A method and system for monitoring patient blood pressure for clinical nursing

Abstract

The present invention relates to the technical field of blood pressure monitoring, and in particular to a method and system for monitoring blood pressure of patients for clinical care. The method obtains the controllability of blood pressure by the distribution fluctuation degree of blood pressure data in the neighborhood range at each moment; obtains the blood pressure disorder by the distribution of blood pressure data corresponding to the sampling moment at the same position in different period segments and the blood pressure controllability; obtains the blood pressure disorder at each sampling moment by the blood pressure disorder in the period segment at the current moment and the blood pressure change in the neighborhood range, obtains the blood pressure disorder based on the distribution of blood pressure disorder, and performs DBSCAN clustering on the blood pressure data in the period segment at the current moment to obtain the monitoring clustering result for blood pressure status monitoring. The present invention combines the blood pressure fluctuation situation and analyzes the abnormality of blood pressure data through multiple periods to adaptively adjust the neighborhood radius of clustering, improve the sensitivity of clustering to abnormal points, obtain more accurate clustering results, and make the monitoring results more reliable.

Description

Technical Field

[0001] The present invention relates to the technical field of blood pressure monitoring, and in particular to a method and system for monitoring the blood pressure of a patient for clinical care. Background Art

[0002] Kidney disease may cause blood flow obstruction in the kidneys, and the body will increase pressure to maintain normal operation. Therefore, kidney conditions are closely related to blood pressure, and blood pressure monitoring is extremely important in the clinical care of kidney disease. For patients with kidney disease, monitoring abnormalities only by blood pressure data not being within the normal range will cause many abnormal conditions to be ignored, such as persistently high blood pressure, high short-term mutations, or high blood pressure at night, etc. Therefore, it is necessary to monitor blood pressure in real time, and through feedback on possible abnormal blood pressure monitoring, timely control of abnormal conditions of patients, and take intervention measures to prevent further deterioration. Therefore, in the clinical care of kidney disease, real-time blood pressure monitoring should be regarded as an important monitoring indicator.

[0003] Due to the irregular fluctuation characteristics of blood pressure data of patients with kidney disease, DBSCAN clustering can be used for abnormal data detection. DBSCAN clustering can cluster data sets of any shape. Even if the blood pressure data presents an irregular shape, it can be effectively clustered. It has no specific requirements for the distribution of data sets and is more flexible in processing unknown structure data sets. It is suitable for various complex data sets. The anomaly detection of DBSCAN clustering depends on parameter settings, that is, the setting of the neighborhood radius and the minimum number of points. Since the blood pressure data of patients with kidney disease is very easy to fluctuate, DBSCAN clustering with fixed parameters will make the clustering process insensitive to abnormal points, which will cause abnormal blood pressure data to be clustered, resulting in inaccurate abnormal identification and low accuracy of blood pressure monitoring results. Summary of the invention

[0004] In order to solve the technical problem that the DBSCAN clustering with fixed parameters in the prior art will make the clustering process insensitive to abnormal points, which will cause abnormal blood pressure data to be clustered, resulting in inaccurate abnormal identification and low accuracy of blood pressure monitoring results, the purpose of the present invention is to provide a patient blood pressure monitoring method and system for clinical care, and the technical solutions adopted are as follows:

[0005] The present invention provides a method for monitoring blood pressure of a patient for clinical nursing, the method comprising:

[0006] Different cycle segments are preset during the monitoring of kidney disease patients to obtain blood pressure data at each sampling moment in the cycle segment; all cycle segments are continuously distributed and have the same length;

[0007] According to the numerical distribution and fluctuation of blood pressure data within a preset neighborhood range at each sampling moment, the controllability of blood pressure at each sampling moment is obtained;

[0008] According to the blood pressure data distribution disorder degree at the same position at the sampling moment in different period segments and the blood pressure controllability at each sampling moment at the same position, the blood pressure disorder degree at each sampling moment is obtained;

[0009] According to the blood pressure disorder degree at each sampling moment in the cycle segment at the current moment and the data change degree of the blood pressure data within the preset neighborhood range at each sampling moment, the blood pressure disorder at each sampling moment in the cycle segment at the current moment is obtained;

[0010] The adjustment neighborhood radius is obtained through the numerical distribution of all blood pressure disorders in the cycle segment at the current moment; DBSCAN clustering is performed on the blood pressure data in the cycle segment at the current moment based on the blood pressure controllability and the adjustment neighborhood radius to obtain the monitoring clustering results; and the blood pressure status of kidney patients is monitored based on the monitoring clustering results.

[0011] Furthermore, the method for obtaining the controllability of blood pressure includes:

[0012] For any sampling moment, the difference between the preset upper limit of blood pressure at the sampling moment and the blood pressure data is taken as the upper limit difference at the sampling moment; the difference between the blood pressure data at the sampling moment and the preset lower limit of blood pressure is taken as the lower limit difference at the sampling moment; the product of the upper limit difference and the lower limit difference is taken as the extreme value difference at the sampling moment;

[0013] Calculate the range of the blood pressure data within the preset neighborhood range at the sampling moment as the numerical range index at the sampling moment; take the difference between the preset blood pressure upper limit value and the preset blood pressure lower limit value as the blood pressure range value; take the ratio of the numerical range index at the sampling moment to the blood pressure range value as the range distribution index at the sampling moment;

[0014] Calculate the variance of all blood pressure data within a preset neighborhood at the sampling moment to obtain a data fluctuation index at the sampling moment;

[0015] According to the range distribution index and data fluctuation index at the sampling time, the neighborhood stability at the sampling time is obtained;

[0016] The controllability of blood pressure at the sampling moment is obtained according to the extreme value difference and neighborhood stability at the sampling moment.

