Pregnant and lying-in woman pregnancy risk assessment method based on previous electronic medical history

Abstract

The invention relates to the technical field of medical data processing, in particular to a pregnant and lying-in woman pregnancy risk assessment method based on past electronic medical history. According to the method, for any important physical sign parameter, the detection state anomaly degree of a local range at each moment is obtained; in combination with the time difference between the adjacent moments, obtaining a detection state change index between the adjacent moments; obtaining a normal time period and an initial abnormal time period to carry out time feature analysis, obtaining an abnormal degree state coefficient of each initial abnormal time period in combination with a corresponding detection state change index in the initial abnormal time period, and screening out an actual abnormal time period; and analyzing detection data of each important sign parameter in a neighborhood range of a real-time moment or a neighborhood range of a next moment, obtaining an abnormal degree state coefficient corresponding to each actual abnormal time period, and screening out a high abnormal time period. The accuracy of pregnancy risk assessment is improved by obtaining important sign parameters and accurate abnormal state distribution.

Description

Technical Field

[0001] The present invention relates to the technical field of medical data processing, and particularly relates to a method for assessing the pregnancy risk of pregnant women based on past electronic medical records. Background Art

[0002] Pregnancy is a crucial stage in a woman's physiological process. During pregnancy, significant changes occur in the mother's physiological functions, and a series of pregnancy complications may occur, such as hypertension, posing potential risks to the health of the mother and baby. Therefore, it is necessary to collect more accurate and comprehensive data from past electronic medical records to assess the pregnancy risk of pregnant women.

[0003] In the prior art, pregnancy risk assessment methods usually rely on basic physical sign data and medical history in past electronic medical records, and make risk judgments based on empirical knowledge. However, due to individual differences among pregnant women, as well as complex environmental and lifestyle factors, when analyzing the complex data in the medical history, it is easily affected by different non-essential physical sign parameters, resulting in a decrease in the efficiency and poor accuracy of the assessment of the pregnancy risk of pregnant women. Summary of the Invention

[0004] In order to solve the technical problem that the efficiency of assessing the pregnancy risk of pregnant women is reduced and the accuracy is poor due to being affected by different non-essential physical sign parameters, the purpose of the present invention is to provide a method for assessing the pregnancy risk of pregnant women based on past electronic medical records, and the specific technical solution adopted is as follows: The present invention proposes a method for assessing the pregnancy risk of pregnant women based on past electronic medical records, and the method includes: Obtaining the detection data of multiple physical sign parameters of a pregnant woman in different examination items according to time sequence; For different examination items, according to the distribution of the detection data of each physical sign parameter at different times, obtaining the abnormal coefficient of each physical sign parameter at each moment, and screening out important physical sign parameters; For any important physical sign parameter, obtaining the fitting curve of the corresponding detection data at all times, and according to the distribution of the fitting data in the local range of each moment on the fitting curve, and the abnormal coefficient corresponding to each moment, obtaining the detection state abnormality degree of the local range of each moment; according to the time difference between adjacent moments, and the detection state abnormality degree of the corresponding local range of the moment, obtaining the detection state change index between adjacent moments; According to the change trend of the detection state change indexes between all adjacent moments, obtaining the normal time period and the initial abnormal time period; according to the time distribution characteristics between each initial abnormal time period and the adjacent normal time period, and the detection state change index corresponding to the initial abnormal time period, obtaining the abnormal degree state coefficient of each initial abnormal time period, and screening out the actual abnormal time period; Obtain the detection data for the next moment based on the detection data at the real-time moment and the corresponding detection status change indicators; analyze the detection data within the neighborhood range of the real-time moment or the neighborhood range of the next moment for each vital sign parameter to obtain the abnormal degree status coefficient for each actual abnormal time period, and screen out the high-abnormal time periods.

[0005] Furthermore, the method for obtaining the abnormal coefficient includes: Obtain the preset normal data range for each sign parameter; Obtain the ratio of the detection data of each sign parameter at each moment to the maximum value within the preset normal data range as the abnormal coefficient of each sign parameter at each moment.

[0006] Furthermore, the method for obtaining the vital sign parameters includes: For any examination item, obtain the number of abnormal times when the detection data of each sign parameter at different times does not belong to the preset normal data range, obtain the number of abnormal times when the detection data of all sign parameters at different times does not belong to the preset normal data range, and obtain the ratio of the number of abnormal times corresponding to each sign parameter and all sign parameters as the abnormal frequency of each sign parameter; According to the fluctuation degree of the abnormal coefficient of each sign parameter at different times and the abnormal frequency, obtain the importance index of each sign parameter. Both the fluctuation degree of the abnormal coefficient and the abnormal frequency are positively correlated with the importance index; If the importance index of the sign parameter is greater than or equal to the preset importance threshold, take the corresponding sign parameter as the vital sign parameter.

[0007] Furthermore, the method for obtaining the detection status abnormality degree includes: According to the fluctuation characteristics of the data difference between the fitting data of adjacent moments within the local range of each moment and the maximum value of the abnormal coefficient among all corresponding moments, obtain the detection status abnormality degree of the local range of the corresponding moment. The fluctuation characteristics of the data difference are negatively correlated with the detection status abnormality degree, and the maximum value of the abnormal coefficient is positively correlated with the detection status abnormality degree.

