Postpartum mastitis risk intelligent early warning system and method based on internet of things

Abstract

The present application relates to the technical field of medical data mining, in particular to a postpartum mastitis risk intelligent early warning system and method based on the Internet of Things, which firstly quantifies risk assessment indexes and detection risk levels based on the overall size of each detection index at the moment of mastitis detection, and determines historical risk characteristic values in the dimension of static analysis based on the risk assessment indexes and detection risk levels at the leading detection moment representing the core state trend in the historical data; then determines time sequence risk characteristic values in the dimension of dynamic analysis according to the mastitis symptom changes reflected by the increasing situation of the detection risk level in the time sequence of the leading detection moment; further, the historical risk characteristic values and the time sequence risk characteristic values are determined on the basis of the risk assessment indexes, so as to more accurately determine the mastitis risk alarm value and make the postpartum mastitis risk early warning effect better according to the mastitis risk alarm value.

Description

Technical Field

[0001] This invention relates to the field of medical data mining technology, specifically to an intelligent early warning system and method for postpartum mastitis risk based on the Internet of Things. Background Technology

[0002] Postpartum mastitis is generally caused by milk stasis, bacterial infection, and decreased immunity. The occurrence of mastitis is often accompanied by multiple indicators, such as body temperature, CRP level, and lump size. Therefore, current technology usually extracts the overall value of each indicator from the most recent mastitis test on an IoT cloud storage platform to quantify the mastitis risk alarm value, and then provides a postpartum mastitis risk warning based on the mastitis risk alarm value.

[0003] However, relying solely on the overall magnitude of various test indicators from the most recent mastitis test cannot reflect the dynamic evolution of the relevant indicators, resulting in a high limitation in the obtained mastitis risk alarm value. This leads to low accuracy of the mastitis risk alarm value obtained by the current technology based solely on the overall magnitude of various test indicators from the most recent mastitis test, making the postpartum mastitis risk warning based on the mastitis risk alarm value ineffective. Summary of the Invention

[0004] To address the issue of low accuracy in existing technologies that rely solely on the overall magnitude of various indicators from the most recent mastitis test for mastitis risk alarm values, this application aims to provide an intelligent early warning system and method for postpartum mastitis risk based on the Internet of Things (IoT). The specific technical solution adopted is as follows:

[0005] The first aspect of this application provides an intelligent early warning method for postpartum mastitis risk based on the Internet of Things, including:

[0006] In the IoT cloud storage platform, the quantitative values ​​of each detection indicator at each mastitis detection time are collected from the postpartum historical data of the early warning users; based on the overall magnitude of the quantitative values ​​of each detection indicator at each mastitis detection time, the corresponding risk assessment index is determined; and based on the risk assessment index, the detection risk level at each mastitis detection time is determined.

[0007] Based on the temporal distribution of detection risk levels, the dominant detection time is determined; based on the frequency distribution of detection risk levels at the dominant detection time and the corresponding risk assessment index, historical risk characteristic values ​​are determined; based on the significant upward trend of the detection risk level at the dominant detection time in a local temporal position, temporal risk characteristic values ​​are determined.

[0008] Based on the distribution of the historical risk characteristic values, the time-series risk characteristic values, and the risk assessment index, a mastitis risk alarm value is determined; and a postpartum mastitis risk warning is issued based on the mastitis risk alarm value.

[0009] The process of obtaining the dominant detection time includes:

[0010] The frequency of occurrence of each risk level at all mastitis detection times is used as the dominant feature value for each risk level.

[0011] The detection risk level with the largest dominant feature value is taken as the dominant risk level; the detection time of mastitis with the detection risk level being the dominant risk level is taken as the dominant detection time.

[0012] Furthermore, the process of obtaining the risk assessment index includes:

[0013] The corresponding risk assessment index is determined based on the average of the quantitative values ​​of all test indicators at each mastitis test time.

[0014] Furthermore, the process of obtaining the detection risk level includes:

[0015] K-means cluster analysis was performed on the risk assessment indices at all mastitis detection times to obtain at least two clusters of assessment indices;

[0016] The mean of all risk assessment indices in each assessment index cluster is used as the corresponding cluster feature value.

[0017] Arrange all evaluation index clusters in ascending order of cluster feature values ​​to obtain the evaluation index cluster sequence;

[0018] The index value of the assessment index cluster in the assessment index cluster sequence at each mastitis detection time is used as the detection risk level at each mastitis detection time.