[0017] Furthermore, the method for obtaining the blood pressure disorder degree includes:

[0018] The set of sampling moments at the same position in different periodic segments is regarded as the same-position set;

[0019] Calculate the variance of the blood pressure data corresponding to all sampling moments in the same-position set to obtain the numerical chaos of the same-position set; calculate the variance of the numerical range index of all sampling moments in the same-position set to obtain the distribution chaos of the same-position set;

[0020] Calculate the average value of the blood pressure controllability at all sampling moments in the same position set, perform negative correlation mapping and normalization processing, and obtain the controllable abnormality of the same position set;

[0021] According to the numerical chaos, distribution chaos and controllable abnormality of the same-position set, the overall blood pressure chaos of the same-position set is obtained; the overall blood pressure chaos is used as the blood pressure chaos at each sampling moment in the same-position set.

[0022] Furthermore, the method for obtaining the blood pressure disorder includes:

[0023] For any sampling moment in the cycle segment where the current moment is located, the difference between every two adjacent blood pressure data within the preset neighborhood range of the sampling moment is calculated, and the accumulated value is calculated to obtain the value change degree of the sampling moment;

[0024] Calculate the average value of all blood pressure data within the preset neighborhood range at the sampling moment to obtain the neighborhood mean at the sampling moment; calculate the difference between the neighborhood mean at the sampling moment and the preset blood pressure upper limit value to obtain the upper extreme value difference at the sampling moment; calculate the difference between the neighborhood mean at the sampling moment and the preset blood pressure lower limit value to obtain the lower extreme value difference at the sampling moment; take the minimum value of the upper extreme value difference and the lower extreme value difference at the sampling moment as the extreme value approximation at the sampling moment;

[0025] The blood pressure disorder at the sampling moment is obtained according to the blood pressure disorder degree, value variation degree and extreme value approximation degree at the sampling moment.

[0026] Furthermore, the method for obtaining the adjusted neighborhood radius includes:

[0027] The sampling moment at which the blood pressure imbalance in the cycle segment at the current moment is greater than the preset imbalance threshold is taken as the impact moment;

[0028] Calculate the average value of blood pressure disorders at all influencing moments, perform negative correlation mapping and normalization processing, and obtain the regulation coefficient of the cycle segment at the current moment;

[0029] The adjustment coefficient of the period segment at the current moment is multiplied by the preset initial neighborhood radius to obtain the adjustment neighborhood radius of the period segment at the current moment.

[0030] Furthermore, the blood pressure data in the cycle segment at the current moment are subjected to DBSCAN clustering based on the blood pressure controllability and the adjustment neighborhood radius to obtain a monitoring clustering result, including:

[0031] The difference in blood pressure controllability between every two sampling moments in the cycle segment where the current moment is located is used as the distance measure of clustering; the adjustment neighborhood radius is used as the neighborhood radius of clustering; DBSCAN clustering is performed on the blood pressure data of all sampling moments in the cycle segment where the current moment is located, and the clustering results obtained are recorded as monitoring clustering results.

[0032] Furthermore, the blood pressure status of the kidney disease patient is monitored according to the monitoring clustering result, including:

[0033] In the monitoring clustering results, when the number of blood pressure data in the cluster is less than the preset abnormal number, the blood pressure data in the corresponding cluster is recorded as suspected abnormal data; a single blood pressure data in the monitoring clustering results is recorded as suspected abnormal data;

[0034] When the total amount of suspected abnormal data exceeds the preset alarm threshold, a monitoring alarm is issued.

[0035] Furthermore, the preset neighborhood range is set to be a range with a sampling moment as the center and 5 sampling moments as the side length.

[0036] Furthermore, the length of the cycle segment is one day.

[0037] The present invention provides a patient blood pressure monitoring system for clinical care, comprising a memory and a processor, wherein the processor executes a computing program stored in the memory to implement a patient blood pressure monitoring method for clinical care as described above.

[0038] The present invention has the following beneficial effects:

[0039] The present invention takes into account the volatility of blood pressure in patients with kidney disease. First, the controllability of blood pressure is obtained by analyzing the distribution fluctuation degree of blood pressure data in the neighborhood range at each moment, which reflects the stability of blood pressure. Clustering based on the stability of blood pressure can reduce the instability of the clustering process caused by excessive fluctuations and improve the quality of clustering. Since the neighborhood radius of clustering will affect the judgment of abnormal data, the neighborhood radius of clustering can be further adaptively adjusted according to the possible abnormal conditions of blood pressure in the periodic segment. Considering the overall periodic characteristics of blood pressure data, the distribution of blood pressure data corresponding to the sampling moments at the same position in different periodic segments, as well as the controllability of blood pressure, is used to obtain the blood pressure. Chaos reflects the randomness of blood pressure data in historical periods. When the data is highly random, it indicates that the blood pressure data at the sampling moment is more likely to be abnormal. Combined with the sudden or persistent characteristics of abnormal blood pressure in patients with kidney disease, the blood pressure disorder at each sampling moment is obtained through the blood pressure chaos at the current moment and the blood pressure changes in the neighborhood range, reflecting the possibility of abnormalities characterized by the blood pressure data at each sampling moment. The overall abnormality in the cycle segment is high, and the neighborhood radius is adjusted. DBSCAN clustering is performed through the adjusted neighborhood radius, so that the clustering process is more sensitive to abnormal points, and then blood pressure status monitoring is performed. The present invention performs clustering in combination with blood pressure fluctuations, and adaptively adjusts the neighborhood radius of clustering by analyzing the abnormality of blood pressure data in multiple cycles, improves the sensitivity of clustering to abnormal points, obtains more accurate clustering results, and makes the monitoring results more reliable. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] 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 for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0041] Figure 1 A flow chart of a method for monitoring blood pressure of a patient for clinical care provided by an embodiment of the present invention;

[0042] Figure 2 A flow chart of a method for obtaining blood pressure disorder degree provided by one embodiment of the present invention;

[0043] Figure 3 A schematic diagram of blood pressure data distribution in a cycle segment provided by an embodiment of the present invention;

[0044] Figure 4 A schematic diagram of selecting a preset range of sampling moments provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0045] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of the patient blood pressure monitoring method and system for clinical care proposed by the present invention, its specific implementation method, structure, characteristics and effects, in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0046] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0047] The following is a detailed description of a method and system for monitoring patient blood pressure for clinical care provided by the present invention in conjunction with the accompanying drawings.

