Health detection method and device and computer equipment

Abstract

The invention relates to the technical field of intelligent health management, and provides a health detection method and device and computer equipment. The method comprises the following steps: obtaining a mean KL divergence component and a variance KL divergence component of various health indexes according to various health index data of a detected object and a sample object without health problems; for any kind of health indexes, obtaining a first detection result according to the magnitude relationship between the mean value KL divergence component and the mean value control limit and the magnitude relationship between the variance KL divergence component and the variance control limit; obtaining a sensitive mean KL divergence component and a sensitive variance KL divergence component according to the sensitive health index data, corresponding to the target health problem, of the detected object and the sample object without the target health problem; a second detection result is obtained according to the size relation between the sensitive mean KL divergence component and the sensitive mean control limit and the size relation between the sensitive variance KL divergence component and the sensitive variance control limit; and obtaining a health detection result according to the first and second detection results. By adopting the method, the health detection precision is improved.

Description

Technical Field

[0001] The present application relates to the field of intelligent health management technology, and in particular to a health detection method, apparatus, computer equipment, computer-readable storage medium, and computer program product. Background Art

[0002] In response to the demand for medical care, the health data of the subject can be collected, and the health condition of the subject can be detected by analyzing the health data.

[0003] However, in the early stages of health problems, health data (health care data) are often easily obscured by some unknown interference and noise. They are generally characterized by low amplitude, concealment, randomness, and small deviation from the normal state, making it difficult to detect health problems in the subjects, resulting in low accuracy in health detection of the subjects. Summary of the invention

[0004] Based on this, it is necessary to provide a health detection method, device, computer equipment, computer readable storage medium and computer program product to address the above technical problems.

[0005] In a first aspect, the present application provides a health detection method, comprising:

[0006] According to the data of various health indicators of the subjects and the data of various health indicators of the sample subjects without health problems, the KL divergence components of various health indicators are obtained to obtain the mean KL divergence components and variance KL divergence components of various health indicators;

[0007] According to the mean KL divergence component and variance KL divergence component of various health indicators without health problems, the mean control limit and variance control limit of various health indicators are obtained respectively;

[0008] According to the size relationship between the mean KL divergence components of various health indicators and the mean control limits of the corresponding health indicators, and according to the size relationship between the variance KL divergence components of various health indicators and the variance control limits of the corresponding health indicators, a first detection result is obtained;

[0009] Determine sensitive health indicators corresponding to target health issues from various health indicators;

[0010] According to the data of the sensitive health indicators of the tested object and the data of the sensitive health indicators of the sample object without the target health problem, the KL divergence component of the sensitive health indicator is obtained to obtain the sensitive mean KL divergence component and the sensitive variance KL divergence component of the sensitive health indicator;

[0011] According to the sensitive mean KL divergence component and sensitive variance KL divergence component of the sensitive health indicators without target health problems, the sensitive mean control limit and sensitive variance control limit of the sensitive health indicators are obtained respectively;

[0012] A second detection result is obtained according to the size relationship between the sensitive mean KL divergence component of the sensitive health indicator and the sensitive mean control limit, and the size relationship between the sensitive variance KL divergence component of the sensitive health indicator and the sensitive variance control limit;

[0013] A health test result of the subject is obtained according to the first test result and the second test result.

[0014] In one embodiment, the KL divergence components of various health indicators are obtained based on the data of various health indicators of the subject and the data of various health indicators of the sample subject without health problems, including:

[0015] Performing dimensionality reduction on the data of various health indicators of the inspected subjects and the data of various health indicators of the sample subjects without health problems, to obtain the inspected data of various health indicators and the sample data of various health indicators after dimensionality reduction;

[0016] In the sliding window, when the various health indicator detected data and the various health indicator sample data slide, the various health indicator detected data segments and the various health indicator sample data segments in the same time window are determined;

[0017] According to the detected data segments of various health indicators in the same time window, first means and first variances of various health indicators are obtained;

[0018] According to the sample data segments of various health indicators in the same time window, second means and second variances of various health indicators are obtained;

[0019] According to the first mean, first variance, second mean and second variance of each health index, the KL divergence component of each health index is obtained.

[0020] In one embodiment, obtaining the mean KL divergence component and the variance KL divergence component of each health indicator includes:

[0021] According to the first mean and the second mean of each health index, the mean difference of each health index is obtained;

[0022] According to the mean difference of each health index and the KL divergence component of each health index, the mean KL divergence component of each health index is obtained;

[0023] According to the first variance and second variance of each health index, the variance square ratio of each health index is obtained;

[0024] According to the variance square ratios of the various health indicators and the KL divergence components of the various health indicators, the variance KL divergence components of the various health indicators are obtained.

[0025] In one embodiment, according to the KL divergence components of the mean values ​​of various health indicators in the absence of health problems, the mean control limits of various health indicators are obtained, including:

[0026] For any mean KL divergence component of any health indicator, obtain an average distance between the mean KL divergence component and the mean KL divergence component of the surrounding area;

[0027] According to the average distance, obtaining the domain density of any mean KL divergence component;

[0028] Taking the average value of the ratio of the domain density of any mean KL divergence component to the domain density of the mean KL divergence component in the surrounding area as the local anomaly factor value of any mean KL divergence component;

[0029] According to the relative size of the local anomaly factor values ​​of each mean KL divergence component, the minimum mean KL divergence component and the maximum mean KL divergence component are obtained;

[0030] The mean control limit of any health indicator is obtained according to the minimum mean KL divergence component and the maximum mean KL divergence component.

[0031] In one embodiment, determining the sensitive health indicator corresponding to the target health problem from various health indicators includes:

[0032] Obtain the contribution rate of various health indicators to target health issues;

[0033] The contribution rates are sorted from large to small, starting from the contribution rate with the first ranking, and the contribution rate of the current ranking is added to the contribution rate before the current ranking to obtain the cumulative contribution rate;

[0034] Obtaining a correlation coefficient ratio according to the cumulative contribution rate and the total added value of the contribution rate;

[0035] According to the relative size between the correlation coefficient ratio and the correlation coefficient ratio threshold, a sensitive health indicator corresponding to the target health problem is determined from various health indicators.

