is considered to be at the error level 2 and not greater than the preset error level 2. It is no longer necessary to adjust the parameter value {0.01,0.02} according to the evaluation corresponding to the error level 2 Adjust the evaluation parameters.

After the third evaluation, the evaluation parameters were determined, and the determined evaluation parameters met the current testing data evaluation requirements.

Wherein, the statistical training adjustment method refers to calculating multiple sample data using currently determined evaluation parameters, and judging whether the multiple sample calculation results statistically meet the error range of medical requirements, and if it meets the requirements, the current training is stopped. The adjustment method, training and adjustment of the evaluation parameters are completed.

Referring to Figure 3, the statistical training method includes:

Step 221: After determining the evaluation parameter of a certain sample data calculation, use the evaluation parameter to recalculate the calculated sample data, or use the evaluation parameter to randomly select multiple sample data from the sample detection data set to calculate again, and obtain multiple Estimated pressure score.

For example, in Figure 2, the current training adjustment method is used to test the same sample data three times, and the final evaluation parameter is {A ₁ ,A ₂ }, then in this method, the evaluation parameter {A ₁ ,A ₂ }Recalculate the calculated sample data or randomly selected sample data to obtain multiple evaluation pressure scores.

Step 222: As shown in Figure 3, compare the obtained multiple estimated pressure scores with the actual pressure scores corresponding to these samples, and count the number of different error levels between them.

Step 223: preset a statistical threshold percentage. In this embodiment, the statistical threshold percentage for the sample data that has been calculated is set to 92%, that is, it is considered that only in the sample data that has been calculated, the calculated evaluation is evaluated according to the evaluation parameters {A ₁ , A ₂ } If the difference between the pressure score and the actual pressure score is not greater than the proportion of the level 2 is greater than 92%, it will be considered that the _{mental stress evaluation algorithm with evaluation parameters {A 1} ,A ₂ } is medically recognized and has statistics Meaning.

Step 224: Calculate the proportion of the number of samples meeting the medically recognized error level to the number of samples recalculated in step 222, and compare it with the statistical threshold percentage.

When the proportion is less than the statistical threshold percentage,

As shown in Figure 3, in this embodiment, for the re-evaluation of the sample data that has been calculated, the error level statistics account for 95%, then the training and adjustment of the evaluation parameters are completed, and the next time the sample data is entered into the mental stress evaluation algorithm The current sample training adjustment method is no longer executed; after entering the test data again, the _{mental stress is evaluated directly according to the evaluation parameters {A 1} _{, A 2} } and the mental stress evaluation algorithm, and the evaluation pressure score obtained after the evaluation is in line with the medical Statistically significant, considered valid.

As shown in Fig. 3, in this embodiment, for the re-evaluation of randomly sampled sample data, the error level statistics account for 83%. If 83% is less than 90% of the preset statistical threshold percentage setting, the training and adjustment of the evaluation parameters are combined. Not completed; the current sample training adjustment method is executed again when the sample data is entered into the mental stress evaluation algorithm next time, and the evaluation parameters are continued to be trained and adjusted until the error level statistics accounted for not less than the statistical threshold percentage.

In this embodiment, the error level gradually becomes smaller, and the adjustment value of the evaluation parameter is also reduced accordingly. The gradually reduced error level adjustment value can make the amplitude of each adjustment gradually decrease, gradually approaching the actual evaluation result, and there will be no large fluctuations.

In this embodiment, the method is also provided with a matching intervention table, in which a plurality of intervention measures corresponding to the evaluation pressure score are set, and the evaluation pressure score is calculated according to the evaluation parameters completed by training and adjustment Obtain the corresponding intervention measures and implement the intervention. For example: for the evaluation of the pressure score, according to the 0-100 points scoring standard, 0-60 points are evaluated as serious and need to be involved; 61-75 points are evaluated as risks and need to be followed; 76-85 points are evaluated as general , No treatment; 86-100 points are evaluated as healthy. After the evaluation, the scores are used to directly send corresponding warnings or information, and the corresponding system can realize the independent evaluation and tracking of mental stress and intervention, which can greatly reduce social risks and life stress.