[0048] See also Figure 1 , which shows a flow chart of a method for monitoring blood pressure of a patient for clinical care provided by an embodiment of the present invention, the method comprising the following steps:

[0049] S1: Different cycle segments are preset during the monitoring of kidney disease patients to obtain blood pressure data at each sampling moment in the cycle segment; all cycle segments are continuously distributed and have the same length.

[0050] In an embodiment of the present invention, the blood pressure data of a patient with kidney disease needs to be monitored in real time. A cuff is installed on the arm of the patient with kidney disease, and a blood pressure monitoring device is installed to monitor the blood pressure in real time. Since blood pressure is usually affected by people's daily activities, such as being generally higher during the day and lower at night, blood pressure data is usually distributed in a daily cycle, so the length of the preset cycle segment can be set to one day. In other embodiments of the present invention, the length of the preset cycle segment can also be set to two days, as long as the blood pressure data has a periodic distribution characteristic, which is not limited here. The interval between sampling moments in each cycle segment is 15 minutes, and the specific numerical value can be adjusted by the implementer, which is not limited here. It should be noted that blood pressure is divided into systolic pressure and diastolic pressure. Since the data ranges of systolic pressure and diastolic pressure are different, systolic pressure and diastolic pressure are analyzed separately. The blood pressure data in the embodiment of the present invention refers to a kind of blood pressure data, and the analysis method of systolic pressure and diastolic pressure is the same.

[0051] In order to conduct a more comprehensive analysis of the distribution of blood pressure data, the embodiment of the present invention conducts analysis based on multiple periodic segments. The number of preset periodic segments is set to 5. All periodic segments need to be distributed continuously and have the same length. That is, in the embodiment of the present invention, the blood pressure data of the 5 consecutive days before the current moment is used as the analysis data, and each day before the current moment is a different periodic segment. Please refer to Figure 3, which shows a schematic diagram of blood pressure data distribution over a period provided by an embodiment of the present invention.

[0052] At this point, the preparation for blood pressure data analysis is completed.

[0053] S2: Obtain the controllability of blood pressure at each sampling moment based on the numerical distribution and fluctuation of blood pressure data within a preset neighborhood at each sampling moment.

[0054] Considering that the blood pressure of patients with kidney disease is prone to more fluctuations, in the subsequent abnormal monitoring through clustering, clustering only based on the value will cause normal fluctuations to be regarded as abnormal conditions. Therefore, in order to reduce the occurrence of errors, the fluctuation of the data is considered more during clustering, so that the clustering results can fit the blood pressure fluctuation of patients with kidney disease. For the fluctuation analysis at each sampling moment, the local distribution change degree of the blood pressure data is analyzed from the value and fluctuation changes of the overall blood pressure within the sampling time range, and the controllability of the blood pressure at the sampling moment is obtained, which reflects the stability of the local blood pressure data at each sampling moment, so that the subsequent clustering abnormal analysis can reduce the misjudgment of normal stable fluctuations. Therefore, according to the numerical distribution and fluctuation of the blood pressure data within the preset neighborhood range of each sampling moment, the controllability of the blood pressure at each sampling moment is obtained.

[0055] Preferably, for any sampling moment, the difference between the preset upper limit of blood pressure at the sampling moment and the blood pressure data is used as the upper limit difference at the sampling moment, reflecting the difference between the blood pressure data and the upper limit of blood pressure within the preset normal range, and the difference between the blood pressure data at the sampling moment and the preset lower limit of blood pressure is used as the lower limit difference at the sampling moment, reflecting the difference between the blood pressure data and the lower limit of blood pressure within the preset normal range. In the embodiment of the present invention, since the variation ranges of systolic and diastolic pressures are inconsistent, the upper limit of blood pressure and the lower limit of blood pressure are preset according to the specifically analyzed blood pressure data. In general, the systolic pressure range is 90~139mmHg, and the diastolic pressure range is 60~89mmHg. The specific numerical settings can be adjusted by the implementer.

[0056] The product of the upper limit difference and the lower limit difference is further taken as the extreme value difference at the sampling moment. When the blood pressure data is closer to the center of the normal range, the extreme value difference is greater, indicating that the blood pressure stability is better and the controllability is higher at this time.

[0057] In the embodiment of the present invention, the preset neighborhood range is set to be a range with the sampling moment as the center and 5 sampling moments as the side length, that is, the range of each sampling moment including the 2 sampling moments before and after is used as the preset neighborhood range of each sampling moment. Figure 4 , which shows a schematic diagram of a preset range selection of a sampling moment provided by an embodiment of the present invention.