[0019] Furthermore, the process of obtaining the historical risk characteristic values ​​includes:

[0020] A reference dominant frequency value is determined based on the ratio between the number of dominant detection times and the number of all mastitis detection times in the postpartum historical data of the warning users; a historical risk characteristic value is determined based on the product of the maximum risk assessment index of all dominant detection times, the reference dominant frequency value, and the dominant risk level.

[0021] Furthermore, the process of obtaining the time-series risk feature values ​​includes:

[0022] The difference between the detection risk level at each dominant detection time and the detection risk level at the next mastitis detection time is used to determine the reference judgment value for each dominant detection time.

[0023] The dominant detection time when the reference judgment value is less than or equal to 0 is taken as the reference risk time; a reference risk time period is obtained, in which the mastitis detection times are continuous and all are reference risk times, and the mastitis detection time before and after the reference risk time period is not a reference risk time.

[0024] Based on the local upward trend of the risk assessment index at the reference risk moment within the reference risk period, the corresponding risk growth characteristic value is determined;

[0025] The number of mastitis detection times between each reference risk time period and the previous reference risk time period is used as the first reference number; the number of mastitis detection times between each reference risk time period and the next reference risk time period is used as the second reference number; the mean values ​​of the first and second reference numbers are negatively correlated to determine the reference interval characteristic value.

[0026] The corresponding local risk characteristic value is determined by multiplying the maximum risk assessment index of all reference risk moments in each reference risk time period, the risk growth characteristic value, the reference interval characteristic value, and the number of reference risk moments; the time-series risk characteristic value is determined by the mean of the local risk characteristic values ​​of all reference risk time periods.

[0027] Furthermore, the process of obtaining the risk growth characteristic value includes:

[0028] The risk assessment indices of all reference risk moments in each reference risk period are arranged in chronological order and then subjected to curve fitting to determine the corresponding risk assessment index curve. On the risk assessment index curve, the mean of the slope of the tangent line for all reference risk moments is normalized to determine the risk growth characteristic value.

[0029] Furthermore, the process of obtaining the mastitis risk alarm value includes:

[0030] The recent risk characteristic value is determined based on the average of the risk assessment index of a preset number of mastitis detection times prior to the current moment;

[0031] The product of the historical risk feature value, the recent risk feature value, and the time-series risk feature value is normalized to determine the mastitis risk alarm value.

[0032] Furthermore, the process of providing postpartum mastitis risk warning based on the mastitis risk alarm value includes:

[0033] When the mastitis risk alarm value is greater than the preset alarm threshold, a high-risk warning signal for postpartum mastitis is issued.

[0034] Secondly, this application provides an intelligent early warning system for postpartum mastitis risk based on the Internet of Things, the system comprising:

[0035] The data acquisition and preprocessing module is used to collect the quantitative values ​​of each detection indicator at each mastitis detection time from the postpartum historical data of early warning users in the IoT cloud storage platform; determine the corresponding risk assessment index based on the overall magnitude of the quantitative values ​​of each detection indicator at each mastitis detection time; and determine the detection risk level at each mastitis detection time based on the risk assessment index.

[0036] The feature value determination module is used to determine the dominant detection time based on the temporal distribution of detection risk levels; determine historical risk feature values ​​based on the frequency distribution of detection risk levels at the dominant detection time and the corresponding risk assessment index values; and determine temporal risk feature values ​​based on the significant upward trend of the detection risk level at a local temporal position of the dominant detection time.

[0037] The process of obtaining the dominant detection time includes:

[0038] The frequency of occurrence of each risk level at all mastitis detection times is used as the dominant feature value for each risk level.

[0039] The detection risk level with the largest dominant feature value is taken as the dominant risk level; the detection time of mastitis with the detection risk level being the dominant risk level is taken as the dominant detection time.

[0040] The mastitis risk warning module is used to determine the mastitis risk alarm value based on the distribution of the historical risk characteristic value, the time-series risk characteristic value, and the risk assessment index; and to provide a postpartum mastitis risk warning based on the mastitis risk alarm value.

[0041] Thirdly, this application provides a computer device including a memory and a processor. The memory is used to store computer program code, and the processor is used to call and run the computer program code from the memory to perform the method as described in the first aspect of this application or any embodiment of the first aspect.

[0042] Fourthly, this application provides a computer program product comprising computer program code, which, when executed, performs the method as described in the first aspect of this application or any embodiment thereof.

[0043] Fifthly, this application provides a computer-readable storage medium that stores computer program code, which, when executed, performs the method as described in the first aspect of this application or any embodiment thereof.

