Data processing method for smart big health system

Abstract

The present invention relates to the field of data processing technology, and specifically to a data processing method for a smart big health system, comprising: collecting health monitoring data; obtaining normal cycles and abnormal cycles in a health monitoring data curve based on the health monitoring data; obtaining abnormal cycles with long-term persistence characteristics in the health monitoring data curve and abnormal cycles without long-term persistence characteristics in the health monitoring data curve based on the abnormal cycles in the health monitoring data curve, obtaining a new health monitoring data curve in combination with the normal cycles in the health monitoring data curve, and compressing the new health monitoring data curve. The present invention increases the redundancy of less important health monitoring data and retains more important health monitoring data, so that the final compression result can take into account both improving the compression effect of the health monitoring data and ensuring that the compressed data can reflect the user's health status.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to a data processing method for a smart big health system. Background Art

[0002] Smart health systems can help people detect potential health issues early. Through regular health monitoring, individuals can better understand their physical condition and take timely intervention measures. Currently, health monitoring devices are commonly used in smart health systems for human health monitoring. They provide real-time health data. For example, by measuring and monitoring heart rate, symptoms of arrhythmia can be detected, stress levels can be assessed, and appropriate measures can be taken to relieve stress and anxiety, thereby promoting health. Because heart rate data is a crucial indicator in health monitoring, more health monitoring devices use HRV (heart rate variability) data for health monitoring. HRV refers to the amplitude and frequency of changes in the heart's spontaneous beat rhythm, which can reflect the functional status of the body's autonomic nervous system. By monitoring HRV, the relative activity of these two nervous systems can be understood, thereby assessing stress levels, emotional state, and overall physical condition. HRV is closely related to cardiovascular health. Abnormal HRV monitoring may be associated with an increased risk of heart disease, so it can also be used to assess cardiovascular health.

[0003] However, since HRV monitoring uses equipment to collect high-frequency data in real time, the collection process generates a massive amount of data information points. Data compression is usually used to process this type of data during monitoring. Although HRV curves have certain periodic similarities, due to various factors such as individual differences, physiological conditions, and environmental factors, the fluctuation characteristics of each period vary to a certain extent. Therefore, using traditional run-length encoding to compress this type of data will still result in low compression efficiency and large storage space. Periodic lossy compression will make it impossible to effectively identify abnormal HRV monitoring data, resulting in unsatisfactory monitoring and storage results. Summary of the Invention

[0004] The present invention provides a data processing method for a smart big health system to solve the existing problem: it is not possible to simultaneously improve the compression effect of health monitoring data and ensure that the compressed data can reflect the user's health status.

[0005] The data processing method for the smart health system of the present invention adopts the following technical solutions:

[0006] One embodiment of the present invention provides a data processing method for a smart big health system, the method comprising the following steps:

[0007] Collect health monitoring data; obtain each cycle in the health monitoring data curve, the data within each cycle in the health monitoring data, and the time span of each cycle in the health monitoring data based on the health monitoring data; obtain a first abnormality degree for each cycle in the health monitoring data curve based on the data within each cycle in the health monitoring data and the time span of each cycle in the health monitoring data; obtain a second abnormality degree for each cycle in the health monitoring data curve based on an inflection point in each cycle in the health monitoring data curve; obtain the abnormality degree of each cycle in the health monitoring data curve based on the first abnormality degree and the second abnormality degree of each cycle in the health monitoring data curve; and divide all cycles in the health monitoring data curve into two categories: abnormal cycles and normal cycles based on the abnormality degree of each cycle in the health monitoring data curve;

[0008] Obtaining the time nodes of all abnormal cycles in the health monitoring data curve; obtaining abnormal cycles with long-term persistence characteristics and abnormal cycles without long-term persistence characteristics in the health monitoring data curve based on the time nodes of all abnormal cycles in the health monitoring data curve;

[0009] According to the normal cycle in the health monitoring data curve, the abnormal cycle with long-term persistence characteristics in the health monitoring data curve, and the abnormal cycle without long-term persistence characteristics in the health monitoring data curve, a new health monitoring data curve is obtained and the new health monitoring data curve is compressed.

