Data acquisition and integration method for adaptively capturing critical core monitoring cluster

Abstract

The invention discloses a data acquisition and integration method for adaptively capturing a critical core monitoring cluster, and relates to the technical field of medical monitoring, and the method comprises the steps: 1, evaluating a change index and a change correlation index of core monitoring data, and dynamically adjusting the acquisition frequency of the core monitoring data according to the change index and the change correlation index; 2, collecting core monitoring data according to the adjusted collection frequency, and packaging the collected core monitoring data by considering the network condition and combining the adjusted collection frequency to obtain a core monitoring data packet; and step 3, sending the core monitoring data packet to a monitoring terminal. According to the data acquisition and integration method for adaptively capturing the critical core monitoring cluster, the data acquisition frequency and the packaging strategy can be dynamically adjusted according to the change condition of the monitoring data and the network condition, and efficient and accurate transmission of the core monitoring data is ensured.

Description

Technical Field

[0001] The present invention relates to the field of medical monitoring technology, and in particular to a data collection and integration method for adaptively capturing a critical care core monitoring cluster. Background Art

[0002] Traditional data collection methods in critical care often use fixed sampling rates and data packet sizes. Although this method is simple and easy to implement, it has many limitations. First, the fixed sampling rate cannot be dynamically adjusted according to the patient's actual physiological state, resulting in a large amount of redundant data when the patient's condition is stable, and critical information may be missed when the condition suddenly changes. Second, the fixed data packet size may cause data loss or transmission delays when the network conditions are poor, affecting the timely judgment of the patient's condition. Summary of the invention

[0003] In view of the defects in the prior art, the present invention proposes a data collection and integration method for adaptively capturing critical core monitoring clusters. The method can dynamically adjust the data collection frequency and packaging strategy according to the changes in monitoring data and network conditions, to ensure efficient and accurate transmission of core monitoring data.

[0004] The technical solution of the present invention is a data collection and integration method for adaptively capturing a critical core monitoring cluster, comprising: Step 1, evaluating the change index and change correlation index of the core monitoring data, and dynamically adjusting the collection frequency of the core monitoring data according to the change index and change correlation index; Step 2: collecting core monitoring data according to the adjusted collection frequency, taking into account the network conditions, and packaging the collected core monitoring data in combination with the adjusted collection frequency to obtain a core monitoring data packet; Step 3: Send the core monitoring data packet to the monitoring terminal.

[0005] Preferably, before step 1, the following steps are included: Step 001, setting a collection threshold for each type of monitoring data, wherein the collection threshold is used to trigger the collection of core monitoring data; Step 002, obtaining historical monitoring data of various monitoring data, making predictions based on the historical monitoring data to obtain monitoring prediction values, calculating the difference between the monitoring prediction values ​​and the monitoring actual values, and dynamically adjusting the collection threshold according to the difference; Step 003: When the monitoring data is not within the adjusted collection threshold, the monitoring data is determined as core monitoring data and the collection of core monitoring data is triggered.

[0006] Preferably, in step 1, the method for evaluating the change index includes: Step 01, determining the length of a sliding window, wherein the sliding window is used to calculate the time correlation of the core monitoring data at each time point; Step 02, for each time point in the historical core monitoring data, calculate the mean and standard deviation of the data in the sliding window; Step 03, within the sliding window, for each core monitoring data, calculate the product of the standard deviation value of the core monitoring data with the current core monitoring data, and sum them up to obtain the time correlation index, calculate the ratio of the second-order derivative of the data to the first-order derivative, determine the data change rate, and determine the change index based on the data change rate and the time correlation coefficient.

[0007] Preferably, in step 1, the method for evaluating the change association index comprises: Step 04, sort the change indexes of various core monitoring data in descending order, and perform normalization processing to obtain a change index sequence; Step 05, calculating the correlation coefficient between two change indexes in the change index sequence; Step 06, constructing an association graph based on the change index and the correlation coefficient. The association graph takes the core monitoring data as points and determines the edge by the difference between the correlation coefficient and the preset coefficient threshold. When the difference is greater than zero, there is an edge. The correlation coefficient is corrected according to the association graph to obtain the change correlation index.