[0008] Furthermore, the method for obtaining the detection status change indicator includes: Perform negative correlation normalization on the time difference between adjacent moments as the time influence coefficient; Fuse the time influence coefficient between adjacent moments with the maximum value of the detection status abnormality degree of the local range of all corresponding moments to obtain the detection status change indicator between adjacent moments.

[0009] Furthermore, the obtaining of the normal time period and the initial abnormal time period includes: Obtain the detection status change index sequence between all adjacent moments according to the time sequence; perform difference processing on the detection status change index sequence to obtain the difference sequence of the detection status change index sequence; If a target element appears in the difference sequence, the target element is greater than the preset difference threshold, and the difference between the corresponding element and the next adjacent element is less than or equal to the preset difference threshold, the continuous time range of the detection status change index corresponding to the target element and the next adjacent element in the detection status change index sequence is used as the initial abnormal time period; the other continuous time ranges except the initial abnormal time period are used as the normal time periods.

[0010] Further, the method for obtaining the abnormal degree status coefficient includes: Obtain the sum of the durations between the corresponding adjacent normal time periods before and after each initial abnormal time period as the first sum value; Obtain the ratio of the duration of each initial abnormal time period to the first sum value and perform normalization as the first abnormal coefficient; Fuse the maximum value of the detection status change index corresponding to the initial abnormal time period and the first abnormal coefficient as the abnormal degree status coefficient of each initial abnormal time period.

[0011] Further, the method for obtaining the actual abnormal time period includes: If the abnormal degree status coefficient of the initial abnormal time period is greater than the preset coefficient threshold, the corresponding initial abnormal time period is used as the actual abnormal time period.

[0012] Further, the method for obtaining the detection data of the next moment includes: For any important vital sign parameter, fuse the detection data at the real-time moment and the corresponding detection status change index to obtain the detection data of the next moment.

[0013] Further, the method for obtaining the high-abnormal time period includes: Select the maximum value of the abnormal degree status coefficient of the actual abnormal time period within the neighborhood range of the real-time moment as the reference abnormal degree; Within the neighborhood range of the next moment, obtain the ratio of the abnormal degree status coefficient of each actual abnormal time period to the reference abnormal degree as the abnormal weight of each actual abnormal time period; If the abnormal weight of the actual abnormal time period is greater than the preset weight threshold, the corresponding actual abnormal time period is used as the high-abnormal time period.

[0014] The present invention has the following beneficial effects: For different examination items, according to the distribution of the detection data of each physical sign parameter at different times, the abnormality coefficient of each physical sign parameter at each time is obtained, and the important physical sign parameters are screened out, and the physical sign parameters that play a key role in diagnosis are screened out; for any important physical sign parameter, the fitting curve of the corresponding detection data at all times is obtained, which reflects the trend of the physical sign parameter changing with time, helps to identify potential change rules, and according to the distribution of the fitting data within the local range of each time on the fitting curve and the abnormality coefficient corresponding to each time, the detection status abnormality degree of the local range of each time is obtained, and the detection status of each time is evaluated more carefully; according to the time difference between adjacent times and the detection status abnormality degree of the local range corresponding to the corresponding time, the detection status change index between adjacent times is obtained, which can quantitatively evaluate the change speed and trend of the physical sign parameter, and helps to identify the rapid change or abnormal fluctuation of the physical sign parameter; according to the change trend of the detection status change index between all adjacent times, the normal time period and the initial abnormal time period are obtained, which helps to narrow the scope of abnormal detection and improve the accuracy and efficiency of diagnosis; according to the time distribution characteristics between each initial abnormal time period and the adjacent normal time period, and the detection status change index corresponding to the initial abnormal time period, the abnormality degree status coefficient of each initial abnormal time period is obtained, and the actual abnormal time period is screened out, and the severity of each abnormal time period is evaluated more accurately; according to the detection data at the real-time moment and the corresponding detection status change index, the detection data at the next moment is obtained, and the possible changes of the physical sign parameter are understood in advance; the detection data within the neighborhood range of the real-time moment or the neighborhood range of the next moment of each important physical sign parameter is analyzed, the abnormality degree status coefficient corresponding to each actual abnormal time period is obtained, and the high-abnormality time period is screened out. The present invention improves the accuracy of pregnancy risk assessment by obtaining important physical sign parameters and accurate abnormal state distributions. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required to be used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0016] Figure 1 It is a flowchart of a method for assessing the pregnancy risk of pregnant women based on the previous electronic medical history provided by an embodiment of the present invention; Figure 2 It is a flowchart of a method for obtaining important physical sign parameters provided by an embodiment of the present invention; Figure 3 It is a flowchart of a method for obtaining the abnormality degree status coefficient provided by an embodiment of the present invention. Detailed implementation manners

[0017] In order to further elaborate on the technical means and effects adopted by the present invention to achieve the intended invention purpose, the following combines the accompanying drawings and preferred embodiments to specifically describe a method for assessing the pregnancy risk of pregnant women based on their previous electronic medical histories, including its specific implementation manners, structures, features, and effects, as follows. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, the specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0018] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which the present invention belongs.