[0058] Calculate the range of blood pressure data within the preset neighborhood range at the sampling moment as the numerical range index at the sampling moment, take the difference between the preset blood pressure upper limit value and the preset blood pressure lower limit value as the blood pressure range value, and take the ratio of the numerical range index at the sampling moment to the blood pressure range value as the range distribution index at the sampling moment. The proportion of the blood pressure range in the neighborhood at the sampling moment in the normal blood pressure range reflects the distribution range of the blood pressure in the neighborhood. The smaller the proportion, the more stable the local distribution of the data at the sampling moment, and the higher the degree to which the blood pressure is stable and controllable.

[0059] The variance of all blood pressure data within the preset neighborhood range at the sampling moment is further calculated to obtain the data fluctuation index at the sampling moment. The variance reflects the complexity of the numerical distribution of the local blood pressure data. The more chaotic the local numerical distribution of the blood pressure data is, the worse the stability and controllability of the blood pressure data at the sampling moment is. According to the range distribution index and data fluctuation index at the sampling moment, the neighborhood stability at the sampling moment is obtained. The stability of the blood pressure data is reflected from the overall distribution change of the neighborhood data. The range distribution index and the data fluctuation index are both negatively correlated with the neighborhood stability. The neighborhood stability is the value after normalization. In the embodiment of the present invention, the expression of the neighborhood stability is:

[0060]

[0061] In the formula, Expressed as The neighborhood stability at the sampling time, Expressed as The maximum value within the preset neighborhood range at the sampling time, Expressed as The minimum value within the preset neighborhood range at the sampling time, Indicates the preset upper limit of blood pressure. Indicates the preset lower limit of blood pressure. Expressed as The total number of sampling moments within the preset neighborhood of the sampling moments, Expressed as The preset neighborhood range of the sampling time Blood pressure data at each sampling moment, Expressed as The average value of the blood pressure data at all sampling moments within a preset neighborhood of a sampling moment. Expressed as an exponential function with a natural constant as base.

[0062] in, It means to calculate the variance of blood pressure data within the preset neighborhood, that is, The data fluctuation index at each sampling moment, Expressed as The numerical range index of the sampling time, Expressed as blood pressure range value, Expressed as From the range distribution index of each sampling moment, it can be determined that the blood pressure range value cannot be zero, so the formula will not have a meaningless situation where the denominator is zero.

[0063] When the data fluctuation index is larger and the range distribution index is larger, it means that the numerical fluctuation degree of the blood pressure data within the preset neighborhood range is large and the fluctuation range is wide, then the controllability at the sampling moment is weaker and the neighborhood stability is lower. The product is multiplied in the form of a negative exponential function to reflect that both the range distribution index and the data fluctuation index are negatively correlated with the neighborhood stability, and the neighborhood stability is the normalized value. In other embodiments of the present invention, other basic mathematical operations can also be used to reflect that both the range distribution index and the data fluctuation index are negatively correlated with the neighborhood stability, such as division or subtraction, etc., which are not limited here.

[0064] Finally, the controllability of blood pressure at the sampling moment is obtained according to the extreme value difference and neighborhood stability at the sampling moment. Combined with the distribution of the values ​​and the distribution change of the local data, the stability and controllability of the blood pressure data at each sampling moment are comprehensively reflected. The extreme value difference and neighborhood stability are positively correlated with the controllability of blood pressure. In the embodiment of the present invention, the expression of blood pressure controllability is:

[0065]

[0066] In the formula, Expressed as The controllability of blood pressure at each sampling moment, Expressed as The neighborhood stability at the sampling time, Indicates the preset upper limit of blood pressure. Indicates the preset lower limit of blood pressure. Expressed as Blood pressure data at each sampling moment.

[0067] in, Expressed as The upper limit difference of the sampling time, Expressed as The lower limit difference of the sampling time, Expressed as The difference of extreme values ​​at each sampling moment is greater, and the greater the extreme value difference, the greater the neighborhood stability, which means that the blood pressure distribution is more stable, the concentrated distribution of local data is more stable, and the blood pressure controllability is greater.

[0068] At this point, the blood pressure data characteristics at each sampling moment are comprehensively characterized based on the numerical distribution of the blood pressure data and the stability of the local data, so as to facilitate the subsequent cluster analysis.

[0069] S3: Obtain the blood pressure disorder degree at each sampling moment according to the blood pressure data distribution disorder degree at the same position at the sampling moment in different cycle segments and the blood pressure controllability at each sampling moment at the same position.

[0070] Since the abnormal state of blood pressure data of kidney patients occurs at an irregular time, and blood pressure in a normal state will show a certain regular distribution, the randomness of blood pressure data at each historical moment can be analyzed through the blood pressure distribution at the corresponding moment in each cycle segment. The higher the randomness of blood pressure at a certain sampling moment, the higher the possibility of abnormal blood pressure at this moment. In the randomness analysis, not only the distribution of blood pressure data at the same position in different cycles is analyzed, but also the local stability represented by the obtained blood pressure controllability is considered. The blood pressure chaos degree is obtained from the local stability and numerical distribution of the same position to characterize the possible degree of abnormality corresponding to each sampling moment.

[0071] Since the blood pressure data distribution between cycle segments has a certain periodicity, the same position is the number of positions of each sampling moment in the cycle. It can be understood that the length of the cycle segment is the same and the sampling method remains unchanged. Therefore, the distance between each sampling moment and the first sampling moment in the cycle represents the position of the sampling moment in the cycle segment. Therefore, according to the degree of blood pressure data distribution chaos at the sampling moments at the same position in different cycle segments and the blood pressure controllability at each sampling moment at the same position, the blood pressure chaos degree at each sampling moment is obtained.