[0008] Preferably, the correlation coefficient is corrected according to the association diagram to obtain a formula for the change association index: In the formula, represents the change correlation index, Indicates The correlation coefficient of the core monitoring data, Indicates the total number of core monitoring data. Indicates The core monitoring data of this class has edge connections The correlation coefficient of the core monitoring data, Indicates The core monitoring data of this class has edge connections The total number of core monitoring data of the class.

[0009] Preferably, the sampling parameters include a collection frequency, and the formula for dynamically adjusting the sampling parameters of the core monitoring data according to the change index and the change correlation index is: In the formula, represents the current time t The frequency of collecting core monitoring data, Indicates the previous time The frequency of collecting core monitoring data, Indicates The change index of the core monitoring data of the class, Indicates Importance weight of core monitoring data, is a constant coefficient.

[0010] Preferably, in step 2, the collected core monitoring data is packaged in consideration of the network conditions and in combination with the adjusted collection frequency, including: Step 021, dynamically adjusting the size of the encoding packet according to the current network bandwidth, the average network bandwidth, the adjusted acquisition frequency and the maximum acquisition frequency; Step 022, determining compression parameters according to the adjusted acquisition frequency and encoding packet size, compressing the core monitoring data according to the compression parameters, and performing random linear encoding on the compressed data to obtain a core monitoring data packet.

[0011] Preferably, in step 021, the formula for dynamically adjusting the size of the encoding packet according to the current network bandwidth, the average network bandwidth, the adjusted acquisition frequency and the maximum acquisition frequency is: In the formula, represents the coded packet size of the i-th type of core monitoring data packet at the current time t, Indicates the size of the encoding packet of the i-th type of core monitoring data at the previous time. represents the current time t The collection frequency of the core monitoring data, that is, the adjusted collection frequency, Indicates the preset maximum sampling frequency of the i-th type of core monitoring data, Indicates the current network bandwidth. represents the average network bandwidth, , , are all constant coefficients.

[0012] The data collection and integration method for adaptively capturing the critical core monitoring cluster proposed in the present invention has shown significant superiority and practicality compared with traditional technologies, which is specifically reflected in the following aspects: 1) Efficient data collection: By dynamically evaluating the change index and change correlation index of the core monitoring data, the present invention can accurately adjust the data collection frequency according to the patient's actual core monitoring data. This means that when the monitoring data is stable, data collection can be automatically reduced, thereby effectively avoiding data redundancy; and when the patient's core monitoring data fluctuates or emerges, the sampling rate can be quickly increased to ensure that key information is captured in a timely manner, providing strong support for the doctor's accurate diagnosis.

[0013] 2) Intelligent data packaging: This invention fully considers the impact of network conditions on data transmission, and realizes intelligent data packaging by dynamically adjusting the encoding packet size and compression parameters. This method not only improves the efficiency of data transmission, but also significantly reduces data loss and transmission delay caused by network congestion or bandwidth limitations, ensuring the integrity and real-time nature of core monitoring data.

[0014] 3) Maximizing resource utilization: Through precise data collection and intelligent data packaging, the present invention effectively reduces unnecessary data processing and transmission, thereby reducing computing overhead and network resource usage. This enables medical monitoring to operate more efficiently and provide high-quality monitoring services for more critically ill patients. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following is a brief introduction to the drawings required for the specific embodiments or the description of the prior art. In all the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, the elements or parts are not necessarily drawn according to the actual scale.

[0016] Figure 1 A flow chart of a method for adaptively capturing data collection and integration of a critical care core monitoring cluster provided in Example 1 of the present invention; Figure 2 This is a flowchart of the method before step 1 provided in Example 1 of the present invention. DETAILED DESCRIPTION

[0017] The following embodiments of the technical solution of the present invention are described in detail in conjunction with the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and are therefore only used as examples, and cannot be used to limit the protection scope of the present invention.

[0018] It should be noted that, unless otherwise specified, the technical terms or scientific terms used in this application should have the common meanings understood by those skilled in the art to which the present invention belongs.