[0019] The following specifically describes the specific solution of a method for assessing the pregnancy risk of pregnant women based on their previous electronic medical histories provided by the present invention with reference to the accompanying drawings.

[0020] Please refer to Figure 1 , which shows a flowchart of a method for assessing the pregnancy risk of pregnant women based on their previous electronic medical histories provided by an embodiment of the present invention. The specific method includes: Step S1: Obtain the detection data of multiple physical sign parameters of pregnant women in different examination items according to the time sequence.

[0021] In the embodiment of the present invention, in order to improve the data analysis efficiency and the accuracy of the diagnosis result of pregnancy risk, during the analysis of the previous electronic medical history of pregnant women, it is necessary to analyze the physical sign parameters to avoid the influence of a large amount of useless data information; first, the previous electronic medical history form includes detection items such as blood routine and clinical tests, and each detection item includes multiple physical sign parameters. For example, the blood routine includes physical sign parameters such as white blood cell count, red blood cell count, and platelet count. It is necessary to sort the previous electronic medical history form according to the sequence of the moments of pregnant women, and sort the examination items in different forms in the same order, which helps to reduce information omission and more conveniently obtain and analyze the detection data of each physical sign parameter of pregnant women; obtain the detection data of multiple physical sign parameters of pregnant women in different examination items according to the time sequence.

[0022] It should be noted that in the embodiment of the present invention, during each detection of pregnant women, the implementer can pre-obtain the corresponding next moment according to the specific situation.

[0023] Step S2: For different examination items, obtain the abnormal coefficient of each physical sign parameter at each moment according to the distribution of the detection data of each physical sign parameter at different moments, and screen out the important physical sign parameters.

[0024] The test results of pregnant and lying-in women at different times are different. By analyzing the distribution of test data of each physical sign parameter at different times, the data fluctuation at different times can be reflected, and the abnormal degree of the physical sign parameter can be quantified. The greater the abnormal degree, the more attention should be paid to the physical sign parameter, which helps the subsequent analysis of the abnormal state. For different examination items, according to the distribution of test data of each physical sign parameter at different times, the abnormal coefficient of each physical sign parameter at each time is obtained, and the important physical sign parameters are screened out.

[0025] Preferably, in an embodiment of the present invention, the method for obtaining the abnormal coefficient includes: Obtain the preset normal data range of each physical sign parameter; Obtain the ratio of the test data of each physical sign parameter at each time to the maximum value within the preset normal data range as the abnormal coefficient of each physical sign parameter at each time.

[0026] It should be noted that in an embodiment of the present invention, the preset normal data range can be used to quantify the deviation degree of the test data of the physical sign parameter from the normal data distribution at different times. The preset normal data range of the physical sign parameter is obtained in advance by the implementer according to relevant professional materials. For example, the normal data range of white blood cell count in adults is 4.0×10 9 ~10.0×10 9 / L.

[0027] Preferably, in an embodiment of the present invention, for the method for obtaining important physical sign parameters, please refer to Figure 2 , which shows a flowchart of a method for obtaining important physical sign parameters, including: Step S201: For any examination item, obtain the number of abnormalities of each physical sign parameter whose test data at different times does not belong to the preset normal data range, obtain the number of abnormalities of all physical sign parameters whose test data at different times does not belong to the preset normal data range, and obtain the ratio of the corresponding number of abnormalities of each physical sign parameter and all physical sign parameters as the abnormal frequency of each physical sign parameter.

[0028] By analyzing the abnormal frequency, it can be identified which physical sign parameters are more likely to deviate from the normal data range. The greater the abnormal frequency, the greater the possibility of the corresponding physical sign parameter being abnormal.

[0029] Step S202: According to the fluctuation degree of the abnormal coefficient of each physical sign parameter at different times and the abnormal frequency, obtain the importance index of each physical sign parameter. Both the fluctuation degree of the abnormal coefficient and the abnormal frequency are positively correlated with the importance index.

[0030] The degree of fluctuation of the anomaly coefficient can reflect the degree of dispersion of abnormal data of the physical sign parameters at different times. The greater the degree of fluctuation, the greater the degree of abnormal dispersion, the more unstable the data change, the more attention is needed, and the greater the importance index; the greater the anomaly frequency, the more times the data of the physical sign parameters are abnormal, which has a greater impact on pregnant women, and the greater the importance index.

[0031] It should be noted that in an embodiment of the present invention, the degree of fluctuation can be represented by calculating the variance. The greater the variance, the greater the degree of fluctuation; the smaller the variance, the smaller the degree of fluctuation. In other embodiments of the present invention, the degree of fluctuation can also be expressed by the standard deviation, range, etc. The specific means are well-known technical means to those skilled in the art and will not be elaborated here.