[0044] This application has the following beneficial effects:

[0045] This application quantifies the risk assessment index and detection risk level based on the overall magnitude of various detection indicators at the time of mastitis detection. Based on the risk assessment index and detection risk level at the dominant detection time, which characterize the core state trend in historical data, historical risk characteristic values ​​are first determined in a static analysis dimension. Then, based on the changes in mastitis symptoms reflected by the increasing detection risk level at the dominant detection time over time, time-series risk characteristic values ​​are determined in a dynamic analysis dimension. Furthermore, based on the risk assessment index, the historical risk characteristic values ​​and time-series risk characteristic values ​​are combined to more accurately determine the mastitis risk alarm value, resulting in a better effect of postpartum mastitis risk warning based on the mastitis risk alarm value. Attached Figure Description

[0046] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0047] Figure 1 A flowchart illustrating an intelligent early warning method for postpartum mastitis risk based on the Internet of Things, provided as an embodiment of the present invention;

[0048] Figure 2 This is a structural diagram of an intelligent early warning system for postpartum mastitis risk based on the Internet of Things, provided in one embodiment of the present invention.

[0049] Figure 3 This is a schematic diagram of a computer device structure provided in one embodiment of the present invention. Detailed Implementation

[0050] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of the IoT-based intelligent early warning system and method for postpartum mastitis risk proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment, and specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form. Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as implying or suggesting relative importance or implicitly indicating the number of indicated technical features. Thus, a feature defined with "first" or "second" may explicitly or implicitly include one or more of that feature.

[0051] Unless otherwise defined, 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 pertains.

[0052] The following description, in conjunction with the accompanying drawings, details the specific scheme of the intelligent early warning system and method for postpartum mastitis risk based on the Internet of Things provided by this invention.

[0053] This application provides an intelligent early warning method for postpartum mastitis risk based on the Internet of Things. Please refer to [link to relevant documentation]. Figure 1 The diagram illustrates a flowchart of an intelligent early warning method for postpartum mastitis risk based on the Internet of Things, according to an embodiment of the present invention. The method includes:

[0054] Step S101: In the IoT cloud storage platform, collect the quantitative values ​​of each detection indicator at each mastitis detection time from the postpartum historical data of the early warning user; determine the corresponding risk assessment index based on the overall magnitude of the quantitative values ​​of each detection indicator at each mastitis detection time; determine the detection risk level at each mastitis detection time based on the risk assessment index.

[0055] In an IoT cloud storage platform storing medical data, historical data on postpartum statistics of users receiving early warnings are collected, including data values ​​of various test indicators at each mastitis detection time. In a specific implementation of this invention, the test indicators include: mastitis lump size, body temperature, CRP value, and white blood cell density, which can be adjusted according to the specific implementation environment. Specifically, the size of the mastitis lump at each detection time is measured using a soft ruler, and the longest diameter of the lump measured by the soft ruler is taken as its corresponding size. The body temperature at each detection time is measured using a thermometer. The CRP value at each detection time is determined using the latex agglutination method. The white blood cell density at each detection time is measured using a blood analyzer. Since the dimensions of the data of different test indicators are different, further quantitative analysis is required. Among these indicators, a body temperature above 37 degrees Celsius usually indicates a possible inflammatory response; therefore, the higher the temperature, the larger the quantitative value of the indicator. CRP (Cytotoxic Reproductive Rate) is a sensitive indicator of acute inflammation; the normal range postpartum is usually less than or equal to 10 mg / L, so the higher the CRP value, the larger the quantitative value of the indicator. Similarly, when there are symptoms of infection, white blood cell density will significantly increase; under normal circumstances, the white blood cell density of healthy individuals is usually less than 10 × 10⁻⁶. 9 / L corresponds to a higher white blood cell density, so the quantitative value of the indicator should also be larger; while the size of the mastitis lump only appears when mastitis symptoms occur, and the larger the size of the mastitis lump, the more obvious the symptoms, so the quantitative value of the indicator should be larger.

[0056] In one specific implementation of this invention, the values ​​of each detection indicator at each mastitis detection time in the postpartum historical data of the early warning user are linearly normalized to determine the corresponding indicator quantification value. This linear normalization method avoids the influence of the dimensions of different types of detection indicators on the subsequent calculation process, making the subsequent analysis process more robust. Furthermore, this embodiment sets the data sampling period to one week prior to the current time, and the mastitis detection frequency is once every 6 hours, meaning the time interval between adjacent mastitis detection times is 6 hours, which can be adjusted according to the specific implementation environment. It should be noted that the data in this embodiment belongs to the same early warning user and is all data authorized for use by the early warning user; and unless otherwise specified, the normalization method in this embodiment is linear normalization, which will not be further elaborated upon below.