[0036] In one embodiment, obtaining the contribution rate of various health indicators to the target health problem includes:

[0037] According to the data of various health indicators of the sample object without the target health problem and the data of various health indicators of another sample object without the target health problem, the KL divergence components of various health indicators for the target health problem are obtained;

[0038] According to the KL divergence components of the various health indicators for the target health problem, a total value of the KL divergence components is obtained;

[0039] The ratio of the KL divergence component of each type of health indicator for the target health problem to the total value of the KL divergence component is used as the contribution rate of each type of health indicator for the target health problem.

[0040] In a second aspect, the present application also provides a health detection device, comprising:

[0041] The first test result acquisition module is used to obtain the KL divergence components of various health indicators based on the data of various health indicators of the tested object and the data of various health indicators of the sample object without health problems, so as to obtain the mean KL divergence components and variance KL divergence components of various health indicators; according to the mean KL divergence components and variance KL divergence components of various health indicators without health problems, respectively obtain the mean control limits and variance control limits of various health indicators; according to the size relationship between the mean KL divergence components of various health indicators and the mean control limits of health indicators of corresponding categories, and according to the size relationship between the variance KL divergence components of various health indicators and the variance control limits of health indicators of corresponding categories, obtain the first test result;

[0042] A sensitive health indicator acquisition module is used to determine the sensitive health indicator corresponding to the target health problem from various health indicators;

[0043] The second detection result acquisition module is used to obtain the KL divergence component of the sensitive health indicator according to the data of the sensitive health indicator of the detected object and the data of the sensitive health indicator of the sample object without target health problems, so as to obtain the sensitive mean KL divergence component and the sensitive variance KL divergence component of the sensitive health indicator; according to the sensitive mean KL divergence component and the sensitive variance KL divergence component of the sensitive health indicator without target health problems, respectively obtain the sensitive mean control limit and the sensitive variance control limit of the sensitive health indicator; according to the size relationship between the sensitive mean KL divergence component and the sensitive mean control limit of the sensitive health indicator, and the size relationship between the sensitive variance KL divergence component and the sensitive variance control limit of the sensitive health indicator, obtain the second detection result;

[0044] The health test result acquisition module is used to obtain the health test result of the object under test according to the first test result and the second test result.

[0045] In a third aspect, the present application further provides a computer device, wherein the computer device comprises a memory and a processor, wherein the memory stores a computer program, and the processor executes the above method.

[0046] In a fourth aspect, the present application further provides a computer-readable storage medium, wherein a computer program is stored on the computer-readable storage medium, and the computer program is executed by a processor to execute the above method.

[0047] In a fifth aspect, the present application further provides a computer program product, wherein the computer program product comprises a computer program, and the computer program is executed by a processor to execute the above method.

[0048] The above-mentioned health detection method, device, computer equipment, computer-readable storage medium and computer program product obtain the KL divergence components of various health indicators based on the data of various health indicators of the detected object and the data of various health indicators of the sample objects without health problems, so as to obtain the mean KL divergence components and variance KL divergence components of various health indicators; obtain the mean control limits and variance control limits of various health indicators respectively according to the mean KL divergence components and variance KL divergence components of various health indicators in the case of no health problems; obtain the first detection result according to the size relationship between the mean KL divergence components of various health indicators and the mean control limits of the health indicators of the corresponding categories, and according to the size relationship between the variance KL divergence components of various health indicators and the variance control limits of the health indicators of the corresponding categories; determine the corresponding health indicators from the various health indicators Sensitive health indicators corresponding to the target health problems; obtaining KL divergence components of the sensitive health indicators based on the data of the sensitive health indicators of the subjects and the data of the sensitive health indicators of the sample subjects without the target health problems, so as to obtain the sensitive mean KL divergence components and sensitive variance KL divergence components of the sensitive health indicators; obtaining the sensitive mean control limits and sensitive variance control limits of the sensitive health indicators respectively based on the sensitive mean KL divergence components and sensitive variance KL divergence components of the sensitive health indicators without the target health problems; obtaining the second test result based on the size relationship between the sensitive mean KL divergence components and the sensitive mean control limits of the sensitive health indicators, and the size relationship between the sensitive variance KL divergence components and the sensitive variance control limits of the sensitive health indicators; obtaining the health test result of the subject based on the first test result and the second test result. This application uses the mean KL divergence component and the variance KL divergence component to measure the probability distribution of the data of various health indicators of the subject, and the information loss between the probability distribution of the data of various health indicators of the sample object without health problems. Even a small deviation can be detected by the mean KL divergence component and the variance KL divergence component. Therefore, the health problems of the subject can be detected in the early stage of the health problems of the subject, thereby improving the accuracy of health detection. This application also determines the sensitive health indicators corresponding to the target health problems from various health indicators, and obtains the second detection result based on the size relationship between the sensitive mean KL divergence component and the sensitive mean control limit of the sensitive health indicator, and the size relationship between the sensitive variance KL divergence component and the sensitive variance control limit of the sensitive health indicator; based on the first detection result and the second detection result, the health detection result of the subject is obtained, which improves the detection accuracy for certain target health problems that are difficult to detect, and further improves the accuracy of health detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the drawings required for use in the embodiments of the present application or related technical descriptions will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0050] Figure 1 A diagram showing an application environment of a health detection method in an embodiment;

[0051] Figure 2 A schematic diagram of a health detection method according to an embodiment of the present invention;

[0052] Figure 3 A schematic diagram of a process for obtaining KL divergence components of various health indicators in one embodiment;

[0053] Figure 4 is a structural block diagram of a health detection device in one embodiment;

[0054] Figure 5 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0055] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0056] It should be noted that the data of various health indicators involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0057] The present application embodiment provides a health detection method, which can be executed by a computer device, such as Figure 1 As shown, the computer device can obtain data on various health indicators of the subject, data on various health indicators of sample subjects without health problems, and data on sensitive health indicators of sample subjects without target health problems, thereby obtaining health test results of the subject. It is understandable that the computer device can be implemented through a server, a terminal, or an interactive system between a terminal and a server. In this embodiment, the method includes Figure 2 The steps shown are:

[0058] Step S201, obtaining KL divergence components of various health indicators based on the data of various health indicators of the subject and the data of various health indicators of the sample subject without health problems, so as to obtain the mean KL divergence component and variance KL divergence component of various health indicators.

[0059] With the help of new health care remote control hardware equipment, data on various health indicators of the subjects can be collected at any time.

[0060] Data on various health indicators of sample subjects without health problems can be obtained from the historical measurement data of the new health care remote control hardware device.