[0032] In an embodiment of the present invention, for any inspection item, the formula of the importance index is expressed as: ; Wherein, represents the importance index of the th physical sign parameter; represents the degree of fluctuation of the anomaly coefficient of the th physical sign parameter at all times; represents the number of anomalies that the detection data of the th physical sign parameter does not belong to the preset normal data range at all times; represents the number of anomalies that the detection data of all physical sign parameters do not belong to the preset normal data range at all times; represents the normalization function; represents the ratio of the th physical sign parameter to the corresponding number of anomalies of all physical sign parameters, as the anomaly frequency.

[0033] Step S203: If the importance index of the physical sign parameter is greater than or equal to the preset importance threshold, the corresponding physical sign parameter is regarded as an important physical sign parameter.

[0034] It should be noted that in an embodiment of the present invention, the preset importance threshold is 0.4. In other embodiments of the present invention, the size of the preset importance threshold can be specifically set according to specific circumstances and will not be limited and elaborated here.

[0035] Based on this, analyze different inspection items to obtain all important physical sign parameters.

[0036] Step S3: For any important vital sign parameter, obtain the fitting curve of the corresponding detection data at all times. According to the distribution of the fitting data within the local range at each time on the fitting curve and the abnormal coefficient corresponding to each time, obtain the detection status abnormality degree of the local range at each time; according to the time difference between adjacent times and the detection status abnormality degree of the local range at the corresponding time, obtain the detection status change index between adjacent times.

[0037] To more intuitively understand the change trend and law of the detection data at different times, for any important vital sign parameter, obtain the fitting curve of the corresponding detection data at all times.

[0038] It should be noted that in an embodiment of the present invention, by performing least squares fitting on the corresponding detection data at all times, the corresponding fitting curve is obtained, that is, the fitting data with the time as the abscissa and the vital sign parameter as the ordinate; in other embodiments of the present invention, curve fitting can also be performed through existing fitting algorithms such as polynomial fitting and spline interpolation; the specific means are well-known technical means to those skilled in the art and will not be elaborated here.

[0039] The distribution of the fitting data reflects the volatility and stability of the fitting data. If the data distribution within the local range is relatively concentrated and stable, it indicates that the data state is relatively normal; on the contrary, if the data distribution is discrete or shows abnormal fluctuations, it may indicate that the data is in an abnormal state; since the data state is also relatively stable in the case of continuous abnormal states, by introducing the abnormal coefficient, it is possible to more accurately identify which times the data is in an abnormal state. According to the distribution of the fitting data within the local range at each time on the fitting curve and the abnormal coefficient corresponding to each time, obtain the detection status abnormality degree of the local range at each time.

[0040] Preferably, in an embodiment of the present invention, the method for obtaining the detection status abnormality degree includes: According to the fluctuation characteristics of the data difference between adjacent times within the local range at each time and the maximum value of the abnormal coefficient among all times, obtain the detection status abnormality degree of the local range at the corresponding time. The fluctuation characteristics of the data difference are negatively correlated with the detection status abnormality degree, and the maximum value of the abnormal coefficient is positively correlated with the detection status abnormality degree.

[0041] Among them, the greater the fluctuation characteristics of the data difference, the less similar the data differences, the more unstable the state performance, the greater the abnormal coefficient, and the greater the detection status abnormality degree.

[0042] It should be noted that in an embodiment of the present invention, the mean value of the ratios between all adjacent data differences within a local range at a moment is calculated as the difference change level; wherein, in order to avoid the denominator of the formula being 0 and the formula being meaningless, a manually set threshold, such as 0.01, is added to the denominator when calculating the ratio; the difference between the difference change level and the positive integer 1 is calculated, and The function is used for normalization to obtain the state stability coefficient, which represents the fluctuation characteristics. That is, the larger the state stability coefficient, the closer the difference between the difference change level and the positive integer 1, and the smaller the fluctuation characteristics of the data difference, indicating that the results of multiple detections change relatively consistently.

[0043] In an embodiment of the present invention, for any local range at a moment, the formula for detecting the state abnormality degree is expressed as: ; Wherein, Represents the detection state abnormality degree of each local range at a moment; Represents the number of moments within the local range at a moment; Represents the Difference between the data of the moment and the fitted data between the Moment; Represents the Difference between the data of the moment and the fitted data between the Moment; Represents the maximum value of the abnormality coefficient within the local range at a moment; Represents the first derivative of the logistic function.

[0044] In the formula for detecting the state abnormality degree, Represents the Ratio of the data difference between the moment and the fitted data between the Moment to the data difference between the Moment and the fitted data between the Moment; Represents the mean value of calculating all ratios within the local range at a moment, that is, the difference change level. The larger the difference change level, the larger the mean value, and the larger the ratio of the data difference, indicating that the change of the data between moments is more unstable; using For normalization, the less the data differences are close, that is, there are differences, the greater the fluctuation characteristics of the data change, and the smaller the detection state abnormality degree; on the contrary, the closer the data differences are, the smaller the fluctuation characteristics of the data change, the larger the maximum value of the abnormality coefficient, and the larger the detection state abnormality degree.