[0057] After determining the quantified values ​​of all detection indicators, a comprehensive risk assessment index can be determined by combining the quantified values ​​of all indicators at each mastitis detection time for subsequent risk grading. Preferably, in some possible implementations of this invention, the process of obtaining the risk assessment index includes: determining the corresponding risk assessment index based on the average of the quantified values ​​of all detection indicators at each mastitis detection time; the higher the risk assessment index, the higher the risk of postpartum mastitis at the corresponding detection time. After determining the risk assessment index, it is further necessary to classify the detection risk levels to provide a data foundation for subsequent analysis.

[0058] Preferably, in some possible implementations of the embodiments of the present invention, the process of obtaining the detection risk level includes:

[0059] A k-means clustering analysis was performed on the risk assessment indices at all mastitis detection times to obtain at least two clusters of assessment indices. In one specific implementation of this invention, the K value in the k-means clustering analysis is set to 4, which can be adjusted according to the specific implementation environment.

[0060] The mean of all risk assessment indices in each assessment index cluster is used as the corresponding cluster feature value. All assessment index clusters are arranged in ascending order of their feature values ​​to obtain an assessment index cluster sequence. Since the cluster analysis focuses on the magnitude of the risk assessment index, for each assessment index cluster, the larger its index value in the sequence of clusters arranged in ascending order of feature values, the larger the risk assessment index within that cluster, and therefore the higher the detection risk level. Therefore, the index value of the assessment index cluster at each mastitis detection time within the assessment index cluster sequence is further used as the detection risk level at each mastitis detection time. A higher detection risk level indicates a higher risk of postpartum mastitis at the corresponding detection time.

[0061] Step S102: Determine the dominant detection time based on the temporal distribution of detection risk levels; determine the historical risk characteristic value based on the frequency distribution of detection risk levels at the dominant detection time and the corresponding risk assessment index value; determine the temporal risk characteristic value based on the significant upward trend of the detection risk level at the local temporal position of the dominant detection time.

[0062] Among all mastitis detection times, the most frequently occurring risk level reflects the main trend of postpartum mastitis status in alerted users. Analysis based on this risk level allows for a more accurate assessment of the mastitis risk status of alerted users. In a specific implementation of this invention, the process of obtaining the dominant detection time includes:

[0063] The frequency of occurrence of each risk level in all mastitis detection times is used as the dominant feature value for each risk level; the risk level with the largest dominant feature value is used as the dominant risk level; the mastitis detection time with the dominant risk level is used as the dominant detection time; that is, the mastitis detection time corresponding to the most frequently occurring risk level is used as the dominant detection time.

[0064] After determining the dominant detection time, the overall mastitis risk is analyzed based on the trend of the main postpartum mastitis status of the warning users as reflected by the dominant risk level, through the overall mastitis risk reflected by the corresponding detection risk level and risk assessment index, to determine a more accurate historical risk characteristic value. That is, the historical risk characteristic value is determined based on the frequency distribution of the detection risk level at the dominant detection time and the value of the corresponding risk assessment index. The larger the historical risk characteristic value, the more obvious the mastitis characteristics are in the static analysis dimension, and the larger the corresponding mastitis risk alarm value should be.

[0065] Preferably, in some possible implementations of the embodiments of the present invention, the process of obtaining historical risk feature values ​​includes:

[0066] The more dominant detection times there are, the higher the reliability of the trend of the main state of mastitis reflected by the corresponding dominant risk level. Therefore, a reference dominant frequency value is determined based on the ratio between the number of dominant detection times and the number of all mastitis detection times in the postpartum historical data of the warning users. The larger the reference dominant frequency value, the higher the reliability of the corresponding dominant risk level in reflecting the risk characteristics of mastitis. Furthermore, using the number of all mastitis detection times in the postpartum historical data of the warning users as the denominator can avoid the influence of dimensions in different implementation environments and improve the robustness of the calculation process. The historical risk characteristic value can be obtained by weighting the dominant risk level with the reference dominant frequency value. In addition, considering that the risk assessment index can also characterize the risk of postpartum mastitis, the historical risk characteristic value is further determined by the product of the maximum value of the risk assessment index of all dominant detection times, the reference dominant frequency value, and the dominant risk level.