[0061] KL divergence (Kullback-Leibler Divergence, KLD), also known as relative entropy or information divergence, is an asymmetric measure of the difference between two probability distributions P and Q. KL divergence represents the information loss when one probability distribution Q is used to approximate another probability distribution P. As a classic loss function, KL divergence is widely used in machine learning tasks such as cluster analysis and parameter estimation.

[0062] This application can use the KL divergence components of various health indicators to quantify the differences in the probability distribution of data of various health indicators under conditions of no health problems and when health problems occur, and determine whether a health problem occurs based on the differences.

[0063] The calculation formula of the KL divergence component is shown in formula (1).

[0064] (1)

[0065] Among them, q(x) represents the probability of the data of various health indicators of the subjects, and p(x) represents the probability of the data of various health indicators of the sample subjects without health problems. represents the difference between probability q and probability p.

[0066] Because the definition of the KL divergence component is asymmetric, the original definition of the KL divergence component is not suitable for use as a distance metric. This application adopts an improved symmetric form of the KL divergence component calculation formula, as shown in formula (2).

[0067] (2)

[0068] If both probability distribution data are normally distributed data, equation (2) can be simplified to equation (3).

[0069] (3)

[0070] Among them, q(x) represents the probability of the data of various health indicators of the subjects, and p(x) represents the probability of the data of various health indicators of the sample subjects without health problems. represents the difference between probability q and probability p, Represents the variance of the data of various health indicators of the subjects, Represents the mean value of various health indicators of the subjects. The variance of the data of various health indicators of sample subjects without health problems, The data represents the mean of various health indicators of sample subjects without health problems.

[0071] The KL divergence components of various health indicators of the inspected object can be obtained according to the probability distribution of the data of various health indicators of the inspected object and the probability distribution of the data of various health indicators of the sample object without health problems. The probability distribution of the data of various health indicators of the inspected object in actual situations and the probability distribution of the data of various health indicators of the sample object without health problems are not necessarily normally distributed. For this, a sliding window can be used to intercept the data of various health indicators of the inspected object and the data of various health indicators of the sample object with health problems, because a section of continuous data in the window can be approximately considered to be normally distributed.

[0072] For the KL divergence component between two different normally distributed data, it can be found that it is mainly affected by the mean and variance. However, not all health indicators in the data of various health indicators will trigger these two factors at the same time. Some health indicators may only produce mean fluctuations, while others may only have variance drifts. Therefore, if the overall data of the KL divergence component is simply considered, the change of one of the factors may be covered. Therefore, it is necessary to pay attention to the changes in the KL divergence components caused by the mean and variance separately.

[0073] According to the KL divergence components of various health indicators, the mean KL divergence components and variance KL divergence components of various health indicators can be obtained. According to the mean KL divergence components and variance KL divergence components of various health indicators, the difference in the probability distribution of the data of various health indicators under the conditions of no health problems and health problems can be quantified, and whether health problems occur can be judged based on the differences.

[0074] Step S202, according to the mean KL divergence component and variance KL divergence component of each health indicator in the case of no health problems, the mean control limit and variance control limit of each health indicator are obtained respectively.

[0075] Based on the data of various health indicators of the sample objects without health problems and the data of various health indicators of another sample object without health problems, the KL divergence components of various health indicators in the case without health problems can be obtained, so as to obtain the mean KL divergence component and variance KL divergence component of various health indicators in the case without health problems.

[0076] According to the mean KL divergence components of various health indicators without health problems, the abnormal degree of the mean KL divergence components of various health indicators without health problems can be judged, thereby obtaining the mean control limits of various health indicators. The mean control limits of various health indicators represent the value range of the mean KL divergence components of various health indicators without health problems.

[0077] According to the variance KL divergence components of various health indicators without health problems, the abnormality of the variance KL divergence components of various health indicators without health problems can be judged, thereby obtaining the variance control limits of various health indicators. The variance control limits of various health indicators represent the value range of the variance KL divergence components of various health indicators without health problems.

[0078] Step S203, obtaining a first detection result according to the size relationship between the mean KL divergence component of each health indicator and the mean control limit of the health indicator of the corresponding category, and according to the size relationship between the variance KL divergence component of each health indicator and the variance control limit of the health indicator of the corresponding category.

[0079] When the mean KL divergence component of any type of health indicator is not within the mean control limit of the health indicator of the corresponding category, or the variance KL divergence component of any type of health indicator is not within the variance control limit of the health indicator of the corresponding category, the probability distribution of the data representing the various types of health indicators of the inspected object deviates from the probability distribution of the data of the various types of health indicators of the sample object without health problems, then it can be determined that the probability of the inspected object having health problems is greater than the threshold, and the first test result can be obtained. The first test result represents that the probability of the inspected object having health problems is greater than the threshold.

[0080] When the mean KL divergence components of various health indicators are within the mean control limits of the corresponding health indicators, and the variance KL divergence components of various health indicators are within the variance control limits of the corresponding health indicators, the probability distribution of the data representing the various health indicators of the inspected object does not deviate from the probability distribution of the data of the various health indicators of the sample object without health problems. At this time, it can be determined that the probability of the inspected object having health problems is less than the threshold, and the first test result can be obtained. The first test result represents that the probability of the inspected object having health problems is less than the threshold.

[0081] Step S204, determining a sensitive health indicator corresponding to the target health problem from various health indicators.

[0082] Different health problems have different sensitive health indicators. Sensitive health indicators of specific health problems can be used for health detection, making health detection more targeted and improving the detection performance of specific health problems that are difficult to detect. Specific health problems that are difficult to detect can be used as target health problems.

[0083] In real health detection scenarios, there is prior knowledge of some specific health problems that are difficult to detect. Based on the prior knowledge of the target health problem, the sensitive health indicators corresponding to the target health problem can be determined from various health indicators. For example, a larger KL divergence component is considered to have a greater impact on health problems. Based on this content, the sensitive health indicators corresponding to the target health problem can be determined from various health indicators. The sensitive health indicators corresponding to the target health problem can be multiple sensitive health indicators, each of which is divided into a single specific small subspace, and the target health problems are detected simultaneously. In addition, for multiple target health problems, the sensitive health indicators corresponding to each target health problem can be determined from various health indicators at the same time, and multiple target health problems can be detected simultaneously.