[0019] Example 1 like Figure 1 As shown, the present invention provides a data collection and integration method for adaptively capturing a critical core monitoring cluster, comprising: Step 1: evaluate the change index and change correlation index of the core monitoring data, and dynamically adjust the collection frequency of the core monitoring data according to the change index and change correlation index.

[0020] Specifically, various medical equipment in the intensive care unit (such as electrocardiographs, blood pressure monitors, blood oxygen saturation monitors, etc.) are used to collect core monitoring data of critically ill patients in real time, and the collected data are preprocessed, including data cleaning (such as removing outliers, processing missing values, etc.), data formatting (such as unifying timestamp format, data units, etc.), etc., to ensure the availability of the data and the accuracy of subsequent analysis.

[0021] After the above processing, the core monitoring data is subjected to change analysis, wherein the change analysis includes change index and change correlation index analysis. The change index analysis mainly focuses on the data change trend in the time dimension.

[0022] Specifically, in step 1, the evaluation method of the change index includes: step 01, determining the length of a sliding window, the sliding window is used to calculate the time correlation of the core monitoring data at each time point; step 02, for each time point in the historical core monitoring data, calculating the mean and standard deviation of the data in the sliding window; step 03, in the sliding window, for each core monitoring data, calculate the product of the core monitoring data and the standard deviation of the current core monitoring data, and sum them to obtain the time correlation index, calculate the ratio of the second-order derivative of the data to the first-order derivative, determine the data change rate, and determine the change index based on the data change rate and the time correlation coefficient.

[0023] The sliding window is the basis for calculating the time correlation of the core monitoring data at each time point. By setting an appropriate sliding window length, the changing trend of the data in the short term can be captured, while avoiding the computational complexity and memory consumption caused by too long a window. The length of the sliding window can be set according to actual needs, usually based on the analysis of historical data and experience. For example, for heart rate data, a window length of several minutes may be selected to be able to capture changes in heart rate in a short period of time.

[0024] For the time correlation coefficient, the time correlation index reflects the correlation of the data in time, that is, the degree of correlation between the current data point and other data points in the sliding window. In the sliding window, for each core monitoring data, calculate the product of its standard deviation with the current core monitoring data. Sum these products to get the time correlation index. The larger this index is, the higher the degree of correlation between the current data point and other data points in the sliding window.

[0025] For the data change rate, the data change rate reflects the speed and direction of data change, that is, whether the data is accelerating, decelerating, or remaining stable. Calculate the ratio of the second-order derivative of the data (that is, the rate of change of the data change rate) to the first-order derivative (that is, the rate of change of the data). This ratio can reflect the speed and trend of data change. For example, when the ratio is large, it means that the data changes faster; when the ratio is small, it means that the data changes more steadily.

[0026] The change index is a composite index that combines the time correlation index and the data change rate, and is used to quantify the degree of change in the data. According to the time correlation index and the data change rate, a calculation formula or rule is set to determine the change index. For example, the time correlation index and the data change rate can be weighted and summed to obtain the change index.

[0027] Through the above steps, the change index of the core monitoring data at each time point can be evaluated. This index not only takes into account the temporal relevance of the data, but also the speed and direction of data change, so it can more comprehensively reflect the changes in the data.

[0028] Furthermore, in step 1, the method for evaluating the change correlation index includes: step 04, sorting the change index of each type of core monitoring data in descending order, and normalizing them to obtain a change index sequence; step 05, calculating the correlation coefficient between two change indexes in the change index sequence; step 06, constructing a correlation graph based on the change index and the correlation coefficient, the correlation graph takes the core monitoring data as points, and determines the edge by the difference between the correlation coefficient and the preset coefficient threshold. When the difference is greater than zero, there is an edge, and the correlation coefficient is corrected according to the correlation graph to obtain the change correlation index.

[0029] In order to more clearly compare the degree of change of different core monitoring data and eliminate dimensional differences, it is necessary to sort and normalize the change indexes of various core monitoring data. Specifically, first, sort the change indexes of all core monitoring data from large to small. In this way, you can intuitively see which data have changed most dramatically. Then, normalize the sorted change indexes. Normalization is the process of scaling data to a specific range (usually 0 to 1), which eliminates dimensional differences between different data and allows them to be compared on the same scale. Normalization can be performed using linear functions, logarithmic functions, and other methods.