[0067] In one specific implementation of this invention, the process of obtaining historical risk characteristic values ​​is expressed by the following formula: ;in, Historical risk characteristic values; The number of dominant detection moments; To provide early warning of the number of all mastitis detection times in the user's postpartum historical data; The maximum value of the risk assessment index at all dominant detection moments; The numerical value of the dominant risk level. In a specific implementation of this invention, the number of all mastitis detection times in the postpartum historical data of the user triggering the warning is used. When the value is 0, the positive number 1 is used to replace the denominator in the calculation to avoid the situation where the formula is meaningless due to the denominator being 0.

[0068] Historical risk characteristic values ​​characterize mastitis risk in a static analysis dimension by measuring the size of historical data. In a dynamic time-series dimension, changes in the detection risk level can reflect the development and changes in mastitis characteristics, thus indirectly characterizing mastitis risk. Therefore, in a dynamic analysis dimension, the temporal risk characteristic value is determined based on the significant upward trend of the detection risk level at the dominant detection time in the local temporal sequence. The larger the temporal risk characteristic value, the more obvious the mastitis characteristics are in the dynamic analysis dimension, and the higher the corresponding mastitis risk alarm value should be.

[0069] Preferably, in a specific implementation of this invention, the process of obtaining the time-series risk feature value includes:

[0070] The difference between the detection risk level at each dominant detection time and the detection risk level at the next mastitis detection time is used to determine the reference judgment value for each dominant detection time. Dominant detection times with a reference judgment value less than or equal to 0 are designated as reference risk times. For each dominant detection time, a reference judgment value less than 0 indicates that the detection risk level at the next mastitis detection time has increased, and mastitis symptoms have worsened; while a reference judgment value equal to 0 indicates that the detection risk level at the next mastitis detection time has not changed, and mastitis symptoms have not been relieved. Therefore, the reference risk time is determined based on whether the reference judgment value is less than or equal to 0.

[0071] A reference risk time period is obtained, in which mastitis detection times are continuous and all are reference risk times. The preceding and following mastitis detection times within a reference risk time period are not reference risk times. By merging continuously distributed reference risk times, a more accurate mastitis risk analysis can be performed using temporal variations within each reference risk time period.

[0072] Since the risk assessment index can characterize the risk features of mastitis to a certain extent, the corresponding risk growth characteristic value is further determined based on the local upward trend of the risk assessment index at each reference risk time within the reference risk period. In a specific implementation of this invention, the process of obtaining the risk growth characteristic value includes: arranging the risk assessment indices of all reference risk times within each reference risk period in chronological order and performing curve fitting to determine the corresponding risk assessment index curve; and normalizing the mean of the tangent slopes at all reference risk times on the risk assessment index curve to determine the risk growth characteristic value. For each reference risk period, the larger the mean of the tangent slopes at all reference risk times, the greater the overall upward trend of the mastitis risk index within that reference risk period; therefore, the larger the risk growth characteristic value, the more pronounced the corresponding mastitis deterioration characteristics.

[0073] The number of mastitis detection times between each reference risk time period and the previous reference risk time period is used as the first reference number; the number of mastitis detection times between each reference risk time period and the next reference risk time period is used as the second reference number. A negative correlation is established between the mean of the first and second reference numbers to determine the reference interval characteristic value. That is, for each reference risk time period, the smaller the time interval between it and two adjacent reference risk time periods, the larger the reference interval characteristic value; conversely, a smaller interval between adjacent reference risk time periods indicates a shorter interval between symptom relief and subsequent worsening of mastitis, and a more pronounced characteristic of mastitis deterioration. Based on the principle of obtaining the reference interval characteristic, the longer the duration of each reference risk time period, i.e., the more reference risk times, the longer the duration for which the detection risk level does not decrease, and the greater the corresponding risk of mastitis.

[0074] Within each reference risk time period, a higher maximum risk assessment index indicates a greater risk of postpartum mastitis. Therefore, by further integrating the maximum risk assessment index, risk growth characteristic value, reference interval characteristic value, and the number of reference risk moments across all reference risk times within each reference risk time period, local risk characteristic values ​​are characterized. By combining the local risk characteristic values ​​across all reference risk time periods, the final determination of the temporal risk characteristic value is made. This ensures that a higher temporal risk characteristic value indicates more pronounced mastitis characteristics and a greater corresponding risk, thus requiring a higher corresponding mastitis risk alarm value.