[0010] Preferably, the collecting of health monitoring data; obtaining each cycle in the health monitoring data curve, data within each cycle in the health monitoring data, and the time span of each cycle in the health monitoring data according to the health monitoring data, includes the following specific methods:

[0011] Allow users to collect health monitoring data through smart bracelets while keeping their bodies relaxed. Through the health monitoring data, the collected health monitoring data is imported into Kubios HRV software to obtain the health monitoring data curve, each cycle in the health monitoring data curve, the data in each cycle of the health monitoring data, and the time span of each cycle of the health monitoring data.

[0012] Preferably, the specific calculation formula for obtaining the first abnormality degree of each cycle in the health monitoring data curve is:

[0013]

[0014] Where, Indicates the first abnormality degree of the i-th cycle in the health monitoring data curve; F i represents the difference between the maximum and minimum values ​​in the i-th cycle of the health monitoring data curve; N represents the number of cycles in the health monitoring data curve; F j represents the difference between the maximum and minimum values ​​in the jth cycle of the health monitoring data curve; T i represents the time span of the i-th cycle in the health monitoring data curve; T j It represents the time span of the jth cycle in the health monitoring data curve; Norm() represents the linear normalization function; exp() represents the exponential function with a natural constant as the base; || represents the absolute value operation.

[0015] Preferably, the method of obtaining the second abnormality degree of each cycle in the health monitoring data curve according to the inflection point in each cycle in the health monitoring data curve includes the following specific methods:

[0016] Each cycle of the health monitoring data curve is divided into several segments according to the inflection points in each cycle of the health monitoring data curve to obtain subsegments of each cycle of the health monitoring data curve. The second abnormality degree of each cycle of the health monitoring data curve is obtained according to the number of inflection points in each cycle of the health monitoring data curve and the cycle subsegments in the health monitoring data curve. The specific calculation formula is:

[0017]

[0018] Where, Indicates the second abnormality degree of the i-th cycle in the health monitoring data curve; M i represents the number of inflection points in the i-th cycle of the health monitoring data curve; N represents the number of cycles in the health monitoring data curve; M j K represents the number of inflection points in the jth cycle of the health monitoring data curve; di,i represents the mean value of the slope of all data on the d-th sub-segment in the ith cycle of the health monitoring data curve; D i represents the number of sub-segments in the i-th cycle of the health monitoring data curve; K dj,j represents the mean value of the slope of all data on the dj-th sub-segment in the j-th cycle of the health monitoring data curve; D j represents the number of sub-segments in the j-th period of the health monitoring data curve; exp() represents an exponential function with a natural constant as the base; A1 and A2 are preset weights; || represents an absolute value operation.

[0019] Preferably, the specific calculation formula for obtaining the abnormality degree of each cycle in the health monitoring data curve is:

[0020]

[0021] Where σ i Indicates the abnormality degree of the i-th cycle in the health monitoring data curve; Indicates the first abnormality degree of the i-th cycle in the health monitoring data curve; Indicates the second abnormality level of the i-th cycle in the health monitoring data curve; μ1 and μ2 are preset weights.

[0022] Preferably, the method of dividing all cycles in the health monitoring data curve into two categories, abnormal cycles and normal cycles, according to the abnormality degree of each cycle in the health monitoring data curve includes the following specific methods:

[0023] An abnormality threshold τ is preset. When the abnormality of a period in the health monitoring data curve is greater than or equal to τ, the period in the health monitoring data curve is an abnormal period. When the abnormality of a period in the health monitoring data curve is less than τ, the period in the health monitoring data curve is a normal period.

[0024] Preferably, the specific method of obtaining the time nodes of all abnormal cycles in the health monitoring data curve includes:

[0025] For the oth abnormal cycle in the health monitoring data curve, the moment when the first data point in the oth abnormal cycle is recorded is selected as the time node of the oth abnormal cycle, and the time nodes of all abnormal cycles in the health monitoring data curve are obtained.