[0030] Calculate the correlation coefficient between two change indices in the change index sequence to quantify the degree of linear correlation between them. Specifically, you can select a suitable type of correlation coefficient, such as Pearson correlation coefficient, Spearman rank correlation coefficient, etc. Use the selected correlation coefficient formula to calculate the correlation coefficient between each pair of change indices in the change index sequence. The correlation coefficient value is between -1 and 1. The closer the value is to 1, the stronger the positive correlation between the two variables is. The closer the value is to -1, the stronger the negative correlation is. The closer the value is to 0, the more there is no linear correlation between the two variables.

[0031] There may be nonlinear relationships and interactions between the core monitoring data. These complex relationships are often ignored when directly calculating the correlation coefficient. Therefore, the correlation coefficient needs to be corrected to make it more specific. Specifically, each core monitoring data is used as a point in the association graph. The edge is determined based on the difference between the correlation coefficient and the preset coefficient threshold. If the difference between the correlation coefficient between two change indices and the preset coefficient threshold is greater than zero, an edge is drawn between them. In this way, the association graph shows which core monitoring data have significant correlations. Then, the correlation is corrected according to the following formula.

[0032] Specifically, the correlation coefficient is corrected according to the correlation diagram to obtain the formula of the change correlation index: In the formula, represents the change correlation index, Indicates The correlation coefficient of the core monitoring data, Indicates the total number of core monitoring data. Indicates The core monitoring data of this class has edge connections The correlation coefficient of the core monitoring data, Indicates The core monitoring data of this class has edge connections The total number of core monitoring data of the class.

[0033] It should be noted that the above correction only considers the impact of the correlation coefficients of other core monitoring data that have edge connections with the core monitoring data whose correlation coefficient is to be corrected. In actual application, the correlation coefficients of the core monitoring data that have edge connections with other core monitoring data will also affect the correction of the correlation coefficient of the corrected core monitoring data. At the same time, considering the computational cost, the embodiment of the present invention may consider two orders, that is, other core monitoring data that have two-hop connections with the core monitoring data whose correlation coefficient is to be corrected all have edge connections.

[0034] It should also be noted that, in some embodiments, the correlation coefficients of all edge connection relationships can be calculated according to the above formula. In other embodiments, the correlation coefficients of direct edge connection relationships can be corrected using the correlation coefficients of indirect edge connection relationships according to the idea of ​​the above formula, and then the corrected correlation coefficients of direct edge connection relationships can be corrected using the corrected correlation coefficients of direct edge connection relationships according to the idea of ​​the above formula.

[0035] Through correction, the complex relationships among data can be considered more comprehensively, and the accuracy and reliability of analysis results can be improved.

[0036] In the embodiment of the present invention, the sampling parameters include the acquisition frequency, and the formula for dynamically adjusting the sampling parameters of the core monitoring data according to the change index and the change correlation index is: In the formula, represents the current time t The frequency of collecting core monitoring data, Indicates the previous time The frequency of collecting core monitoring data, Indicates The change index of the core monitoring data of the class, Indicates Importance weight of core monitoring data, is a constant coefficient.

[0037] Step 2: collect core monitoring data according to the adjusted collection frequency, consider the network situation, and package the collected core monitoring data in combination with the adjusted collection frequency to obtain a core monitoring data packet.

[0038] In an embodiment of the present invention, in step 2, the network conditions are taken into consideration, and the collected core monitoring data is packaged in combination with the adjusted acquisition frequency, including: step 021, dynamically adjusting the encoding packet size according to the current network bandwidth, the average network bandwidth, the adjusted acquisition frequency and the maximum acquisition frequency; step 022, determining the compression parameters according to the adjusted acquisition frequency and the encoding packet size, compressing the core monitoring data according to the compression parameters, and performing random linear encoding on the compressed data to obtain a core monitoring data packet.