[0075] In one specific implementation of this invention, a local risk characteristic value is determined based on the product of the maximum risk assessment index, risk growth characteristic value, reference interval characteristic value, and the number of reference risk moments for each reference risk time period; a time-series risk characteristic value is determined based on the mean of the local risk characteristic values ​​for all reference risk time periods. The process of obtaining the time-series risk characteristic value is expressed by the following formula: ;in, This represents the time-series risk characteristic value; For reference, the number of risk periods; For the first The maximum risk assessment index value for all reference risk moments within a reference risk time period; For the first Risk growth characteristic values ​​for a reference risk period; For the first The first reference quantity for a reference risk time period; For the first The second reference quantity for a reference risk time period; It is an exponential function with the natural constant as its base; For the first The reference interval characteristic value for each reference risk time period; For the first The number of reference risk moments within each reference risk time period. It should be noted that, to avoid a denominator of 0, when the number of reference risk time periods is 0, the time-series risk characteristic value is directly set to 0. A value of 0 for the number of reference risk time periods indicates that there are no reference risk time periods, suggesting that no obvious and continuous trend of worsening mastitis risk was detected in historical data; therefore, the time-series risk characteristic value is set to 0 for subsequent analysis.

[0076] Step S103: Determine the mastitis risk alarm value based on the distribution of historical risk characteristic values, time-series risk characteristic values, and risk assessment index; issue a postpartum mastitis risk warning based on the mastitis risk alarm value.

[0077] Both historical risk characteristic values ​​and time-series risk characteristic values ​​can clearly characterize the features of mastitis. Furthermore, the larger the historical risk characteristic value and the larger the time-series risk characteristic value, the higher the mastitis risk alarm value should be. In addition, it is necessary to consider that the risk assessment index can also characterize mastitis risk characteristics. The risk assessment index closer to the current moment has higher reliability. Therefore, further comprehensive characterization of the mastitis risk alarm value is achieved based on the distribution of historical risk characteristic values, time-series risk characteristic values, and risk assessment index, resulting in a more accurate mastitis risk alarm value.

[0078] Preferably, in some possible implementations of the embodiments of the present invention, the process of obtaining the mastitis risk alarm value includes:

[0079] Based on the average of the risk assessment indices of a preset number of mastitis detection times prior to the current moment, a recent risk characteristic value is determined. The product of the historical risk characteristic value, the recent risk characteristic value, and the time-series risk characteristic value is normalized to determine the mastitis risk alarm value. In one specific implementation of this invention, the preset number is set to 4, which can be adjusted according to the specific implementation environment. The larger the recent risk characteristic value, the larger the overall risk assessment index of the recent mastitis detection times, and the greater the represented mastitis risk; therefore, a larger mastitis risk alarm value needs to be assigned. Furthermore, the normalization method here limits the value of the mastitis risk alarm value to between 0 and 1, facilitating more detailed postpartum mastitis risk warnings. Specifically, the process of issuing a postpartum mastitis risk warning based on the mastitis risk alarm value includes: when the mastitis risk alarm value is greater than a preset alarm threshold, a high-risk postpartum mastitis warning signal is issued. In one specific implementation of this invention, the preset alarm threshold is set to 0.6, which can be adjusted according to the specific implementation environment and will not be further elaborated here.

[0080] In summary, an IoT-based intelligent early warning method for postpartum mastitis risk quantifies the risk assessment index and detection risk level based on the overall magnitude of various detection indicators at the time of mastitis detection. It first determines historical risk characteristic values ​​in a static analysis dimension based on the dominant detection time's risk assessment index and detection risk level, which characterize the core state trend in historical data. Then, based on the temporal changes in mastitis symptoms reflected by the increasing detection risk level at the dominant detection time, it determines temporal risk characteristic values ​​in a dynamic analysis dimension. Finally, based on the risk assessment index, it combines historical and temporal risk characteristic values ​​to more accurately determine the mastitis risk alarm value, resulting in a better effect of postpartum mastitis risk early warning based on the mastitis risk alarm value.

[0081] This application also provides an intelligent early warning system for postpartum mastitis risk based on the Internet of Things (IoT). Please refer to [link / reference]. Figure 2 The diagram shows a structural diagram of an intelligent early warning system for postpartum mastitis risk based on the Internet of Things provided in an embodiment of the present invention. The system includes: a data acquisition and preprocessing module 201, a feature value determination module 202, and a mastitis risk early warning module 203.