[0072] Preferably, the set of sampling moments at the same position in different periodic segments is taken as a co-location set. In the embodiment of the present invention, the sampling moments in each periodic segment are numbered starting from 1 in chronological order, and the sampling moments with the same number in all periodic segments are taken as a co-location set. The number of sampling moments in a periodic segment is also the number of co-location sets. The blood pressure confusion degree is obtained by distributing the blood pressure data corresponding to the sampling moments in each co-location set. Figure 2 , which shows a flow chart of a method for obtaining blood pressure disorder degree provided by an embodiment of the present invention, the method comprising the following steps:

[0073] S301: Obtaining a numerical chaos degree according to the numerical chaos of the blood pressure data at the sampling time in the same position set.

[0074] By calculating the variance of the blood pressure data corresponding to all sampling moments in the same location set, the numerical chaos of the same location set is obtained. The larger the variance, the more unstable the numerical distribution of the blood pressure data is, and the higher the numerical chaos is.

[0075] In other embodiments of the present invention, the information entropy of the blood pressure data is also calculated as the degree of numerical chaos. The larger the information entropy, the more chaotic the distribution of the blood pressure data is. At this time, the sampling time represented in the same position set is more likely to be the time when an abnormality occurs.

[0076] S302: Obtain the distribution disorder degree according to the local data fluctuation of the blood pressure data at the sampling time in the same position set.

[0077] By calculating the variance of the numerical range index of all sampling moments in the same location set, the distribution disorder of the same location set is obtained. For the sampling moments at the same location, the distribution range of the local blood pressure data is more similar. The variance calculation reflects the degree of disorder of the local blood pressure distribution. The larger the variance, the worse the consistency of the local distribution of the blood pressure data and the higher the distribution disorder.

[0078] In other embodiments of the present invention, the range of the numerical range index at all sampling moments can be calculated as the distribution disorder. The range reflects the degree of local change difference. The larger the range, the higher the difference in the numerical range index, and the higher the possibility of abnormality in the local range at the sampling moment.

[0079] S303: Obtain the controllable abnormality degree according to the stability and controllability of the blood pressure data at the sampling time in the same position set.

[0080] By calculating the average value of the blood pressure controllability at all sampling moments in the same position set, performing negative correlation mapping and normalization processing, the controllable abnormality of the same position set is obtained. The smaller the average blood pressure controllability at the sampling moment, the smaller the overall controllable stability of the blood pressure at the same sampling moment, and the position may have an abnormal state, so the controllable abnormality is higher.

[0081] S304: Obtain the overall blood pressure disorder of the same-position set according to the numerical disorder, distribution disorder and controllable abnormality; and use the overall blood pressure disorder as the blood pressure disorder at each sampling moment in the same-position set.

[0082] According to the numerical chaos, distribution chaos and controllable abnormality of the same position set, the overall blood pressure chaos of the same position set is obtained, and the blood pressure chaos distribution is obtained by integrating the blood pressure data of all sampling moments at the same position. The numerical chaos, distribution chaos and controllable abnormality are all positively correlated with the overall blood pressure chaos. The overall blood pressure chaos of the same position set is the blood pressure chaos at each sampling moment in the same position set. In the embodiment of the present invention, the expression of the overall blood pressure chaos is:

[0083]

[0084] In the formula, Expressed as The blood pressure disorder at each sampling moment, Expressed as The total number of sampling moments in the same location set at the sampling moment, Expressed as The first Blood pressure data at each sampling moment, Expressed as The average value of the blood pressure data at all sampling times in the same location set at the sampling time, Shown as The first The numerical range index of the sampling time, Expressed as The average value of the value range index of all sampling moments in the same location set at the sampling moment, Expressed as The first The controllability of blood pressure at each sampling moment, Expressed as an exponential function with a natural constant as base.

[0085] in, Expressed as The numerical chaos of the set of samples at the same position at the sampling moment, Expressed as The distribution disorder of the set of samples at the same location at the sampling time is Expressed as The controllable abnormality of the same position set at each sampling moment is as follows: when the numerical chaos, distribution chaos and controllable abnormality are greater, it means that the blood pressure data at the sampling moment in the same position set is more random, and the possibility of abnormality is higher, so the blood pressure chaos is higher. The numerical chaos is reflected in the form of multiplication, and the distribution chaos and controllable abnormality are both positively correlated with the blood pressure chaos. In other embodiments of the present invention, other basic mathematical operations can be used to reflect the numerical chaos, distribution chaos and controllable abnormality are both positively correlated with the blood pressure chaos, such as addition or power operation, etc., which are not limited here.

[0086] At this point, based on the analysis of multiple cycle segments, the randomness of the blood pressure data at the sampling moment at the same position in the cycle is analyzed, and the blood pressure disorder degree is obtained, reflecting the possible abnormality of the blood pressure data at the sampling moment.

[0087] S4: Obtain the blood pressure disorder at each sampling moment in the cycle segment at the current moment according to the blood pressure disorder degree at each sampling moment in the cycle segment at the current moment and the data change degree of the blood pressure data within a preset neighborhood range at each sampling moment.

[0088] There are various states of abnormal blood pressure data of patients with kidney disease. For example, although the blood pressure does not exceed the positive and negative range, the blood pressure continues to be high, or there are obvious ups and downs in a short period of time. Another example is the chaos of normal changes in blood pressure data such as nocturnal hypertension. These abnormal conditions will cause further damage to the kidneys. Therefore, in the analysis of abnormal blood pressure, not only the blood pressure chaos at the sampling time is analyzed, but also the abnormality is analyzed from the changes in blood pressure data in the local time. The blood pressure disorder is obtained by combining the chaotic random conditions at the time and the data changes in the local range, so as to more comprehensively characterize the abnormal blood pressure at the time, so as to facilitate the subsequent screening of clustering abnormalities.