[0039] Specifically, first, you need to obtain the current network bandwidth and average network bandwidth information. This can be achieved through network monitoring tools or APIs. You also need to consider the adjusted acquisition frequency and maximum acquisition frequency. The acquisition frequency determines the speed at which data is generated, while the maximum acquisition frequency is a limit value used to ensure that data is not sent at a speed that exceeds the network processing capacity. Based on the above information, dynamically adjust the size of each encoding packet. When the network bandwidth is large and the acquisition frequency is low, the size of the encoding packet can be increased to reduce the number of transmissions; conversely, when the network bandwidth is limited or the acquisition frequency is high, the size of the encoding packet should be reduced to avoid network congestion.

[0040] After determining the size of the coded packet, the next step is to determine the compression parameters of the data based on this size and the adjusted acquisition frequency, and compress and randomly linearly encode the core monitoring data to generate a core monitoring data packet. Specifically, each coded packet size is set with matching compression parameters. These compression parameters can include compression ratio, compression algorithm, etc., which are designed to ensure that the data can still maintain sufficient accuracy and readability after compression, while reducing the data size to fit the size limit of the coded packet. Compress the core monitoring data using the determined compression parameters. Compression can significantly reduce the size of the data, thereby reducing transmission costs and time. Randomly linearly encode the compressed data. This is a data encoding technology that can increase the redundancy and robustness of the data, so that even if part of the data is lost or damaged during transmission, the original data or content close to the original data can be restored through the decoding algorithm. Finally, the compressed and randomly linearly encoded data is packaged into a core monitoring data packet for transmission.

[0041] In the embodiment of the present invention, in step 021, the formula for dynamically adjusting the size of the encoding packet according to the current network bandwidth, the average network bandwidth, the adjusted acquisition frequency and the maximum acquisition frequency is: In the formula, represents the coded packet size of the i-th type of core monitoring data packet at the current time t, Indicates the size of the encoding packet of the i-th type of core monitoring data at the previous time. represents the current time t The collection frequency of the core monitoring data, that is, the adjusted collection frequency, Indicates the preset maximum sampling frequency of the i-th type of core monitoring data, Indicates the current network bandwidth. represents the average network bandwidth, , , are all constant coefficients.

[0042] Step 3: Send the core monitoring data packet to the monitoring terminal.

[0043] Based on the above-mentioned embodiment 1 of the invention, the step before step 1 includes: Step 001, set a collection threshold for each type of monitoring data, and the collection threshold is used to trigger the collection of core monitoring data; Step 002, obtain historical monitoring data of each type of monitoring data, make predictions based on the historical monitoring data, obtain monitoring prediction values, calculate the difference between the monitoring prediction values ​​and the monitoring actual values, and dynamically adjust the collection threshold based on the difference; Step 003, when the monitoring data is not within the adjusted collection threshold, determine the monitoring data as core monitoring data and trigger the collection of core monitoring data.

[0044] The embodiment of the present invention triggers the core monitoring data collection only when there is an abnormality in the data, thereby avoiding recording unnecessary data and reducing computing overhead.

[0045] In summary, through the above specific implementation methods, a data collection and integration method for adaptively capturing critical core monitoring clusters of the present invention is described in detail. The method can dynamically adjust the data collection and transmission strategy according to the changes in monitoring data and network conditions, ensure the efficient and accurate transmission of core monitoring data, and provide strong support for the monitoring and treatment of critically ill patients.

[0046] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein by equivalents. These modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be included in the scope of the claims and specification of the present invention.

Claims

1. A data collection and integration method for adaptively capturing a critical core monitoring cluster, characterized in that: include: Step 1, evaluating the change index and change correlation index of the core monitoring data, and dynamically adjusting the collection frequency of the core monitoring data according to the change index and change correlation index; Step 2: collecting core monitoring data according to the adjusted collection frequency, taking into account the network conditions, and packaging the collected core monitoring data in combination with the adjusted collection frequency to obtain a core monitoring data packet; Step 3: Send the core monitoring data packet to the monitoring terminal.