[0045] It should be noted that in an embodiment of the present invention, the time local range is mainly based on each moment and forms a range with the two adjacent subsequent moments, that is, traversing the fitting curve and analyzing the obtained different time local ranges in sequence; in other embodiments of the present invention, the size of the time local range can be specifically set according to specific circumstances, and no limitation and elaboration will be made here.

[0046] There are differences in the time span of pregnant and lying-in women during examinations. If there is no abnormal condition or the degree of abnormality is low, the time interval for the next examination of pregnant and lying-in women is relatively long, and there may be a situation where other diseases are not examined; by analyzing the magnitude of the time difference between adjacent moments, the fineness of the state change can be captured. The smaller the time difference, the more attention can be paid to the subtle changes in the abnormality degree of the detection state, and the greater the impact on the abnormality degree of the detection state. The larger the time difference, the more likely some important change information will be missed, and the smaller the impact on the abnormality degree of the detection state; according to the time difference between adjacent moments and the abnormality degree of the detection state in the corresponding time local range, the detection state change index between adjacent moments is obtained.

[0047] Preferably, in an embodiment of the present invention, the method for obtaining the detection state change index includes: Performing negative correlation normalization on the time difference between adjacent moments as the time influence coefficient; Fusing the time influence coefficient between adjacent moments with the maximum value of the abnormality degree of the detection state in the corresponding time local range of all moments to obtain the detection state change index between adjacent moments.

[0048] In an embodiment of the present invention, for the th moment and the th moment, the formula for the detection state change index is expressed as: ; Wherein, represents the detection state change index between adjacent moments; represents the time difference between the th moment and the th moment; represents the maximum value of the time differences between all adjacent moments; represents the maximum value of the abnormality degree of the detection state in the corresponding time local range of all moments; represents the maximum-minimum normalization function.

[0049] In the formula of the detection state change index, It represents the ratio of the maximum calculated time difference to the time difference between adjacent moments, and normalization is performed, that is, negative correlation normalization is performed on the time difference between adjacent moments to obtain a time influence coefficient. The larger the time influence coefficient, the greater the time difference between adjacent moments, the smaller the influence of the detection data on the status of the pregnant and lying-in women, the smaller the abnormality degree of the detection status, and the smaller the detection status change index.

[0050] Step S4: Obtain the normal time period and the initial abnormal time period according to the change trend of the detection status change index between all adjacent moments; according to the time distribution characteristics between each initial abnormal time period and the adjacent normal time period, and the corresponding detection status change index within the initial abnormal time period, obtain the abnormal degree status coefficient of each initial abnormal time period, and screen out the actual abnormal time period.

[0051] The detection status change index reflects the change trend of the status of pregnant and lying-in women at different moments. By dividing the normal time period, it helps to analyze the status change within each time period more carefully. According to the change trend of the detection status change index between different adjacent moments, the normal time period and the initial abnormal time period are obtained.

[0052] Preferably, in an embodiment of the present invention, the method for obtaining the normal time period and the initial abnormal time period includes: Obtain the detection status change index sequence between all adjacent moments in chronological order; perform a difference operation on the detection status change index sequence to obtain the difference sequence of the detection status change index sequence; If a target element appears in the difference sequence, the target element is greater than the preset difference threshold, and the difference between the corresponding element and the next adjacent element is less than or equal to the preset difference threshold, the continuous time range of the detection status change index corresponding to each target element and the next adjacent element in the detection status change index sequence is used as the initial abnormal time period; the other continuous time ranges except the initial abnormal time period are used as the normal time period.

[0053] It should be noted that, in an embodiment of the present invention, the preset difference threshold is 0.7, and the preset difference threshold is 0.1; in other embodiments of the present invention, the preset difference threshold and the preset difference threshold can be specifically set by the implementer according to the specific situation, and will not be limited and elaborated here.

[0054] Make an example. If there is a sequence of detection status change indicators 0.1, 0.2, 0.2, 0.9, 0.3, 0.2, and the corresponding adjacent times are 12, 23, 35, 57, 78, 89, the obtained difference sequence is 0.1, 0, 0.7, 0.6, 0.1. The third element satisfies being greater than the preset difference threshold, and the difference from the next adjacent element is less than or equal to the preset difference threshold, that is, the third element is the target element. The third and fourth elements are the detection status change indicators 0.2, 0.9, 0.3 in the detection status change indicator sequence, and the time range of the corresponding adjacent times 35, 57, 78 is 3 - 8 as the initial abnormal time period, and other continuous time ranges 1 - 3 or 8 - 9 are used as normal time periods.

[0055] The body of pregnant and lying-in women is in a stable state for a long time, and the similarity of the states before and after the occurrence of abnormal states is relatively high. By analyzing the time distribution characteristics between each initial abnormal time period and the adjacent normal time periods, the persistence and change trend of the abnormality can be reflected; the detection status change indicator is an important indicator for evaluating the state change within the abnormal time period. The larger the detection status change indicator, the greater the possibility of state change and the greater the degree of influence by the abnormality; through comprehensive analysis, a more comprehensive understanding of the actual situation of the abnormal time period can be obtained. According to the time distribution characteristics between each initial abnormal time period and the adjacent normal time periods, as well as the corresponding detection status change indicators within the initial abnormal time period, the abnormality degree status coefficient of each initial abnormal time period is obtained, and the actual abnormal time periods are screened out.