[0026] Preferably, the specific method of obtaining abnormal periods with long-term persistence characteristics in the health monitoring data curve and abnormal periods without long-term persistence characteristics in the health monitoring data curve is as follows:

[0027] For the oth abnormal cycle in the health monitoring data curve, the oth abnormal cycle is combined with each other abnormal cycle to obtain several abnormal cycle pairs of the oth abnormal cycle, and each abnormal cycle pair is combined with each other abnormal cycle pair to obtain several abnormal cycle pair groups of the oth abnormal cycle; the absolute value of the time node difference between the two abnormal cycles in each abnormal cycle pair is used as the time difference value of each abnormal cycle pair; the ratio of the small time difference value to the large time difference value of the two abnormal cycle pairs in the abnormal cycle pair group is used as the time difference similarity of each abnormal cycle pair group; finally, a threshold ξ is preset. When the ratio of the time difference similarity of any abnormal cycle pair group is greater than or equal to ξ, the oth abnormal cycle has the long-term persistence feature. When the ratio of the time difference similarity of all abnormal cycle pair groups is less than ξ, the oth abnormal cycle does not have the long-term persistence feature.

[0028] Preferably, the obtaining of a new health monitoring data curve and the compression processing of the new health monitoring data curve include the following specific methods:

[0029] Perform mean fitting on all normal cycles in the health monitoring data curve to obtain fitting results of all normal cycles in the health monitoring data curve; use the fitting results of all normal cycles in the health monitoring data curve to replace all normal cycles in the health monitoring data curve and all abnormal cycles that do not have long-term persistence characteristics to obtain a new health monitoring data curve, and finally use run-length encoding to compress the data corresponding to the new health monitoring data curve.

[0030] The beneficial effect of the technical solution of the present invention is that traditional data compression methods cannot, when compressing health monitoring data, simultaneously improve the compression effect of the health monitoring data and ensure that the compressed data can reflect the health condition of the user; while the present invention classifies the health monitoring data according to its importance, obtains health monitoring data of high importance and health monitoring data of low importance, increases the redundancy of the health monitoring data of low importance, retains the health monitoring data of high importance, obtains new health monitoring data, and compresses the new health monitoring data to achieve a final compression result that can simultaneously improve the compression effect of the health monitoring data and ensure that the compressed data can reflect the health condition of the user. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0032] Figure 1 This is a flowchart of the steps of the data processing method for the smart big health system of the present invention. DETAILED DESCRIPTION

[0033] To further illustrate the technical means and effects of the present invention to achieve the intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, describes in detail the data processing method for a smart health system proposed by the present invention, its specific implementation, structure, features, and effects. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, specific features, structures, or characteristics of one or more embodiments may be combined in any suitable form.

[0034] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0035] The specific scheme of the data processing method for the smart big health system provided by the present invention is described in detail below with reference to the accompanying drawings.

[0036] See also Figure 1 , which shows a flowchart of a data processing method for a smart big health system provided by one embodiment of the present invention, the method comprising the following steps:

[0037] Step S001: Collect health monitoring data.

[0038] It should be noted that the health monitoring data in this embodiment is HRV data. Since HRV data is high-frequency data collected in real time by the device, the collection process will generate a large number of data information points, so data compression is usually used to process the health monitoring data; although the health monitoring data has a certain degree of periodic similarity, under the influence of factors such as the user's health condition and psychological condition, there are certain differences in data of different periods; therefore, if the health monitoring data is directly compressed using traditional run-length encoding, a good compression effect cannot be obtained. If the health monitoring data is simply lossily compressed, the health monitoring data after lossy compression will lose some information, resulting in the health monitoring data after lossy compression cannot accurately reflect the user's health condition. Therefore, this embodiment proposes a data compression method that can improve the compression effect of health monitoring data while ensuring that the compressed data can reflect the user's health condition; therefore, it is necessary to collect health monitoring data first.

[0039] Specifically, users can collect health monitoring data through smart bracelets while keeping their bodies relaxed.

[0040] At this point, the health monitoring data is obtained.

[0041] Step S002: obtaining a health monitoring data curve according to the health monitoring data, classifying the cycles in the health monitoring data curve, and obtaining normal cycles and abnormal cycles in the health monitoring data curve.

[0042] It's important to note that run-length encoding data compression only works well for continuous, identical data segments. However, the data within each cycle of health monitoring data is not identical, so run-length encoding data compression doesn't work well for health monitoring data. However, the importance of normal data in health monitoring data is far lower than that of abnormal data. Therefore, lossy compression can be performed on normal data while fully retaining abnormal data. This improves the compression effect while ensuring that the compressed data reflects the user's health status. Therefore, it's necessary to first determine the degree of abnormality for each cycle in the health monitoring data curve.