[0084] Step S205, obtaining the KL divergence component of the sensitive health indicator based on the data of the sensitive health indicator of the subject and the data of the sensitive health indicator of the sample subject without target health problems, so as to obtain the sensitive mean KL divergence component and the sensitive variance KL divergence component of the sensitive health indicator.

[0085] The sensitive mean KL divergence component and the sensitive variance KL divergence component can be used to quantify the differences in the probability distribution of data of various health indicators under the conditions of no target health problem and when the target health problem occurs, and whether the target health problem occurs can be determined based on the differences.

[0086] The KL divergence component of the sensitive health indicator of the inspected object can be obtained according to the probability distribution of the data of the sensitive health indicator of the inspected object and the probability distribution of the data of the sensitive health indicator of the sample object without the target health problem.

[0087] According to the KL divergence components of the sensitive health indicators, the sensitive mean KL divergence components and the sensitive variance KL divergence components of the sensitive health indicators are obtained.

[0088] Step S206, according to the sensitive mean KL divergence component and the sensitive variance KL divergence component of the sensitive health indicator in the case of no target health problem, respectively obtain the sensitive mean control limit and the sensitive variance control limit of the sensitive health indicator.

[0089] Based on the data of various sensitive health indicators of the sample objects without target health problems and the data of various sensitive health indicators of another sample object without target health problems, the KL divergence components of the various sensitive health indicators in the case without target health problems can be obtained, so as to obtain the sensitive mean KL divergence component and sensitive variance KL divergence component of the sensitive health indicators in the case without target health problems.

[0090] According to the sensitive mean KL divergence components of various sensitive health indicators without target health problems, the abnormal degree of the sensitive mean KL divergence components of various sensitive health indicators without target health problems can be judged, thereby obtaining the sensitive mean control limits of various sensitive health indicators. Among them, the sensitive mean control limits of various sensitive health indicators represent the value range of the sensitive mean KL divergence components of various sensitive health indicators without target health problems.

[0091] According to the KL divergence components of the sensitive variances of various sensitive health indicators without target health problems, the abnormality of the KL divergence components of the sensitive variances of various sensitive health indicators without target health problems can be judged, thereby obtaining the sensitive variance control limits of various sensitive health indicators. Among them, the sensitive variance control limits of various sensitive health indicators represent the value range of the KL divergence components of the sensitive variances of various sensitive health indicators without target health problems.

[0092] Step S207, obtaining a second detection result according to the size relationship between the sensitive mean KL divergence component of the sensitive health indicator and the sensitive mean control limit, and the size relationship between the sensitive variance KL divergence component of the sensitive health indicator and the sensitive variance control limit.

[0093] When the sensitive mean KL divergence component of the sensitive health indicator is not within the sensitive mean control limit, or the sensitive variance KL divergence component of the sensitive health indicator is not within the sensitive variance control limit, the probability distribution of the data representing the various sensitive health indicators of the inspected object deviates from the probability distribution of the data of the various sensitive health indicators of the sample object without the target health problem, and it can be determined that the probability of the inspected object having the target health problem is greater than the threshold, and the second test result can be obtained at this time. The second test result represents that the probability of the inspected object having the target health problem is greater than the threshold.

[0094] When the sensitive mean KL divergence component of the sensitive health indicator is within the sensitive mean control limit, and the sensitive variance KL divergence component of the sensitive health indicator is within the sensitive variance control limit, the probability distribution of the data representing the sensitive health indicator of the inspected object does not deviate from the probability distribution of the data of various sensitive health indicators of the sample object without the target health problem, and it can be determined that the probability of the inspected object having the target health problem is less than the threshold value, and the second test result can be obtained at this time. The second test result represents that the probability of the inspected object having the target health problem is less than the threshold value.

[0095] Step S208, obtaining a health test result of the subject according to the first test result and the second test result.

[0096] Based on Bayesian theory, the first detection result and the second detection result can be fused to obtain the health detection result of the object being detected.

[0097] In the above health detection method, the probability distribution of the data of various health indicators of the object being detected is measured by the mean KL divergence component and the variance KL divergence component, and the information loss between the probability distribution of the data of various health indicators of the sample object without health problems, even a small deviation can be detected by the mean KL divergence component and the variance KL divergence component, so the health problems of the object being detected can be detected in the early stage of the health problems of the object being detected, thereby improving the accuracy of health detection. The present application also determines the sensitive health indicators corresponding to the target health problems from various health indicators, and obtains the second detection result based on the size relationship between the sensitive mean KL divergence component and the sensitive mean control limit of the sensitive health indicator, and the size relationship between the sensitive variance KL divergence component and the sensitive variance control limit of the sensitive health indicator; based on the first detection result and the second detection result, the health detection result of the object being detected is obtained, the detection accuracy for certain target health problems that are difficult to detect is improved, and the accuracy of health detection is further improved.

[0098] In one embodiment, the KL divergence components of various health indicators are obtained based on the data of various health indicators of the subject and the data of various health indicators of the sample subject without health problems. The specific steps are as follows: Figure 3 As shown, step S301, the data of various health indicators of the inspected objects and the data of various health indicators of the sample objects without health problems are reduced in dimension to obtain the inspected data of various health indicators and the sample data of various health indicators after dimension reduction; step S302, in the sliding window, the inspected data segments of various health indicators and the sample data segments of various health indicators in the same time window are determined; step S303, according to the inspected data segments of various health indicators in the same time window, the first mean and the first variance of various health indicators are obtained; step S304, according to the sample data segments of various health indicators in the same time window, the second mean and the second variance of various health indicators are obtained; step S305, according to the first mean, the first variance, the second mean and the second variance of various health indicators, the KL divergence components of various health indicators are obtained.

[0099] This embodiment is introduced by taking the KL divergence component of the i-th type of health indicator as an example.

[0100] According to the Neighborhood Preserving Embedding (NPE) algorithm, the data of the i-th health indicator of the inspected object and the data of the i-th health indicator of the sample object without health problems can be reduced in dimension to obtain the i-th health indicator inspected data and the i-th health indicator sample data after the dimension reduction. According to the i-th health indicator inspected data and the i-th health indicator sample data after the dimension reduction, the projection matrix of the i-th health indicator inspected data and the i-th health indicator sample data, that is, the state space, can be obtained.

[0101] The i-th health indicator detected data segment and the i-th health indicator sample data segment in the same time window can be determined when the i-th health indicator detected data and the i-th health indicator sample data are sliding in the sliding window. The i-th health indicator detected data segment is shown in formula (4).