[0056] Preferably, in an embodiment of the present invention, for the method of obtaining the abnormality degree status coefficient, please refer to Figure 3 , which shows a flowchart of a method for obtaining the abnormality degree status coefficient, including: Step S301: Obtain the sum of the durations between the corresponding adjacent normal time periods before and after each initial abnormal time period as the first sum value; obtain the ratio of the duration of each initial abnormal time period to the first sum value and perform normalization as the first abnormality coefficient.

[0057] Through the duration, the degree of change can be captured. The ratio of the duration of each abnormal time period to the first sum value reflects the comprehensive performance of the duration of the abnormal time period in the time relationship with the adjacent normal time periods. The shorter the duration of the abnormal time period, the more likely it is a short-term abnormality, and the smaller the first abnormality coefficient.

[0058] Step S302: Fuse the maximum value of the detection status change indicator within the initial abnormal time period and the first abnormality coefficient as the abnormality degree status coefficient of each initial abnormal time period.

[0059] In some embodiments of the present invention, fusion is performed by addition or multiplication, and the specific means are well-known technical means to those skilled in the art, which will not be limited and elaborated herein.

[0060] In one embodiment of the present invention, the formula for the abnormal degree status coefficient is expressed as: ; wherein, represents the abnormal degree status coefficient of the th initial abnormal time period; represents the duration of the th initial abnormal time period; represents the duration of the immediately preceding normal time period adjacent to the th initial abnormal time period; represents the duration of the immediately following normal time period adjacent to the th initial abnormal time period; represents the maximum value of the corresponding detection status change index within the th initial abnormal time period; represents a non-linear activation function.

[0061] In the formula for the abnormal degree status coefficient, represents the first sum value; represents the ratio of calculating the duration of the th initial abnormal time period to the first sum value; represents normalizing the ratio of the duration of the th initial abnormal time period to the first sum value, that is, the first abnormal coefficient. The larger the ratio, the smaller the sum of the durations of the immediately preceding normal time period and the immediately following normal time period adjacent to the th initial abnormal time period, the larger the duration of the th initial abnormal time period, the larger the first abnormal coefficient, and the larger the abnormal degree status coefficient; the larger the maximum value of the corresponding detection status change index within the th initial abnormal time period, the higher the degree of abnormal influence, and the smaller the abnormal degree status coefficient.

[0062] Considering that improper diet and excessive salt intake may lead to short-term parameter abnormalities, but this type of abnormality has no value for pregnancy risk. By screening the actual abnormal time periods, it helps to improve the detection efficiency.

[0063] Preferably, in one embodiment of the present invention, the method for obtaining the actual abnormal time period includes: If the abnormal degree status coefficient of the abnormal time period is greater than the preset coefficient threshold, the corresponding initial abnormal time period is used as the actual abnormal time period.

[0064] It should be noted that, in an embodiment of the present invention, the preset coefficient threshold is 0.4; in other embodiments of the present invention, the size of the preset coefficient threshold can be specifically set according to specific circumstances, and no limitation and elaboration will be made here.

[0065] Step S5: Obtain the detection data of the next moment according to the detection data at the real-time moment and the corresponding detection status change index; analyze the detection data within the neighborhood range of the real-time moment or the neighborhood range of the next moment for each important vital sign parameter to obtain the abnormal degree status coefficient corresponding to each actual abnormal time period, and screen out the high-abnormal time periods.

[0066] By monitoring the real-time changes of the patient's vital sign parameters, the health status of the patient can be understood in a timely manner; the detection status change index provides information about the change trend of the vital sign parameters, which can predict the detection data of the next moment, so as to understand the possible health status of the patient in advance. Obtain the detection data of the next moment according to the detection data at the real-time moment and the corresponding detection status change index.

[0067] Preferably, in an embodiment of the present invention, the method for obtaining the detection data of the next moment includes: For any important vital sign parameter, fuse the detection data at the real-time moment and the corresponding detection status change index to obtain the detection data of the next moment.

[0068] It should be noted that, in some embodiments of the present invention, the method of addition or multiplication can be used for fusion, and the specific means are well-known technical means to those skilled in the art, and no elaboration will be made here.

[0069] The abnormal degree status coefficient can quantify the abnormal degree of the vital sign parameters during the abnormal time period, so as to more accurately understand the health status of the patient. Analyze the detection data within the neighborhood range of the real-time moment or the neighborhood range of the next moment for each important vital sign parameter to obtain the abnormal degree status coefficient corresponding to each actual abnormal time period, and screen out the high-abnormal time periods, which helps to understand the health status of the patient in a timely manner and improve the accuracy and efficiency of diagnosis.