2. According to claim 1, a data collection and integration method for adaptively capturing a critical care core monitoring cluster is characterized in that: Before step 1, also include: Step 001, setting a collection threshold for each type of monitoring data, wherein the collection threshold is used to trigger the collection of core monitoring data; Step 002, obtaining historical monitoring data of various monitoring data, making predictions based on the historical monitoring data to obtain monitoring prediction values, calculating the difference between the monitoring prediction values ​​and the monitoring actual values, and dynamically adjusting the collection threshold according to the difference; Step 003: When the monitoring data is not within the adjusted collection threshold, the monitoring data is determined as core monitoring data and the collection of core monitoring data is triggered.

3. According to claim 1, a data collection and integration method for adaptively capturing a critical core monitoring cluster is characterized in that: In step 1, the evaluation method of the change index includes: Step 01, determining the length of a sliding window, wherein the sliding window is used to calculate the time correlation of the core monitoring data at each time point; Step 02, for each time point in the historical core monitoring data, calculate the mean and standard deviation of the data in the sliding window; Step 03, within the sliding window, for each core monitoring data, calculate the product of the standard deviation value of the core monitoring data with the current core monitoring data, and sum them up to obtain the time correlation index, calculate the ratio of the second-order derivative of the data to the first-order derivative, determine the data change rate, and determine the change index based on the data change rate and the time correlation coefficient.

4. According to claim 3, a data collection and integration method for adaptively capturing a critical care core monitoring cluster is characterized in that: In step 1, the methods for evaluating the change association index include: Step 04, sorting the change indexes of various core monitoring data in descending order and normalizing them to obtain a change index sequence; Step 05, calculating the correlation coefficient between two change indexes in the change index sequence; Step 06, constructing an association graph based on the change index and the correlation coefficient. The association graph takes the core monitoring data as points and determines the edge by the difference between the correlation coefficient and the preset coefficient threshold. When the difference is greater than zero, there is an edge. The correlation coefficient is corrected according to the association graph to obtain the change correlation index.

5. According to claim 4, a data collection and integration method for adaptively capturing a critical core monitoring cluster is characterized in that: The correlation coefficient is corrected according to the correlation diagram to obtain the formula of the change correlation index: In the formula, represents the change correlation index, Indicates The correlation coefficient of the core monitoring data, Indicates the total number of core monitoring data. Indicates The core monitoring data of this class has edge connections The correlation coefficient of the core monitoring data, Indicates The core monitoring data of this class has edge connections The total number of core monitoring data of the class.

6. According to the data collection and integration method of adaptively capturing a critical core monitoring cluster according to claim 5, it is characterized in that: The formula for dynamically adjusting the sampling frequency of core monitoring data according to the change index and the change correlation index is: In the formula, represents the current time t The frequency of collecting core monitoring data, Indicates the previous time The frequency of collecting core monitoring data, Indicates The change index of the core monitoring data of the class, Indicates Importance weight of core monitoring data, is a constant coefficient.

7. The data collection and integration method for adaptively capturing a critical care core monitoring cluster according to claim 1, characterized in that: In step 2, the collected core monitoring data is packaged considering the network conditions and the adjusted collection frequency, including: Step 021, dynamically adjusting the size of the encoding packet according to the current network bandwidth, the average network bandwidth, the adjusted acquisition frequency and the maximum acquisition frequency; Step 022, determining compression parameters according to the adjusted acquisition frequency and encoding packet size, compressing the core monitoring data according to the compression parameters, and performing random linear encoding on the compressed data to obtain a core monitoring data packet.

8. The data collection and integration method for adaptively capturing a critical care core monitoring cluster according to claim 7, characterized in that: Step 021, according to the current network bandwidth, the average network bandwidth, the adjusted acquisition frequency and the maximum acquisition frequency, the formula for dynamically adjusting the encoding packet size is: In the formula, represents the coded packet size of the i-th type of core monitoring data packet at the current time t, Indicates the size of the encoding packet of the i-th type of core monitoring data at the previous time. represents the current time t The collection frequency of the core monitoring data, that is, the adjusted collection frequency, Indicates the preset maximum sampling frequency of the i-th type of core monitoring data, Indicates the current network bandwidth. represents the average network bandwidth, , , are all constant coefficients.

Images