[0089] Preferably, for any sampling moment in the cycle segment in which the current moment is located, the difference between each two adjacent blood pressure data within the preset neighborhood range of the sampling moment is calculated, and the cumulative value is calculated to obtain the numerical change degree of the sampling moment. The difference cumulative value of the blood pressure data reflects the degree of change of the data within the neighborhood range. The higher the degree of change, the more obvious the abnormality.

[0090] Furthermore, the average value of all blood pressure data within the preset neighborhood range at the sampling moment is calculated to obtain the neighborhood mean at the sampling moment, reflecting the overall distribution of blood pressure data in the neighborhood range. The difference between the neighborhood mean at the sampling moment and the preset upper limit of blood pressure is calculated to obtain the upper extreme difference at the sampling moment. The difference between the neighborhood mean at the sampling moment and the preset lower limit of blood pressure is calculated to obtain the lower extreme difference at the sampling moment. The minimum value of the upper extreme difference and the lower extreme difference at the sampling moment is used as the extreme approximation at the sampling moment. The degree of proximity of the overall distribution trend to the normal range reflects the possibility of abnormality in the overall blood pressure data distribution. The closer the blood pressure data is to the boundary data of the normal range, the more attention should be paid to the abnormal characterization of the blood pressure data.

[0091] Finally, the blood pressure disorder at the sampling moment is obtained according to the blood pressure disorder, value variation and extreme value approximation at the sampling moment, and the blood pressure disorder reflects the abnormal characterization degree of the blood pressure at each sampling moment in the current cycle segment through the blood pressure disorder. The blood pressure disorder and value variation are both positively correlated with the blood pressure disorder, and the extreme value approximation is negatively correlated with the blood pressure disorder. The blood pressure disorder is the value after normalization. In the embodiment of the present invention, the expression of blood pressure disorder is:

[0092]

[0093] In the formula, Expressed as Blood pressure disorder at each sampling moment, Expressed as The blood pressure disorder at each sampling moment, Indicates the preset upper limit of blood pressure. Indicates the preset lower limit of blood pressure. Expressed as The total number of sampling moments within the preset neighborhood of the sampling moments, Expressed as The preset neighborhood range of the sampling time Blood pressure data at each sampling moment, Expressed as The preset neighborhood range of the sampling time Blood pressure data at each sampling moment, Expressed as The neighborhood mean at sampling time, is represented as a minimum extraction function, Expressed as an absolute value extraction function, It is represented as a preset adjustment coefficient, which is set to 0.001 in the embodiment of the present invention, and its purpose is to prevent the denominator from being zero, making the formula meaningless. It is represented as a normalization processing function. It should be noted that normalization is a technical means well known to those skilled in the art. The normalization function can be linear normalization or standard normalization, etc. The specific normalization method is not limited here.

[0094] in, Expressed as The value change at each sampling moment, Expressed as The difference of the upper extreme value at each sampling moment, Expressed as The difference of the lower extreme value at each sampling moment, Expressed as The extreme value approximation of each sampling moment. The blood pressure chaos and value variation are positively correlated with blood pressure disorder in the form of ratio, and the extreme value approximation is negatively correlated with blood pressure disorder. The greater the value variation, the more significant the local change at the sampling moment. The greater the blood pressure chaos, the higher the randomness of the blood pressure data at the sampling moment. The smaller the extreme value approximation, the closer the local data at the sampling moment is to the range boundary, which reflects that the abnormal situation at the sampling moment is more obvious and the blood pressure disorder is higher.

[0095] At this point, based on the local data characteristics at each sampling moment in the cycle segment where the current moment is located and the blood pressure disorder degree, the blood pressure disorder reflecting the abnormal degree is obtained.

[0096] S5: Obtain the adjustment neighborhood radius through the numerical distribution of all blood pressure disorders in the cycle segment at the current moment; perform DBSCAN clustering on the blood pressure data in the cycle segment at the current moment based on the blood pressure controllability and the adjustment neighborhood radius to obtain the monitoring clustering result; monitor the blood pressure status of kidney patients based on the monitoring clustering result.

[0097] According to the degree of abnormality in the data, the neighborhood radius of DBSCAN clustering is adjusted to screen out more abnormal situations in the clustering process. Therefore, the neighborhood radius is adjusted through the numerical distribution of all blood pressure disorders in the cycle segment at the current moment, thereby realizing adaptive adjustment of the neighborhood radius.

[0098] Preferably, the sampling moment at which the blood pressure imbalance in the cycle segment of the current moment is greater than the preset imbalance threshold is used as the influencing moment. When the blood pressure imbalance is greater than the preset imbalance threshold, it means that the abnormality of the blood pressure data at the corresponding sampling moment is more significant, and then adjustments are made according to the influencing moment when the abnormality is more significant. In the embodiment of the present invention, the preset imbalance threshold is set to 0.7, and the specific value can be adjusted by the implementer.

[0099] The average value of the blood pressure disorder at all the influencing moments is further calculated, negative correlation mapping is performed and normalized, and the adjustment coefficient of the cycle segment at the current moment is obtained. When the overall blood pressure disorder at the influencing moment is higher, it means that the blood pressure changes in the abnormal situation in the cycle segment are more, and the degree to which the neighborhood radius needs to be adjusted is higher, so the adjustment coefficient is larger. Therefore, the product of the adjustment coefficient of the cycle segment at the current moment and the preset initial neighborhood radius is obtained to obtain the adjustment neighborhood radius of the cycle segment at the current moment. In the embodiment of the present invention, the expression for adjusting the neighborhood radius is:

[0100]

[0101] In the formula, Represented as the adjusted neighborhood radius, Expressed as the total number of impact moments, Expressed as Blood pressure disorders that affect the moment, Represented as the preset initial neighborhood radius, is expressed as a hyperbolic tangent function. Where, Expressed as adjustment coefficient.