[0102] (4)

[0103] in, Represents the data of the i-th health indicator at time te.

[0104] The data segments of various health indicators constitute a time series describing the dynamic trend of data, reflecting the evolution of various health indicators. In order to enhance the changing trend of data in the evolution of health problems, it is necessary to give different weights to health indicators at different times to highlight their dynamic changes.

[0105] The exponentially weighted moving-average (EWMA) processing can be performed on the i-th health indicator test data segment in the same time window, as shown in formula (5). According to the i-th health indicator test data segment after the exponentially weighted moving-average processing, the first mean and the first variance of the i-th health indicator are obtained.

[0106] (5)

[0107] in, It represents the inspected data of the i-th health indicator at time t, ω represents the weight, w represents the total number of inspected health indicator data, j is each time, and the value of j is accumulated from 1 to w.

[0108] The larger ω value indicates that the detected data of the health indicator at the current moment plays an important role, while the detected data of the health indicator at the past moment has a weaker influence. However, a larger ω also leads to an increase in the impact of noise. On the contrary, a smaller ω can smooth the impact of noise and reflect the process trend. However, a smaller ω will weaken the importance of the new detected data of the health indicator and may increase the delay of health detection.

[0109] The i-th health indicator sample data segment in the same time window may be subjected to exponentially weighted moving average processing, and the second mean and second variance of the i-th health indicator may be obtained based on the i-th health indicator sample data segment after the exponentially weighted moving average processing.

[0110] The KL divergence component of the i-th health indicator can be obtained according to the first mean, first variance, second mean and second variance of the i-th health indicator, as shown in formula (3). Among them, the various health indicator detected data segments and various health indicator sample data segments in a certain time window can be used as the various health indicator detected data segments and various health indicator sample data segments corresponding to the boundary moment of the window, so that the various health indicator detected data segments and various health indicator sample data segments corresponding to each moment can be obtained, so as to obtain the KL divergence components of various health indicators at each moment.

[0111] (3)

[0112] In this embodiment, based on the various health indicator tested data and various health indicator sample data after dimensionality reduction, the various health indicator tested data segments and various health indicator sample data segments in the same time window are determined; based on the first mean and first variance of the various health indicator tested data segments in the same time window, as well as the second mean and second variance of the various health indicator sample data segments in the same time window, the KL divergence components of various health indicators can be quickly obtained.

[0113] In one of the embodiments, the mean KL divergence component and variance KL divergence component of each health indicator are obtained, and the specific steps are as follows: according to the first mean and the second mean of each health indicator, the mean difference of each health indicator is obtained; according to the mean difference of each health indicator and the KL divergence component of each health indicator, the mean KL divergence component of each health indicator is obtained; according to the first variance and the second variance of each health indicator, the variance square ratio of each health indicator is obtained; according to the variance square ratio of each health indicator and the KL divergence component of each health indicator, the variance KL divergence component of each health indicator is obtained.

[0114] Through the method introduced in this embodiment, the mean KL divergence component and variance KL divergence component of each type of health indicator can be obtained, and the mean KL divergence component and variance KL divergence component of the i-th type of health indicator are taken as an example for introduction.

[0115] The first mean of the i-th health indicator can be Subtract the second mean , get the mean difference of the i-th health indicator The product of the mean difference of the i-th health indicator and the KL divergence component of the i-th health indicator can be used as the mean KL divergence component of the i-th health indicator, as shown in formula (6).

[0116] (6)

[0117] According to the first variance of the i-th health indicator and the second variance , get the variance square ratio of the i-th health index The product of the square ratio of the variance of the i-th health indicator and the KL divergence component of the i-th health indicator can be used as the variance KL divergence component of the i-th health indicator, as shown in formula (7).

[0118] (7)

[0119] In this embodiment, the mean KL divergence component of each health indicator is obtained according to the mean difference of each health indicator and the KL divergence component of each health indicator; the variance KL divergence component of each health indicator is obtained according to the variance square ratio of each health indicator and the KL divergence component of each health indicator, so as to prepare data for quantifying the difference in probability distribution of data of each health indicator under the conditions of no health problems and when health problems occur according to the mean KL divergence component and the variance KL divergence component of each health indicator.

[0120] In one of the embodiments, the mean control limits of various health indicators are obtained based on the mean KL divergence components of various health indicators in the absence of health problems. The specific steps are as follows: for any mean KL divergence component of any health indicator, the average distance between any mean KL divergence component and the mean KL divergence component of the surrounding area is obtained; based on the average distance, the field density of any mean KL divergence component is obtained; the average value of the ratio of the field density of any mean KL divergence component to the field density of the mean KL divergence component of the surrounding area is used as the local anomaly factor value of any mean KL divergence component; based on the relative sizes of the local anomaly factor values ​​of each mean KL divergence component, the minimum mean KL divergence component and the maximum mean KL divergence component are obtained; based on the minimum mean KL divergence component and the maximum mean KL divergence component, the mean control limit of any health indicator is obtained.

[0121] Through the method introduced in this embodiment, the mean control limit of each type of health indicator can be obtained, and the method of obtaining the mean control limit of the i-th type of health indicator is used as an example for introduction.

[0122] After obtaining multiple mean KL divergence components of the i-th health indicator, the local anomaly factor value of each mean KL divergence component can be determined. Taking the local anomaly factor value of the n-th mean KL divergence component (KLD_i_n) of the i-th health indicator as an example, the process of obtaining the local anomaly factor value is introduced.

[0123] For the mean KL divergence component KLD_i_n of the i-th health indicator, obtain the distance between the mean KL divergence component KLD_i_n and each mean KL divergence component of the surrounding area; solve the mean of the distance between the mean KL divergence component KLD_i_n and each mean KL divergence component of the surrounding area to obtain the average distance between the mean KL divergence component KLD_i_n and the mean KL divergence component of the surrounding area.

[0124] According to the average distance, the domain density of the mean KL divergence component KLD_i_n is obtained; the average value of the ratio of the domain density of the mean KL divergence component KLD_i_n to the domain density of the mean KL divergence component of the surrounding area is taken as the local outlier factor (LOF) value of the mean KL divergence component KLD_i_n. The size of the local outlier factor value indicates the degree of outlier value of the mean KL divergence component. Since the data of various health indicators of sample subjects without health problems are normally distributed, the mean KL divergence component at each time point is roughly in the same range.