[0070] Preferably, in an embodiment of the present invention, the method for obtaining the high-abnormal time period includes: Select the maximum value of the abnormal degree status coefficient of the actual abnormal time period within the neighborhood range of the real-time moment as the reference abnormal degree; Within the neighborhood range of the next moment, obtain the ratio of the abnormal degree status coefficient of each actual abnormal time period to the reference abnormal degree as the abnormal weight value of each actual abnormal time period; If the abnormal weight value of the actual abnormal time period is greater than the preset weight threshold, the corresponding actual abnormal time period is used as the high-abnormal time period.

[0071] It should be noted that, in an embodiment of the present invention, the preset weight threshold is 0.5; in other embodiments of the present invention, the size of the preset weight threshold can be specifically set according to specific circumstances, and no limitation and elaboration will be made here.

[0072] It should be noted that, in an embodiment of the present invention, the method for obtaining the neighborhood range is based on each moment and forms a range with all historical moments; in other embodiments of the present invention, the size of the neighborhood range can be specifically set according to specific circumstances, and no limitation and elaboration will be made here.

[0073] Based on this, by analyzing all important vital sign parameters and obtaining the high-abnormality time period, the status of pregnant women can be analyzed in detail, improving the accuracy of analyzing the abnormal status of pregnant women and enhancing the detection efficiency.

[0074] In summary, for any important vital sign parameter of the present invention, the detection status abnormality degree of the local range at each moment is obtained; by combining the time difference between adjacent moments, the detection status change index between adjacent moments is obtained; the normal time period and the initial abnormal time period are obtained for time feature analysis, and by combining the detection status change index corresponding to the initial abnormal time period, the abnormality degree status coefficient of each initial abnormal time period is obtained, and the actual abnormal time period is screened out; the detection data within the neighborhood range of the real-time moment or the neighborhood range of the next moment for each important vital sign parameter is analyzed, the abnormality degree status coefficient corresponding to each actual abnormal time period is obtained, and the high-abnormality time period is screened out. The present invention improves the accuracy of pregnancy risk assessment by obtaining important vital sign parameters and accurate abnormal status distributions.

[0075] It should be noted that: the above sequence of embodiments of the present invention is only for description and does not represent the superiority or inferiority of the embodiments. The processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0076] Each embodiment in this specification is described in a progressive manner. The same or similar parts among the embodiments can be referred to each other, and the key points of each embodiment are the differences from other embodiments.

Claims

1. A method for assessing the pregnancy risk of pregnant women based on their previous electronic medical history, characterized in that, The method includes: Obtaining the detection data of multiple physical sign parameters of pregnant and lying-in women in different examination items according to the time sequence; For different examination items, according to the distribution of the detection data of each physical sign parameter at different times, obtaining the abnormality coefficient of each physical sign parameter at each moment, and screening out the important physical sign parameters; For any important physical sign parameter, obtaining the fitting curve of the corresponding detection data at all times, and according to the distribution of the fitting data in the local range of each moment on the fitting curve and the abnormality coefficient corresponding to each moment, obtaining the detection state abnormality degree of the local range of each moment; according to the time difference between adjacent moments and the detection state abnormality degree of the local range of the corresponding moment, obtaining the detection state change index between adjacent moments; According to the change trend of the detection state change index between all adjacent moments, obtaining the normal time period and the initial abnormal time period; according to the time distribution characteristics between each initial abnormal time period and the adjacent normal time period, and the detection state change index corresponding to the initial abnormal time period, obtaining the abnormality degree state coefficient of each initial abnormal time period, and screening out the actual abnormal time period; According to the detection data at the real-time moment and the corresponding detection state change index, obtaining the detection data at the next moment; analyzing the detection data in the neighborhood range of the real-time moment or the neighborhood range of the next moment for each important physical sign parameter, obtaining the abnormality degree state coefficient corresponding to each actual abnormal time period, and screening out the high-abnormality time period.

2. The maternal pregnancy risk assessment method based on previous electronic medical history according to claim 1, wherein The method for obtaining the abnormality coefficient includes: Obtaining the preset normal data range of each physical sign parameter; Obtaining the ratio of the detection data of each physical sign parameter at each moment to the maximum value within the preset normal data range as the abnormality coefficient of each physical sign parameter at each moment.

3. The method for assessing the pregnancy risk of pregnant women based on previous electronic medical records according to claim 2, wherein The method for obtaining the important physical sign parameters includes: For any examination item, obtaining the number of abnormalities of the detection data of each physical sign parameter that does not belong to the preset normal data range at different times, obtaining the number of abnormalities of the detection data of all physical sign parameters that do not belong to the preset normal data range at different times, and obtaining the ratio of the number of abnormalities corresponding to each physical sign parameter and all physical sign parameters as the abnormality frequency of each physical sign parameter; According to the fluctuation degree of the abnormality coefficient of each physical sign parameter at different times and the abnormality frequency, obtaining the importance index of each physical sign parameter, and both the fluctuation degree of the abnormality coefficient and the abnormality frequency are positively correlated with the importance index; The formula of the importance index is expressed as: ； Among them, represents the importance index of the th feature parameter; represents the fluctuation degree of the anomaly coefficient of the th feature parameter at all times; represents the number of anomalies where the detection data of the th feature parameter does not belong to the preset normal data range at all times; represents the number of anomalies where the detection data of all feature parameters does not belong to the preset normal data range at all times; represents the normalization function; represents the ratio of the th feature parameter and the corresponding number of anomalies of all feature parameters, as the anomaly frequency; If the importance index of the physical sign parameter is greater than or equal to the preset importance threshold, the corresponding physical sign parameter is used as an important physical sign parameter.