[0102] When the abnormal performance in the current cycle segment is analyzed to be more significant, the neighborhood radius is adjusted smaller during clustering to increase the sensitivity of clustering to abnormal points. In the embodiment of the present invention, the preset neighborhood radius is set according to the specific implementation situation and is not limited here. After completing the adjustment of the neighborhood radius, DBSCAN clustering can be performed on the blood pressure data in the cycle segment at the current moment based on the blood pressure controllability and the adjusted neighborhood radius to obtain the monitoring clustering result.

[0103] In one embodiment of the present invention, the difference in blood pressure controllability between every two sampling moments in the cycle segment where the current moment is located is used as the distance metric for clustering. The distance metric determines the similarity and difference of the data in the clustering process and is a key judgment influencing factor in the clustering algorithm. The greater the difference in blood pressure controllability, the greater the distance metric, indicating that the similarity between the data is lower. The adjusted neighborhood radius is used as the neighborhood radius for clustering, and DBSCAN clustering is performed on the blood pressure data at all sampling moments in the cycle segment where the current moment is located. The clustering results obtained are recorded as monitoring clustering results, and the monitoring clustering results include single blood pressure data and cluster clusters that are not clustered. It should be noted that the specific clustering method of DBSCAN clustering is a technical means well known to those skilled in the art and will not be further elaborated here.

[0104] Finally, the blood pressure status of the kidney patients is monitored according to the monitoring clustering results. In the embodiment of the present invention, in the monitoring clustering results, when the number of blood pressure data in the clustering cluster is less than the preset abnormal number, it means that the clustering cluster may be the clustering cluster corresponding to the abnormal situation, and the blood pressure data in the corresponding clustering cluster is recorded as suspected abnormal data, and the single blood pressure data in the monitoring clustering results is recorded as suspected abnormal data, and the blood pressure data that is not clustered is also the data of the abnormal situation. When the total number of suspected abnormal data is greater than the preset alarm threshold, it means that the abnormal state is significant and the monitoring alarm is performed. In the embodiment of the present invention, the preset abnormal number is set to 4, and the preset alarm threshold is set to 7. The specific numerical value implementer can adjust it according to the specific implementation situation, which is not limited here. It should be noted that due to the certainty of historical moments and cycle segments, the newly collected blood pressure data in the monitoring process only makes different cycle segments move back one moment, so only the blood pressure controllability and blood pressure disorder analysis at the current sampling moment are performed. If the impact moment is not added or reduced, the neighborhood radius does not need to be readjusted to improve the efficiency of real-time monitoring.

[0105] It should be noted that, in order to facilitate calculation, all indicator data involved in the calculation in the embodiment of the present invention are preprocessed to eliminate the dimension effect. The specific means of eliminating the dimension effect are technical means well known to those skilled in the art and are not limited here.

[0106] In summary, the present invention takes into account the volatility of blood pressure in patients with kidney disease. First, the controllability of blood pressure is obtained by the distribution fluctuation degree of blood pressure data in the neighborhood range at each moment, which reflects the stability of blood pressure. Clustering based on the stability of blood pressure can reduce the instability of the clustering process caused by excessive fluctuations and improve the quality of clustering. Since the neighborhood radius of clustering will affect the judgment of abnormal data, the neighborhood radius of clustering can be further adaptively adjusted according to the possible abnormal conditions of blood pressure in the periodic segment. Considering the overall periodic characteristics of blood pressure data, the distribution of blood pressure data corresponding to the sampling moments at the same position in different periodic segments is used to obtain the controllability of blood pressure. And the controllability of blood pressure, the blood pressure disorder is obtained, which reflects the randomness of blood pressure data in the historical period. When the data randomness is strong, it means that the blood pressure data at the sampling moment is likely to be abnormal. Combined with the sudden change or persistence characteristics of kidney patients when blood pressure is abnormal, the blood pressure disorder at each sampling moment is obtained through the blood pressure disorder at the current moment and the blood pressure changes in the neighborhood range, reflecting the abnormal possibility characterized by the blood pressure data at each sampling moment. Through the overall abnormality in the cycle segment, the neighborhood radius is adjusted to make the DBSCAN clustering more sensitive to abnormal points, and then the blood pressure state is monitored. The present invention performs clustering in combination with the blood pressure fluctuation situation, and adaptively adjusts the neighborhood radius of the clustering by analyzing the abnormality of blood pressure data in multiple cycles, improves the sensitivity of clustering to abnormal points, obtains more accurate clustering results, and makes the monitoring results more reliable.

[0107] The present invention provides a patient blood pressure monitoring system for clinical care, comprising a memory and a processor, wherein the processor executes a computing program stored in the memory to implement a patient blood pressure monitoring method for clinical care as described above.

[0108] It should be noted that the sequence of the above embodiments of the present invention is only for description and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying 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.