[0082] The data acquisition and preprocessing module 201 is used to collect the quantitative values ​​of each detection indicator at each mastitis detection time in the postpartum historical data of early warning users in the Internet of Things cloud storage platform; determine the corresponding risk assessment index based on the overall magnitude of the quantitative values ​​of each detection indicator at each mastitis detection time; and determine the detection risk level at each mastitis detection time based on the risk assessment index.

[0083] The feature value determination module 202 is used to determine the dominant detection time based on the temporal distribution of the detection risk level; determine the historical risk feature value based on the frequency distribution of the detection risk level at the dominant detection time and the corresponding risk assessment index value; and determine the temporal risk feature value based on the significant upward trend of the detection risk level at the local temporal position of the dominant detection time in the time sequence.

[0084] The mastitis risk warning module 203 is used to determine the mastitis risk alarm value based on the distribution of historical risk characteristic values, time-series risk characteristic values, and risk assessment index; and to provide postpartum mastitis risk warning based on the mastitis risk alarm value.

[0085] It should be noted that the system provided in the above embodiments is only an example of the division of the above functional modules. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the computer device can be divided into different functional modules to complete all or part of the functions described above. In addition, the intelligent early warning system for postpartum mastitis risk based on the Internet of Things and the intelligent early warning method for postpartum mastitis risk based on the Internet of Things provided in the above embodiments belong to the same concept. The specific implementation process is detailed in the method embodiment and will not be repeated here.

[0086] This application also provides a computer device; please refer to [link / reference]. Figure 3 The illustration shows a schematic diagram of a computer device structure provided by an embodiment of the present invention. The computer device includes a memory 301, a processor 302, and a computer program 303 stored in the memory 301 and running on the processor 302. When the processor 302 executes the computer program 303, the computer device can execute any of the aforementioned Internet of Things-based intelligent early warning methods for postpartum mastitis risk.

[0087] This application also provides a computer program product that, when run on a computer device, enables the computer device to execute any of the aforementioned IoT-based intelligent early warning methods for postpartum mastitis risk.

[0088] This application also provides a computer-readable storage medium storing computer program code. When the computer program code is run on a computer device, the computer device can execute any of the aforementioned IoT-based intelligent early warning methods for postpartum mastitis risk.

[0089] In the embodiments provided in this application, it should be understood that the computer device, computer program product and computer-readable storage medium provided are all used to perform the corresponding methods provided above, and therefore the beneficial effects they can achieve can be referred to the beneficial effects of the methods provided above, which will not be repeated here.

[0090] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0091] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. An Internet of Things-based intelligent warning method for postpartum mastitis risk, characterized in that, The method comprises: In the Internet of Things cloud storage platform, the index quantitative value of each detection index of each mastitis detection moment in the early warning user postpartum historical data is collected; the risk assessment index corresponding to each mastitis detection moment is determined according to the overall size of the index quantitative value of each detection index of each mastitis detection moment; and the detection risk level of each mastitis detection moment is determined according to the risk assessment index. The dominant detection moment is determined according to the centralized situation of the time sequence distribution of the detection risk level; the historical risk characteristic value is determined according to the frequency numerical distribution of the detection risk level of the dominant detection moment and the numerical size of the corresponding risk assessment index; and the time sequence risk characteristic value is determined according to the significant situation of the rising trend of the detection risk level of the dominant detection moment in the local time sequence position in the time sequence. The mastitis risk alarm value is determined according to the distribution of the historical risk characteristic value, the time sequence risk characteristic value and the risk assessment index; and the postpartum mastitis risk early warning is performed according to the mastitis risk alarm value. The acquisition process of the dominant detection moment comprises: The occurrence number of each detection risk level in all mastitis detection moments is taken as the dominant characteristic value of each detection risk level. The detection risk level with the maximum dominant characteristic value is taken as the dominant risk level; and the mastitis detection moment with the detection risk level being the dominant risk level is taken as the dominant detection moment. The acquisition process of the detection risk level comprises: The k-means clustering analysis is performed on the risk assessment indexes of all mastitis detection moments to obtain at least two assessment index clustering clusters. The mean value of all risk assessment indexes in each assessment index clustering cluster is taken as the characteristic value of the corresponding clustering cluster. All assessment index clustering clusters are arranged in the order of the characteristic values from small to large to obtain an assessment index clustering cluster sequence. The index value of the assessment index clustering cluster in which the risk assessment index of each mastitis detection moment is located in the assessment index clustering cluster sequence is taken as the detection risk level of each mastitis detection moment. The acquisition process of the time sequence risk characteristic value comprises: The reference judgment value of each dominant detection moment is determined according to the difference between the detection risk level of each dominant detection moment and the detection risk level of the next mastitis detection moment. The dominant detection moment with the reference judgment value being less than or equal to 0 is taken as the reference risk moment; the reference risk time period is obtained, the mastitis detection moments in the reference risk time period are continuous and are all reference risk moments, and the previous mastitis detection moment and the next mastitis detection moment of the reference risk time period are not reference risk moments. The risk growth characteristic value corresponding to the reference risk moment in the reference risk time period is determined according to the local rising trend of the risk assessment index. The first reference number is determined according to the number of mastitis detection moments between each reference risk time period and the previous reference risk time period; the second reference number is determined according to the number of mastitis detection moments between each reference risk time period and the next reference risk time period; and the reference interval characteristic value is determined by negatively correlating the mean value between the first reference number and the second reference number. Determine a corresponding local risk characteristic value according to the product among the maximum value of risk assessment indexes of all reference risk moments in each reference risk time period, the risk growth characteristic value, the reference interval characteristic value, and the number of reference risk moments; and determine a time sequence risk characteristic value according to the average value of local risk characteristic values of all reference risk time periods. 2.The postpartum mastitis risk intelligent early warning method based on the Internet of Things according to claim 1, characterized in that, The risk assessment index acquisition process comprises: Determine a corresponding risk assessment index according to the average value of index quantization values of all detection indexes at each mastitis detection moment. 3.The postpartum mastitis risk intelligent early warning method based on the Internet of Things according to claim 1, characterized in that, The historical risk characteristic value acquisition process comprises: Determine a reference dominant frequency value according to the ratio between the number of dominant detection moments and the number of all mastitis detection moments in the postpartum historical data of the early warning user; and determine a historical risk characteristic value according to the product among the maximum value of risk assessment indexes of all dominant detection moments, the reference dominant frequency value, and the product among the reference dominant frequency value and the dominant risk level.