[0125] According to the above method, the local anomaly factor value of each mean KL divergence component of the i-th health indicator can be obtained.

[0126] According to the relative size of the local anomaly factor values ​​of each mean KL divergence component of the i-th health indicator, the maximum local anomaly factor value and the minimum local anomaly factor value are obtained. The KL divergence component corresponding to the minimum local anomaly factor value is used as the minimum mean KL divergence component; the KL divergence component corresponding to the maximum local anomaly factor value is used as the maximum mean KL divergence component. The minimum mean KL divergence component is used as the lower limit of the mean control limit of the i-th health indicator, and the maximum mean KL divergence component is used as the upper limit of the mean control limit of the i-th health indicator.

[0127] In this embodiment, for any mean KL divergence component of any health indicator, according to the distance between any mean KL divergence component and the mean KL divergence component of the surrounding area, the average value of the ratio of the field density of any mean KL divergence component to the field density of the mean KL divergence component of the surrounding area is obtained, which is used as the local anomaly factor value of any mean KL divergence component. According to the relative sizes of the local anomaly factor values ​​of each mean KL divergence component, a more accurate mean control limit of any health indicator is obtained.

[0128] In one of the embodiments, sensitive health indicators corresponding to the target health problem are determined from various health indicators, and the specific steps are as follows: obtain the contribution rate of various health indicators to the target health problem; sort the contribution rates from large to small, starting with the first-ranked contribution rate, and sequentially add the contribution rate of the current ranking with the contribution rate before the current ranking to obtain the cumulative contribution rate; obtain the correlation coefficient ratio based on the cumulative contribution rate and the total sum of the contribution rates; determine the sensitive health indicator corresponding to the target health problem from various health indicators based on the relative size between the correlation coefficient ratio and the correlation coefficient ratio threshold.

[0129] In real health detection scenarios, there is prior knowledge of some specific health problems that are difficult to detect. Based on the prior knowledge of the target health problem, sensitive health indicators corresponding to the target health problem can be determined from various health indicators. For example, a larger KL divergence component is considered to have a greater impact on the health problem, so based on the contribution rate of various health indicators to the target health problem, sensitive health indicators corresponding to the target health problem can be determined from various health indicators.

[0130] Specifically, the contribution rates are sorted from large to small, starting from the first-ranked contribution rate, and the contribution rate of the current ranking is added to the contribution rate before the current ranking to obtain the cumulative contribution rate; according to the cumulative contribution rate and the total added value of the contribution rate, the correlation coefficient ratio (CCR) is obtained, as shown in formula (8).

[0131] (8)

[0132] Where CCR represents the correlation coefficient ratio, It represents the contribution rate of the ith health indicator to the jth target health problem, d represents the current ranking, and m represents the total number of categories of health indicators.

[0133] When the correlation coefficient ratio is greater than or equal to the correlation coefficient ratio threshold, determine the contribution rate that is less than or equal to the current ranking; based on the contribution rate that is less than or equal to the current ranking, determine the health indicator that is less than or equal to the current ranking; determine the health indicators among various health indicators that are less than or equal to the current ranking as sensitive health indicators corresponding to the target health problem.

[0134] Exemplarily, assuming that the total number of categories m of health indicators is 6, the contribution rates A, B, C, D, E, and F of the six categories of health indicators for the target health problems are obtained, where the contribution rate of the first category of health indicators is A, the contribution rate of the second category of health indicators is B, the contribution rate of the third category of health indicators is C, the contribution rate of the fourth category of health indicators is D, the contribution rate of the fifth category of health indicators is E, and the contribution rate of the sixth category of health indicators is F; the contribution rates are sorted from large to small to obtain a contribution rate sequence {C, A, B, D, F, E}, where the contribution rate C corresponds to a ranking of 1, the contribution rate A corresponds to a ranking of 2, the contribution rate B corresponds to a ranking of 3, the contribution rate D corresponds to a ranking of 4, the contribution rate F corresponds to a ranking of 5, and the contribution rate E corresponds to a ranking of 6.

[0135] First, let the current ranking d be equal to 1, then take the contribution rate C as the cumulative contribution rate, and take the ratio of the cumulative contribution rate to the total added value of the contribution rates as the correlation coefficient ratio when d is equal to 1. If the correlation coefficient ratio when d is equal to 1 is less than the correlation coefficient ratio threshold, then let the current ranking d be equal to 2; add the contribution rate C and the contribution rate A to obtain the cumulative contribution rate, and take the ratio of the cumulative contribution rate to the total added value of the contribution rates as the correlation coefficient ratio when d is equal to 2. If the correlation coefficient ratio when d is equal to 2 is greater than or equal to the correlation coefficient ratio threshold, then determine the contribution rates corresponding to less than or equal to the current ranking 2, which are contribution rate C and contribution rate A; the health indicator corresponding to the contribution rate C is health indicator 3, and the health indicator corresponding to the contribution rate A is health indicator 1. Therefore, the health indicators corresponding to less than or equal to the current ranking 2 include health indicator 1 and health indicator 3, and these two types of health indicators are determined as sensitive health indicators corresponding to the target health problems.

[0136] In this embodiment, the correlation coefficient ratio is obtained based on the cumulative contribution rate and the total sum of the contribution rates; based on the relative size between the correlation coefficient ratio and the correlation coefficient ratio threshold, the health indicators with a larger contribution rate to the detection target health problem are selected from various health indicators and determined as sensitive health indicators corresponding to the target health problem, so as to obtain relatively accurate sensitive health indicators corresponding to the target health problem.

[0137] In one of the embodiments, the contribution rate of each health indicator to the target health problem is obtained, and the specific steps are as follows: based on the data of each health indicator of a sample object without the target health problem and the data of each health indicator of another sample object without the target health problem, the KL divergence components of each health indicator for the target health problem are obtained; based on the KL divergence components of each health indicator for the target health problem, the total value of the KL divergence components is obtained; and the ratio of the KL divergence components of each health indicator for the target health problem to the total value of the KL divergence components is used as the contribution rate of each health indicator to the target health problem.

[0138] This embodiment is introduced by taking the contribution rate of the i-th type of health indicator to the target health problem as an example.

[0139] According to the probability distribution of the data of the i-th health indicator of the sample object without the target health problem and the probability distribution of the data of the i-th health indicator of another sample object without the target health problem, the KL divergence component of the i-th health indicator for the target health problem is obtained.