Referring to FIG. 4, the present invention also provides a human mental stress testing system. The system includes: a movable emotional catharsis object 10; a plurality of detection sensors 11 arranged on the emotional catharsis object 10, the detection sensor 11 is used to detect the human body’s behavior towards the emotional catharsis object 10 and send detection data; a data calculation center 20 with a pre-set mental stress evaluation algorithm is built in, and the calculation algorithm calculates and evaluates the detection data through at least one preset evaluation parameter. Human mental stress, and continuously train and adjust the evaluation parameters to determine the final evaluation parameters, so that the evaluation error of the mental pressure evaluation algorithm with the final evaluation parameters meets the medically recognized error range statistically; The data computing center 20 uses a mental stress evaluation algorithm with final evaluation parameters to evaluate the human mental stress to obtain a test stress score.

Further, the system includes: the data calculation center 20 that calculates the detection data and obtains the estimated pressure score 30; the actual pressure score 40 obtained after passing the artificial mental stress test; The error evaluation unit for evaluating the actual pressure score 40 and the estimated pressure score 30 error levels; according to the error level, the parameter feedback modification unit for adjusting the evaluation parameters according to the corresponding relationship table between the preset error level and the evaluation parameter adjustment value 50; The data calculation center 20 re-evaluates the human mental stress according to the evaluation parameters adjusted by the parameter feedback modification unit 50, and continuously evaluates the human body pressure after multiple adjustments, so as to achieve the actual pressure score of 40 and the estimated pressure score The value of 30 gradually reduces the error until it is within the medically recognized controllable range.

Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the scope of protection of the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand The technical solution of the present invention can be modified or equivalently replaced without departing from the essence and scope of the technical solution of the present invention.

Claims

A method for testing human mental stress, which is characterized in that it comprises the following steps:

Step 1: Set up at least one detection sensor on a movable emotional catharsis object to detect the behavior of the sample crowd towards the emotional catharsis object to obtain a sample detection data set;

Step 2: Preset a mental stress evaluation algorithm, the mental stress evaluation algorithm has at least one evaluation parameter, the mental stress evaluation algorithm is used to evaluate the mental stress of the samples in the sample detection data set, and the evaluation parameters are trained and adjusted so that The evaluation error of the mental stress evaluation algorithm with the final determined evaluation parameters meets the medically recognized error range statistically;

Step 3: Stop the evaluation parameter training and adjustment, and use the final evaluation parameters to test and evaluate the test samples, so as to obtain a test pressure score that can be recognized by medicine.
The human mental stress test method according to claim 1, wherein the sample data set includes a plurality of sample data, each of the test samples generates one sample data, and each of the sample data includes a plurality of sample elements , Each of the detection sensors obtains one of the sample elements after detection.
The human mental stress test method according to claim 2, wherein the emotional catharsis object is divided into a plurality of different areas, and the detection sensor includes a position sensor, a force sensor, a counting sensor and a combination thereof; a pair of detection sensors The vent position, the sample element is obtained for the vent force and the vent frequency, and the sample elements of a plurality of detection sensors constitute the sample data.
The human mental stress test method according to claim 1, wherein the method of training and adjusting the evaluation parameters in the step 2 includes a current training adjustment method and a statistical training adjustment method implemented in sequence; the current training adjustment The method refers to performing multiple calculations on a certain sample data and adjusting the evaluation parameters, so that the calculation results obtained by calculating the sample data according to the currently determined evaluation parameters meet the error range of the medical requirements; the statistical training adjustment method is Refers to using the currently determined evaluation parameters to calculate multiple sample data, and to determine whether the multiple sample calculation results statistically meet the error range of medical requirements, if it is satisfied, stop implementing the current training adjustment method, train and adjust the evaluation The parameters are complete.
The method for testing human mental stress according to claim 3, wherein the current training adjustment method comprises:

Step 211: Use a mental stress evaluation algorithm to calculate a certain sample data in the sample detection data set to obtain an estimated stress score;

Step 212: Perform artificial mental stress detection on a sample corresponding to the sample data to obtain an actual stress score;

Step 213: Evaluate the error level between the estimated pressure score and the actual pressure score;

Step 214: Adjust the evaluation parameters according to the preset correspondence table, and the mapping relationship between the error level and the adjustment value of the evaluation parameter is pre-stored in the correspondence table;

Step 215: After each adjustment of the evaluation parameters, the mental stress evaluation algorithm is used to calculate the sample data again, and the calculated estimated pressure score and the actual pressure score are evaluated for the error level until the error level meets the preset level So far.
The method for testing human mental stress according to claim 3 or 4, wherein the statistical training method comprises:

Step 221: After determining the evaluation parameter of a certain sample data calculation, use the evaluation parameter to recalculate the calculated sample data, or use the evaluation parameter to randomly select multiple sample data from the sample detection data set to calculate again, and obtain multiple Estimated pressure score;

Step 222: Compare the obtained multiple estimated pressure scores with the actual pressure scores corresponding to these samples, and count the number of different error levels between them;

Step 223: preset a percentage of the statistical threshold;

Step 224: Calculate the proportion of the number of samples meeting the medically recognized error level to the number of samples recalculated in step 222, and compare it with the threshold percentage;

When the proportion is less than the threshold percentage, the current sample training and adjustment method will continue to be executed the next time the sample data is entered into the mental stress evaluation algorithm; otherwise, the training and adjustment of the evaluation parameters are completed, and the next time the sample data is entered into the mental stress evaluation algorithm The current sample training adjustment method is no longer executed.
7. The human mental stress testing method according to claim 6, wherein the error level gradually becomes smaller, and the adjustment value of the evaluation parameter is also reduced correspondingly.
The human mental stress test method according to any one of claims 1 to 5, wherein a matching intervention table is set, and a plurality of intervention measures corresponding to the evaluation stress score are set in the matching intervention table, and according to training The evaluation pressure score calculated by adjusting the evaluation parameters to obtain the corresponding intervention measures and implement the intervention.
A human mental stress testing system, characterized in that it is used to implement the human mental stress testing method of claim 1 to claim 8, and the system comprises:

An emotional catharsis object, the emotional catharsis object can be moved;

A plurality of detection sensors arranged on the emotional catharsis object, the detection sensors are used to detect the human body's behavior on the emotional catharsis object and send detection data;

A data calculation center with a preset mental stress evaluation algorithm is built in. The calculation algorithm calculates the detection data through at least one preset evaluation parameter, evaluates the human mental pressure, and continuously trains and adjusts the evaluation parameters to determine the final evaluation Parameters, so that the evaluation error of the mental stress evaluation algorithm with the final evaluation parameters meets the medically recognized error range statistically;

The data computing center uses a mental stress evaluation algorithm with final evaluation parameters to evaluate the human mental stress to obtain a test stress score.
The human mental stress testing system according to claim 9, wherein the system comprises:

The data calculation center that calculates the detection data and obtains the estimated pressure score;

The actual stress score obtained after passing the artificial mental stress test;

An error evaluation unit for evaluating at least one error level of the actual pressure score and the estimated pressure score;

According to the error level, adjust the parameter feedback modification unit of the evaluation parameter according to the preset error level and the corresponding relationship table of the evaluation parameter adjustment value;

The data calculation center re-evaluates the human mental pressure according to the evaluation parameters adjusted by the parameter feedback modification unit, and continuously evaluates the human body pressure after multiple adjustments, thereby achieving a gradual reduction in the error between the actual pressure score and the estimated pressure score Until it is within the medically recognized controllable range.