4. The method for assessing the pregnancy risk of pregnant women based on previous electronic medical history according to claim 1, wherein The method for obtaining the detection state abnormality degree includes: According to the fluctuation characteristics of the data difference between adjacent moments in the local range of each moment and the maximum value of the abnormality coefficient among all moments, obtaining the detection state abnormality degree of the local range of the corresponding moment, the fluctuation characteristics of the data difference are negatively correlated with the detection state abnormality degree, and the maximum value of the abnormality coefficient is positively correlated with the detection state abnormality degree; The formula of the detection state abnormality degree is expressed as: ； Among them, represents the detection status abnormality degree of the local range at each moment; represents the number of moments within the local range of the moment; represents the th moment and the th moment, the data difference of the fitting data between them; represents the th moment and the th moment, the data difference of the fitting data between them; represents the maximum value of the abnormality coefficient within the local range of the moment; represents the first derivative of the logistic function.

5. The maternal pregnancy risk assessment method based on previous electronic medical history according to claim 1, characterized in that, The method for obtaining the detection state change index includes: Perform negative correlation normalization on the time difference between adjacent moments as the time influence coefficient; Fuse the time influence coefficient between adjacent moments with the maximum value of the detection status abnormality degree in the local range of all corresponding moments to obtain the detection status change index between adjacent moments; The negative correlation normalization of the time difference between adjacent moments includes: calculating the ratio of the maximum time difference and the time difference between adjacent moments and then performing normalization; The formula for the detection status change index is expressed as: ； Among them, represents the detection status change index between adjacent time instants; represents the time instant and the time difference between time instants; represents the maximum value of the time differences between all adjacent time instants; represents the maximum value of the detection status abnormality degree in the local range corresponding to all time instants; represents the maximum-minimum normalization function.

6. The method for assessing the pregnancy risk of pregnant women based on their previous electronic medical history according to claim 1, characterized in that, The obtaining of the normal time period and the initial abnormal time period includes: Obtain the detection status change index sequence between all adjacent moments in chronological order; perform difference processing on the detection status change index sequence to obtain the difference sequence of the detection status change index sequence; If a target element appears in the difference sequence, the target element is greater than the preset difference threshold, and the difference between the corresponding element and the next adjacent element is less than or equal to the preset difference threshold, the continuous moment range of the detection status change index corresponding to the target element and the next adjacent element in the detection status change index sequence is used as the initial abnormal time period; the other continuous moment ranges except the initial abnormal time period are used as the normal time period.

7. The method for assessing the pregnancy risk of pregnant women based on previous electronic medical records according to claim 6, wherein The obtaining method of the abnormal degree status coefficient includes: Obtain the sum of the durations between the corresponding adjacent normal time periods before and after each initial abnormal time period as the first sum value; Obtain the ratio of the duration of each initial abnormal time period to the first sum value and perform normalization as the first abnormal coefficient; Fuse the maximum value of the detection status change index corresponding to the initial abnormal time period and the first abnormal coefficient as the abnormal degree status coefficient of each initial abnormal time period; The formula for the abnormal degree status coefficient is expressed as: ； Among them, represents the abnormal degree status coefficient of the th initial abnormal time period; represents the duration of the th initial abnormal time; represents the duration of the adjacent previous normal time period of the th initial abnormal time period; represents the duration of the adjacent next normal time period of the th initial abnormal time period; represents the maximum value of the corresponding detection status change index within the th initial abnormal time period; represents the non-linear activation function.

8. A method for assessing the pregnancy risk of pregnant women based on their previous electronic medical history according to claim 1, characterized in that, The obtaining method of the actual abnormal time period includes: If the abnormal degree status coefficient of the initial abnormal time period is greater than the preset coefficient threshold, the corresponding initial abnormal time period is used as the actual abnormal time period.

9. The method for assessing the pregnancy risk of pregnant women based on previous electronic medical history according to claim 1, wherein The obtaining method of the detection data of the next moment includes: For any important vital sign parameter, fuse the detection data at the real-time moment and the corresponding detection status change index to obtain the detection data of the next moment.

10. The method for assessing the pregnancy risk of pregnant women based on their previous electronic medical history according to claim 1, wherein The obtaining method of the high-abnormal time period includes: Select the maximum value of the abnormal degree status coefficient of the actual abnormal time period within the neighborhood range of the real-time moment as the reference abnormal degree; Within the neighborhood range of the next moment, obtain the ratio of the abnormal degree status coefficient of each actual abnormal time period to the reference abnormal degree as the abnormal weight of each actual abnormal time period; If the abnormal weight of the actual abnormal time period is greater than the preset weight threshold, the corresponding actual abnormal time period is used as the high-abnormal time period.

Images