[0140] Assume that there are m health indicators, add the KL divergence components of the i-th health indicator for the target health problem, and get the total value of the KL divergence components. The ratio of the KL divergence component of the i-th health indicator for the target health problem to the total value of the KL divergence components is taken as the contribution rate of the i-th health indicator for the target health problem, as shown in formula (9).

[0141] (9)

[0142] in, represents the contribution rate of the i-th health indicator to the j-th target health problem, represents the KL divergence component of the i-th health indicator for the j-th target health problem, It represents the total value of the KL divergence component when the total number of categories of health indicators is m.

[0143] In this embodiment, the ratio of the KL divergence component of each health indicator for the target health problem to the total value of the KL divergence component is used as the contribution rate of each health indicator for the target health problem, so as to prepare data for determining the sensitive health indicators corresponding to the target health problem.

[0144] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0145] Based on the same inventive concept, the embodiment of the present application also provides a health detection device for implementing the health detection method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more health detection device embodiments provided below can refer to the limitations on the health detection method above, and will not be repeated here.

[0146] In an exemplary embodiment, Figure 4 As shown, a health detection device is provided, wherein:

[0147] The first test result acquisition module 401 is used to obtain the KL divergence components of various health indicators based on the data of various health indicators of the tested object and the data of various health indicators of the sample object without health problems, so as to obtain the mean KL divergence components and variance KL divergence components of various health indicators; according to the mean KL divergence components and variance KL divergence components of various health indicators without health problems, respectively obtain the mean control limits and variance control limits of various health indicators; according to the size relationship between the mean KL divergence components of various health indicators and the mean control limits of the health indicators of the corresponding categories, and according to the size relationship between the variance KL divergence components of various health indicators and the variance control limits of the health indicators of the corresponding categories, obtain the first test result;

[0148] A sensitive health indicator acquisition module 402 is used to determine a sensitive health indicator corresponding to a target health problem from various health indicators;

[0149] The second detection result acquisition module 403 is used to obtain the KL divergence component of the sensitive health indicator according to the data of the sensitive health indicator of the detected object and the data of the sensitive health indicator of the sample object without target health problems, so as to obtain the sensitive mean KL divergence component and the sensitive variance KL divergence component of the sensitive health indicator; according to the sensitive mean KL divergence component and the sensitive variance KL divergence component of the sensitive health indicator without target health problems, respectively obtain the sensitive mean control limit and the sensitive variance control limit of the sensitive health indicator; according to the size relationship between the sensitive mean KL divergence component and the sensitive mean control limit of the sensitive health indicator, and the size relationship between the sensitive variance KL divergence component and the sensitive variance control limit of the sensitive health indicator, obtain the second detection result;

[0150] The health test result acquisition module 404 is used to obtain the health test result of the object under test according to the first test result and the second test result.

[0151] In one of the embodiments, the first test result acquisition module 401 is also used to: reduce the dimension of the data of various health indicators of the tested objects and the data of various health indicators of the sample objects without health problems, and obtain the tested data of various health indicators and the sample data of various health indicators after dimension reduction; determine the tested data segments of various health indicators and the sample data segments of various health indicators in the same time window while the tested data of various health indicators and the sample data of various health indicators are sliding in the sliding window; obtain the first mean and the first variance of various health indicators according to the tested data segments of various health indicators in the same time window; obtain the second mean and the second variance of various health indicators according to the sample data segments of various health indicators in the same time window; obtain the KL divergence component of various health indicators according to the first mean, the first variance, the second mean and the second variance of various health indicators.

[0152] In one embodiment, the first detection result acquisition module 401 is also used to: obtain the mean difference of each health indicator according to the first mean and the second mean of each health indicator; obtain the mean KL divergence component of each health indicator according to the mean difference of each health indicator and the KL divergence component of each health indicator; obtain the variance square ratio of each health indicator according to the first variance and the second variance of each health indicator; obtain the variance KL divergence component of each health indicator according to the variance square ratio of each health indicator and the KL divergence component of each health indicator.

[0153] In one embodiment, the first detection result acquisition module 401 is also used to: for any mean KL divergence component of any health indicator, obtain the average distance between the any mean KL divergence component and the mean KL divergence component of the surrounding area; obtain the area density of the any mean KL divergence component based on the average distance; take the average value of the ratio of the area density of the any mean KL divergence component to the area density of the mean KL divergence component of the surrounding area as the local anomaly factor value of the any mean KL divergence component; obtain the minimum mean KL divergence component and the maximum mean KL divergence component based on the relative sizes of the local anomaly factor values ​​of each mean KL divergence component; and obtain the mean control limit of any health indicator based on the minimum mean KL divergence component and the maximum mean KL divergence component.

[0154] In one of the embodiments, the sensitive health indicator acquisition module 402 is also used to: obtain the contribution rate of each type of health indicator to the target health problem; sort the contribution rates from large to small, starting from the first-ranked contribution rate, and sequentially add the contribution rate of the current ranking with the contribution rate before the current ranking to obtain the cumulative contribution rate; obtain the correlation coefficient ratio according to the cumulative contribution rate and the total added value of the contribution rate; determine the sensitive health indicator corresponding to the target health problem from each type of health indicator according to the relative size between the correlation coefficient ratio and the correlation coefficient ratio threshold.

[0155] In one of the embodiments, the sensitive health indicator acquisition module 402 is also used to: obtain the KL divergence components of various health indicators for the target health problem based on the data of various health indicators of the sample objects without the target health problem and the data of various health indicators of another sample object without the target health problem; obtain the total value of the KL divergence components based on the KL divergence components of the various health indicators for the target health problem; and use the ratio of the KL divergence components of the various health indicators for the target health problem to the total value of the KL divergence components as the contribution rate of the various health indicators to the target health problem.

[0156] Each module in the above health detection device can be implemented in whole or in part by software, hardware and their combination. Each module can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to each module above.

[0157] In an exemplary embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as shown in FIG. Figure 5 As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. Among them, the processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data of an embodiment of the health detection method. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a health detection method is implemented.

[0158] Those skilled in the art will understand that Figure 5 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0159] In one embodiment, a computer device is further provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above method embodiments when executing the computer program.

[0160] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0161] In one embodiment, a computer program product is provided, including a computer program, which implements the steps in the above method embodiments when executed by a processor.

[0162] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, an artificial intelligence (AI) processor, etc., but are not limited to this.