[0043] Specifically, the collected health monitoring data is first imported into the Kubios HRV software to obtain the health monitoring data curve, each cycle in the health monitoring data curve, the data in each cycle of the health monitoring data, and the time span of each cycle of the health monitoring data.

[0044] It should be noted that the horizontal axis of the health monitoring data curve is the time when the health monitoring data is collected, and the vertical axis is the collected health monitoring data.

[0045] Then, based on the data in each cycle of the health monitoring data curve, the first abnormality degree of each cycle in the health monitoring data curve is obtained. The specific calculation formula is:

[0046]

[0047] Where, Indicates the first abnormality degree of the i-th cycle in the health monitoring data curve; F i represents the difference between the maximum and minimum values ​​in the i-th cycle of the health monitoring data curve; N represents the number of cycles in the health monitoring data curve; F j represents the difference between the maximum and minimum values ​​in the jth cycle of the health monitoring data curve; T i represents the time span of the i-th cycle in the health monitoring data curve; T j It represents the time span of the jth cycle in the health monitoring data curve; Norm() represents the linear normalization function; exp() represents the exponential function with a natural constant as the base, and || represents the absolute value operation.

[0048] It needs to be further explained that It represents the difference between the health data of the ith cycle in the health monitoring data curve and all cycles in the health monitoring data curve, so The larger the value of , the more abnormal the i-th cycle in the health monitoring data curve is; It represents the degree of difference in time span between the i-th cycle in the health monitoring data curve and all cycles in the health monitoring data curve; The larger the value of is, the more abnormal the i-th cycle in the health monitoring data curve is; so the obtained The larger it is, the more abnormal the i-th cycle in the health monitoring data curve is.

[0049] At this point, the first abnormality degree of each cycle in the health monitoring data curve is obtained.

[0050] Then, each cycle of the health monitoring data curve is divided into several segments according to the inflection points in each cycle of the health monitoring data curve to obtain subsegments of each cycle of the health monitoring data curve. According to the number of inflection points in each cycle of the health monitoring data curve and the cycle subsegments in the health monitoring data curve, the second abnormality degree of each cycle of the health monitoring data curve is obtained. The specific calculation formula is:

[0051]

[0052] Where, Indicates the second abnormality degree of the i-th cycle in the health monitoring data curve; M i represents the number of inflection points in the i-th cycle of the health monitoring data curve; N represents the number of cycles in the health monitoring data curve; M j K represents the number of inflection points in the jth cycle of the health monitoring data curve; di,i represents the mean value of the slope of all data on the d-th sub-segment in the ith cycle of the health monitoring data curve; D i represents the number of sub-segments in the i-th cycle of the health monitoring data curve; K dj,j represents the mean value of the slope of all data on the dj-th sub-segment in the j-th cycle of the health monitoring data curve; D j represents the number of sub-segments in the j-th period of the health monitoring data curve; exp() represents an exponential function with a natural constant as the base; A1 and A2 are preset weights. The sizes of A1 and A2 can be set according to actual conditions. This embodiment does not make a rigid requirement. In this embodiment, A1 and A2 are described as 0.6 and 0.4 respectively; || represents absolute value operation.

[0053] It needs to be further explained that It represents the difference between the number of inflection points in the ith cycle of the health monitoring data curve and the number of inflection points in all cycles of the health monitoring data curve. The larger the value of , the more abnormal the i-th cycle in the health monitoring data curve is; It represents the difference between the sub-segment in the ith cycle of the health monitoring data curve and the sub-segments in all cycles of the health monitoring data curve. The larger the value of , the The larger it is, the more abnormal the i-th cycle in the health monitoring data curve is.

[0054] At this point, the second abnormality degree of each cycle in the health monitoring data curve is obtained.