[0109] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

Claims

1. A method for monitoring blood pressure of a patient for clinical nursing, characterized in that: The method comprises: Different cycle segments are preset during the monitoring of kidney disease patients to obtain blood pressure data at each sampling moment in the cycle segment; all cycle segments are continuously distributed and have the same length; According to the numerical distribution and fluctuation of blood pressure data within a preset neighborhood range at each sampling moment, the controllability of blood pressure at each sampling moment is obtained; According to the blood pressure data distribution disorder degree at the same position at the sampling moment in different period segments and the blood pressure controllability at each sampling moment at the same position, the blood pressure disorder degree at each sampling moment is obtained; According to the blood pressure disorder degree at each sampling moment in the cycle segment at the current moment and the data change degree of the blood pressure data within the preset neighborhood range at each sampling moment, the blood pressure disorder at each sampling moment in the cycle segment at the current moment is obtained; The adjustment neighborhood radius is obtained through the numerical distribution of all blood pressure disorders in the cycle segment at the current moment; the blood pressure data in the cycle segment at the current moment are clustered by DBSCAN based on the blood pressure controllability and the adjustment neighborhood radius to obtain the monitoring clustering result; the blood pressure status of the kidney patients is monitored according to the monitoring clustering result; The method for obtaining the controllability of blood pressure includes: For any sampling moment, the difference between the preset upper limit of blood pressure at the sampling moment and the blood pressure data is taken as the upper limit difference at the sampling moment; the difference between the blood pressure data at the sampling moment and the preset lower limit of blood pressure is taken as the lower limit difference at the sampling moment; the product of the upper limit difference and the lower limit difference is taken as the extreme value difference at the sampling moment; Calculate the range of the blood pressure data within the preset neighborhood range at the sampling moment as the numerical range index at the sampling moment; take the difference between the preset blood pressure upper limit value and the preset blood pressure lower limit value as the blood pressure range value; take the ratio of the numerical range index at the sampling moment to the blood pressure range value as the range distribution index at the sampling moment; Calculate the variance of all blood pressure data within a preset neighborhood at the sampling moment to obtain a data fluctuation index at the sampling moment; According to the range distribution index and data fluctuation index at the sampling time, the neighborhood stability at the sampling time is obtained; According to the extreme value difference and neighborhood stability at the sampling moment, the controllability of the blood pressure at the sampling moment is obtained; The method for obtaining the blood pressure disorder degree comprises: The set of sampling moments at the same position in different periodic segments is regarded as the same-position set; Calculate the variance of the blood pressure data corresponding to all sampling moments in the same-position set to obtain the numerical chaos of the same-position set; calculate the variance of the numerical range index of all sampling moments in the same-position set to obtain the distribution chaos of the same-position set; Calculate the average value of the blood pressure controllability at all sampling moments in the same position set, perform negative correlation mapping and normalization processing, and obtain the controllable abnormality of the same position set; According to the numerical chaos, distribution chaos and controllable abnormality of the same-position set, the overall blood pressure chaos of the same-position set is obtained; the overall blood pressure chaos is used as the blood pressure chaos at each sampling moment in the same-position set.

2. A method for monitoring blood pressure of a patient for clinical care according to claim 1, characterized in that: The method for obtaining the blood pressure disorder comprises: For any sampling moment in the cycle segment where the current moment is located, the difference between every two adjacent blood pressure data within the preset neighborhood range of the sampling moment is calculated, and the accumulated value is calculated to obtain the value change degree of the sampling moment; Calculate the average value of all blood pressure data within the preset neighborhood range at the sampling moment to obtain the neighborhood mean at the sampling moment; calculate the difference between the neighborhood mean at the sampling moment and the preset blood pressure upper limit value to obtain the upper extreme value difference at the sampling moment; calculate the difference between the neighborhood mean at the sampling moment and the preset blood pressure lower limit value to obtain the lower extreme value difference at the sampling moment; take the minimum value of the upper extreme value difference and the lower extreme value difference at the sampling moment as the extreme value approximation at the sampling moment; The blood pressure disorder at the sampling moment is obtained according to the blood pressure disorder degree, value variation degree and extreme value approximation degree at the sampling moment.

3. A method for monitoring blood pressure of a patient for clinical care according to claim 1, characterized in that: The method for obtaining the adjusted neighborhood radius includes: The sampling moment at which the blood pressure imbalance in the cycle segment at the current moment is greater than the preset imbalance threshold is taken as the impact moment; Calculate the average value of blood pressure disorders at all influencing moments, perform negative correlation mapping and normalization processing, and obtain the regulation coefficient of the cycle segment at the current moment; The adjustment coefficient of the period segment at the current moment is multiplied by the preset initial neighborhood radius to obtain the adjustment neighborhood radius of the period segment at the current moment.

4. A method for monitoring blood pressure of a patient for clinical care according to claim 1, characterized in that: The blood pressure data in the cycle segment at the current moment are clustered by DBSCAN based on the blood pressure controllability and the adjustment neighborhood radius to obtain the monitoring clustering results, including: The difference in blood pressure controllability between every two sampling moments in the cycle segment where the current moment is located is used as the distance measure of clustering; the adjustment neighborhood radius is used as the neighborhood radius of clustering; DBSCAN clustering is performed on the blood pressure data of all sampling moments in the cycle segment where the current moment is located, and the clustering results obtained are recorded as monitoring clustering results.

5. A method for monitoring blood pressure of a patient for clinical care according to claim 1, characterized in that: The blood pressure status of the kidney disease patient is monitored according to the monitoring clustering result, including: In the monitoring clustering results, when the number of blood pressure data in the cluster is less than the preset abnormal number, the blood pressure data in the corresponding cluster is recorded as suspected abnormal data; a single blood pressure data in the monitoring clustering results is recorded as suspected abnormal data; When the total amount of suspected abnormal data exceeds the preset alarm threshold, a monitoring alarm is issued.

6. A method for monitoring blood pressure of a patient for clinical care according to claim 1, characterized in that: The preset neighborhood range is set to be a range with the sampling moment as the center and 5 sampling moments as the side length.

7. A method for monitoring blood pressure of a patient for clinical care according to claim 1, characterized in that: The length of the cycle segment is one day.

8. A patient blood pressure monitoring system for clinical care, comprising a memory and a processor, characterized in that: The processor executes the computing program stored in the memory to implement a patient blood pressure monitoring method for clinical care as described in any one of claims 1-7.

Images