[0163] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0164] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A health detection method, characterized in that: The method comprises: According to the data of various health indicators of the subjects and the data of various health indicators of the sample subjects without health problems, the KL divergence components of various health indicators are obtained to obtain the mean KL divergence components and variance KL divergence components of various health indicators; According to the mean KL divergence component and variance KL divergence component of various health indicators without health problems, the mean control limit and variance control limit of various health indicators are obtained respectively; According to the size relationship between the mean KL divergence components of various health indicators and the mean control limits of the corresponding health indicators, and according to the size relationship between the variance KL divergence components of various health indicators and the variance control limits of the corresponding health indicators, a first detection result is obtained; Determine sensitive health indicators corresponding to target health issues from various health indicators; According to the data of the sensitive health indicators of the tested object and the data of the sensitive health indicators of the sample object without the target health problem, the KL divergence component of the sensitive health indicator is obtained to obtain the sensitive mean KL divergence component and the sensitive variance KL divergence component of the sensitive health indicator; According to the sensitive mean KL divergence component and sensitive variance KL divergence component of the sensitive health indicators without target health problems, the sensitive mean control limit and sensitive variance control limit of the sensitive health indicators are obtained respectively; A second detection result is obtained according to the size relationship between the sensitive mean KL divergence component of the sensitive health indicator and the sensitive mean control limit, and the size relationship between the sensitive variance KL divergence component of the sensitive health indicator and the sensitive variance control limit; A health test result of the subject is obtained according to the first test result and the second test result.

2. The method according to claim 1, characterized in that The KL divergence components of various health indicators are obtained based on the data of various health indicators of the subject and the data of various health indicators of the sample subject without health problems, including: Performing dimensionality reduction on the data of various health indicators of the inspected subjects and the data of various health indicators of the sample subjects without health problems, to obtain the inspected data of various health indicators and the sample data of various health indicators after dimensionality reduction; In the sliding window, when the various health indicator detected data and the various health indicator sample data slide, the various health indicator detected data segments and the various health indicator sample data segments in the same time window are determined; According to the detected data segments of various health indicators in the same time window, first means and first variances of various health indicators are obtained; According to the sample data segments of various health indicators in the same time window, second means and second variances of various health indicators are obtained; According to the first mean, first variance, second mean and second variance of each health index, the KL divergence component of each health index is obtained.

3. The method according to claim 2, characterized in that Get the mean KL divergence component and variance KL divergence component of various health indicators, including: According to the first mean and the second mean of each health index, the mean difference of each health index is obtained; According to the mean difference of each health index and the KL divergence component of each health index, the mean KL divergence component of each health index is obtained; According to the first variance and second variance of each health index, the variance square ratio of each health index is obtained; According to the variance square ratios of the various health indicators and the KL divergence components of the various health indicators, the variance KL divergence components of the various health indicators are obtained.

4. The method according to claim 1, characterized in that: According to the KL divergence components of the mean values ​​of various health indicators without health problems, the mean control limits of various health indicators are obtained, including: For any mean KL divergence component of any health indicator, obtain an average distance between the mean KL divergence component and the mean KL divergence component of the surrounding area; According to the average distance, obtaining the domain density of any mean KL divergence component; Taking the average value of the ratio of the domain density of any mean KL divergence component to the domain density of the mean KL divergence component in the surrounding area as the local anomaly factor value of any mean KL divergence component; According to the relative size of the local anomaly factor values ​​of each mean KL divergence component, the minimum mean KL divergence component and the maximum mean KL divergence component are obtained; The mean control limit of any health indicator is obtained according to the minimum mean KL divergence component and the maximum mean KL divergence component.

5. The method according to claim 1, characterized in that The step of determining sensitive health indicators corresponding to target health problems from various health indicators includes: Obtain the contribution rate of various health indicators to target health issues; The contribution rates are sorted from large to small, starting from the contribution rate with the first ranking, and the contribution rate of the current ranking is added to the contribution rate before the current ranking to obtain the cumulative contribution rate; Obtaining a correlation coefficient ratio according to the cumulative contribution rate and the total added value of the contribution rate; According to the relative size between the correlation coefficient ratio and the correlation coefficient ratio threshold, a sensitive health indicator corresponding to the target health problem is determined from various health indicators.

6. The method according to claim 1, characterized in that The contribution rate of various health indicators to the target health problems is obtained, including: According to the data of various health indicators of the sample object without the target health problem and the data of various health indicators of another sample object without the target health problem, the KL divergence components of various health indicators for the target health problem are obtained; According to the KL divergence components of the various health indicators for the target health problem, a total value of the KL divergence components is obtained; The ratio of the KL divergence component of each type of health indicator for the target health problem to the total value of the KL divergence component is used as the contribution rate of each type of health indicator for the target health problem.

7. A health detection device, characterized in that: The device comprises: The first test result acquisition module is used to obtain the KL divergence components of various health indicators based on the data of various health indicators of the tested object and the data of various health indicators of the sample object without health problems, so as to obtain the mean KL divergence components and variance KL divergence components of various health indicators; according to the mean KL divergence components and variance KL divergence components of various health indicators without health problems, respectively obtain the mean control limits and variance control limits of various health indicators; according to the size relationship between the mean KL divergence components of various health indicators and the mean control limits of health indicators of corresponding categories, and according to the size relationship between the variance KL divergence components of various health indicators and the variance control limits of health indicators of corresponding categories, obtain the first test result; A sensitive health indicator acquisition module is used to determine the sensitive health indicator corresponding to the target health problem from various health indicators; The second detection result acquisition module is used to obtain the KL divergence component of the sensitive health indicator according to the data of the sensitive health indicator of the detected object and the data of the sensitive health indicator of the sample object without target health problems, so as to obtain the sensitive mean KL divergence component and the sensitive variance KL divergence component of the sensitive health indicator; according to the sensitive mean KL divergence component and the sensitive variance KL divergence component of the sensitive health indicator without target health problems, respectively obtain the sensitive mean control limit and the sensitive variance control limit of the sensitive health indicator; according to the size relationship between the sensitive mean KL divergence component and the sensitive mean control limit of the sensitive health indicator, and the size relationship between the sensitive variance KL divergence component and the sensitive variance control limit of the sensitive health indicator, obtain the second detection result; The health test result acquisition module is used to obtain the health test result of the object under test according to the first test result and the second test result.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images