[0055] Next, according to the first abnormality degree of each cycle in the health monitoring data curve and the second abnormality degree of each cycle in the health monitoring data curve, the abnormality degree of each cycle in the health monitoring data curve is obtained. The specific calculation formula is:

[0056]

[0057] Where σ i Indicates the abnormality degree of the i-th cycle in the health monitoring data curve; Indicates the first abnormality degree of the i-th cycle in the health monitoring data curve; Indicates the second abnormality degree of the i-th cycle in the health monitoring data curve; μ1 and μ2 are preset weights. The sizes of μ1 and μ2 can be set according to actual conditions. This embodiment does not make a rigid requirement. In this embodiment, the calculation is performed with μ1 and μ2 both equal to 0.5.

[0058] It needs to be further explained that σ i The larger the value of , the more abnormal the i-th cycle in the health monitoring data curve is. Finally, according to the degree of abnormality of each cycle in the health monitoring data curve, each cycle in the health monitoring data curve is divided into two categories: abnormal cycle and normal cycle.

[0059] Specifically, an abnormality threshold τ is preset. The size of τ can be set according to actual conditions. This embodiment does not make a rigid requirement. In this embodiment, τ=0.7 is used for description. When the abnormality of the period in the health monitoring data curve is greater than or equal to τ, the period in the health monitoring data curve is an abnormal period. When the abnormality of the period in the health monitoring data curve is less than τ, the period in the health monitoring data curve is a normal period.

[0060] At this point, the normal period and abnormal period in the health monitoring data curve are obtained.

[0061] Step S003: Obtain the time nodes of all abnormal cycles in the health monitoring data curve; according to the time nodes of all abnormal cycles in the health monitoring data curve, obtain the abnormal cycles with long-term persistence characteristics in the health monitoring data curve and the abnormal cycles without long-term persistence characteristics in the health monitoring data curve.

[0062] It should be noted that there are many factors that can cause abnormal cycles in health monitoring data curves. If the abnormal cycles in the health monitoring data curve do not have long-term persistence characteristics, then the abnormal cycles in the health monitoring data curve are most likely caused by short-term factors such as exercise consumption or mood changes. Cycle abnormalities caused by short-term factors such as exercise consumption or mood changes are less important and can be considered normal cycles and lossy compressed together with normal cycles. If the abnormal cycles in the health monitoring data curve have long-term persistence characteristics, then they are most likely caused by factors such as heart disease. Cycle abnormalities caused by factors such as heart disease are more important and cannot be considered normal cycles. All characteristics of these abnormal cycles need to be retained.

[0063] Specifically, for the oth abnormal cycle in the health monitoring data curve, the oth abnormal cycle is combined with each other abnormal cycle to obtain several abnormal cycle pairs of the oth abnormal cycle, and each abnormal cycle pair is combined with each other abnormal cycle pair to obtain several abnormal cycle pair groups of the oth abnormal cycle; the absolute value of the time node difference between the two abnormal cycles in each abnormal cycle pair is used as the time difference value of each abnormal cycle pair; the ratio of the small time difference value to the large time difference value of the two abnormal cycle pairs in the abnormal cycle pair group is used as the time difference similarity of each abnormal cycle pair group; finally, a threshold ξ is preset. When the ratio of the time difference similarities of any abnormal cycle pair group is greater than or equal to ξ, the oth abnormal cycle has a long-term persistence feature. When the ratio of the time difference similarities of all abnormal cycle pair groups is less than ξ, the oth abnormal cycle does not have a long-term persistence feature.

[0064] Thus, the abnormal period with long-term persistence characteristics in the health monitoring data curve and the abnormal period without long-term persistence characteristics in the health monitoring data curve are obtained.

[0065] Step S004: according to the normal cycle in the health monitoring data curve, the abnormal cycle with long-term persistence characteristics in the health monitoring data curve, and the abnormal cycle without long-term persistence characteristics in the health monitoring data curve, a new health monitoring data curve is obtained and compressed.

[0066] It should be noted that since the importance of normal cycles and abnormal cycles without long-term persistence characteristics in the health monitoring data curve is low, while the importance of abnormal cycles with long-term persistence characteristics in the health monitoring data curve is high; therefore, when processing the health monitoring data, the data corresponding to the normal cycles and abnormal cycles without long-term persistence characteristics in the health monitoring data curve can be lossy compressed; the data corresponding to the abnormal cycles with long-term persistence characteristics in the health monitoring data curve are completely retained.