4. The postpartum mastitis risk intelligent early warning method based on the Internet of Things according to claim 1, characterized in that, The risk growth characteristic value acquisition process comprises: Arrange the risk assessment indexes of all reference risk moments in each reference risk time period in time sequence, perform curve fitting, and determine a corresponding risk assessment index curve; and normalize the average value of tangent slopes of all reference risk moments on the risk assessment index curve, and determine a risk growth characteristic value.

5. The postpartum mastitis risk intelligent early warning method based on the Internet of Things according to claim 1, characterized in that, The mastitis risk alarm value acquisition process comprises: Determine a recent risk characteristic value according to the average value of risk assessment indexes of four mastitis detection moments before the current moment; Normalize the product among the historical risk characteristic value, the recent risk characteristic value, and the time sequence risk characteristic value, and determine a mastitis risk alarm value. 6.The postpartum mastitis risk intelligent early warning method based on the Internet of Things according to claim 1, characterized in that, The postpartum mastitis risk early warning process according to the mastitis risk alarm value comprises: When the mastitis risk alarm value is greater than a preset alarm threshold, issue a high-risk postpartum mastitis early warning signal.

7. An Internet of Things-based intelligent warning system for postpartum mastitis risk, characterized in that, The system is used to implement the postpartum mastitis risk intelligent early warning method based on the Internet of Things, and comprises: A data acquisition and preprocessing module is used to acquire index quantization values of each detection index at each mastitis detection moment in the postpartum historical data of the early warning user in the Internet of Things cloud storage platform; determine a corresponding risk assessment index according to the overall size of index quantization values of each detection index at each mastitis detection moment; and determine a detection risk level of each mastitis detection moment according to the risk assessment index; A characteristic value determination module is used to determine a dominant detection moment according to the time sequence distribution of detection risk levels; determine a historical risk characteristic value according to the frequency value distribution of detection risk levels of the dominant detection moment and the numerical size of the corresponding risk assessment index; and determine a time sequence risk characteristic value according to the significant change rising trend of detection risk levels of the dominant detection moment in the local time sequence position in time sequence. The dominant detection moment acquisition process comprises: The number of appearances of each detection risk level in all mastitis detection moments is taken as a dominant characteristic value of each detection risk level. The detection risk level with the largest principal eigenvalue is taken as a principal risk level, and a mastitis detection time with the detection risk level being the principal risk level is taken as a principal detection time; The mastitis risk early warning module is configured to determine a mastitis risk alarm value according to the historical risk eigenvalue, the time-series risk eigenvalue, and a distribution of the risk assessment index, and perform postpartum mastitis risk early warning according to the mastitis risk alarm value.

Images