[0067] Specifically, mean fitting is performed on all normal cycles in the health monitoring data curve to obtain fitting results of all normal cycles in the health monitoring data curve. Since mean fitting is a well-known technology, it will not be described in detail in this embodiment; the fitting results of all normal cycles in the health monitoring data curve replace all normal cycles in the health monitoring data curve and all abnormal cycles that do not have long-term persistence characteristics to obtain a new health monitoring data curve, and finally run-length encoding is used to compress the data corresponding to the new health monitoring data curve.

[0068] It should be further explained that, in this embodiment, a new health monitoring data curve is obtained by performing mean fitting on the less important periods in the health monitoring data curve, so that the redundancy of the new health monitoring data curve is improved while retaining the important periods in the new health monitoring data curve, thereby improving the compression effect of the health monitoring data while ensuring that the compressed data can reflect the user's health status.

[0069] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A data processing method for a smart health system, characterized in that: The method comprises the following steps: Collect health monitoring data, which is HRV data; obtain each cycle in the health monitoring data curve, the data in each cycle in the health monitoring data, and the time span of each cycle of the health monitoring data based on the health monitoring data; obtain a first abnormality degree for each cycle in the health monitoring data curve based on the data in each cycle in the health monitoring data and the time span of each cycle of the health monitoring data, the first abnormality degree is used to characterize the abnormality degree of each cycle in the health monitoring data and all cycles in health data and time span; obtain a second abnormality degree for each cycle in the health monitoring data curve based on the inflection point in each cycle in the health monitoring data curve, the second abnormality degree is used to characterize the difference between the number of inflection points in each cycle in the health monitoring data curve and the number of inflection points in all cycles, and the difference between the sub-segment of each cycle and the sub-segments of all cycles; obtain the abnormality degree of each cycle in the health monitoring data curve based on the first abnormality degree and the second abnormality degree of each cycle in the health monitoring data curve; divide all cycles in the health monitoring data curve into two categories: abnormal cycles and normal cycles based on the abnormality degree of each cycle in the health monitoring data curve; Obtaining the time nodes of all abnormal cycles in the health monitoring data curve; obtaining abnormal cycles with long-term persistence characteristics and abnormal cycles without long-term persistence characteristics in the health monitoring data curve based on the time nodes of all abnormal cycles in the health monitoring data curve; According to the normal cycle in the health monitoring data curve, the abnormal cycle with long-term persistence characteristics in the health monitoring data curve, and the abnormal cycle without long-term persistence characteristics in the health monitoring data curve, a new health monitoring data curve is obtained and the new health monitoring data curve is compressed.

2. The data processing method for the smart health system according to claim 1, characterized in that: The collecting of health monitoring data; obtaining each cycle in the health monitoring data curve, the data within each cycle in the health monitoring data, and the time span of each cycle in the health monitoring data according to the health monitoring data, includes the following specific methods: Allow users to collect health monitoring data through smart bracelets while keeping their bodies relaxed. Through the health monitoring data, the collected health monitoring data is imported into Kubios HRV software to obtain the health monitoring data curve, each cycle in the health monitoring data curve, the data in each cycle of the health monitoring data, and the time span of each cycle of the health monitoring data.

3. The data processing method for the smart health system according to claim 1, characterized in that: The specific calculation formula for obtaining the first abnormality degree of each cycle in the health monitoring data curve is: Where, Indicates the first abnormality degree of the i-th cycle in the health monitoring data curve; F i Represents the difference between the maximum and minimum values ​​in the i-th cycle of the health monitoring data curve; N represents the number of cycles in the health monitoring data curve; F j Represents the difference between the maximum and minimum values ​​in the jth cycle of the health monitoring data curve; T i Represents the time span of the i-th cycle in the health monitoring data curve; T j It represents the time span of the jth cycle in the health monitoring data curve; Norm() represents the linear normalization function; exp() represents the exponential function with a natural constant as the base; || represents the absolute value operation.

4. The data processing method for a smart health system according to claim 1, characterized in that: The method of obtaining the second abnormality degree of each cycle in the health monitoring data curve according to the inflection point in each cycle in the health monitoring data curve includes the following specific methods: Each cycle of the health monitoring data curve is divided into several segments according to the inflection points in each cycle of the health monitoring data curve to obtain subsegments of each cycle of the health monitoring data curve. The second abnormality degree of each cycle of the health monitoring data curve is obtained according to the number of inflection points in each cycle of the health monitoring data curve and the cycle subsegments in the health monitoring data curve. The specific calculation formula is: Where, Indicates the second abnormality degree of the i-th cycle in the health monitoring data curve; M i represents the number of inflection points in the i-th cycle of the health monitoring data curve; N represents the number of cycles in the health monitoring data curve; M j K represents the number of inflection points in the jth cycle of the health monitoring data curve; di,i represents the mean value of the slope of all data on the d-th sub-segment in the ith cycle of the health monitoring data curve; D i represents the number of sub-segments in the i-th cycle of the health monitoring data curve; K dj,j represents the mean value of the slope of all data on the dj-th sub-segment in the j-th cycle of the health monitoring data curve; D j represents the number of sub-segments in the j-th period of the health monitoring data curve; exp() represents an exponential function with a natural constant as the base; A1 and A2 are preset weights; || represents an absolute value operation.

5. The data processing method for the smart health system according to claim 1, characterized in that: The specific calculation formula for obtaining the abnormality degree of each cycle in the health monitoring data curve is as follows: Where σ i Indicates the abnormality degree of the i-th cycle in the health monitoring data curve; Indicates the first abnormality degree of the i-th cycle in the health monitoring data curve; Indicates the second abnormality level of the i-th cycle in the health monitoring data curve; μ1 and μ2 are preset weights.

6. The data processing method for the smart health system according to claim 1, characterized in that: The method of dividing all cycles in the health monitoring data curve into abnormal cycles and normal cycles according to the abnormality degree of each cycle in the health monitoring data curve includes the following specific methods: An abnormality threshold τ is preset. When the abnormality of a period in the health monitoring data curve is greater than or equal to τ, the period in the health monitoring data curve is an abnormal period. When the abnormality of a period in the health monitoring data curve is less than τ, the period in the health monitoring data curve is a normal period.

7. The data processing method for a smart health system according to claim 1, characterized in that: The specific method of obtaining the time nodes of all abnormal cycles in the health monitoring data curve includes: For the oth abnormal cycle in the health monitoring data curve, the moment when the first data point in the oth abnormal cycle is recorded is selected as the time node of the oth abnormal cycle, and the time nodes of all abnormal cycles in the health monitoring data curve are obtained.

8. The data processing method for a smart health system according to claim 1, characterized in that: The specific method for obtaining abnormal periods with long-term persistence characteristics in the health monitoring data curve and abnormal periods without long-term persistence characteristics in the health monitoring data curve is as follows: For the oth abnormal cycle in the health monitoring data curve, the oth abnormal cycle is combined with each other abnormal cycle to obtain several abnormal cycle pairs of the oth abnormal cycle, and each abnormal cycle pair is combined with each other abnormal cycle pair to obtain several abnormal cycle pair groups of the oth abnormal cycle; the absolute value of the time node difference between the two abnormal cycles in each abnormal cycle pair is used as the time difference value of each abnormal cycle pair; the ratio of the small time difference value to the large time difference value of the two abnormal cycle pairs in the abnormal cycle pair group is used as the time difference similarity of each abnormal cycle pair group; finally, a threshold ξ is preset. When the ratio of the time difference similarity of any abnormal cycle pair group is greater than or equal to ξ, the oth abnormal cycle has the long-term persistence feature. When the ratio of the time difference similarity of all abnormal cycle pair groups is less than ξ, the oth abnormal cycle does not have the long-term persistence feature.

9. The data processing method for a smart health system according to claim 1, characterized in that: The specific method of obtaining a new health monitoring data curve and compressing the new health monitoring data curve is as follows: Perform mean fitting on all normal cycles in the health monitoring data curve to obtain fitting results of all normal cycles in the health monitoring data curve; use the fitting results of all normal cycles in the health monitoring data curve to replace all normal cycles in the health monitoring data curve and all abnormal cycles that do not have long-term persistence characteristics to obtain a new health monitoring data curve, and finally use run-length encoding to compress the data corresponding to the new health monitoring data curve.

Images