Pathological data monitoring system based on Internet of Things

Abstract

The invention discloses a pathology data monitoring system based on the Internet of Things, relates to the technical field of pathology data monitoring, solves the problem that a plurality of groups of different pathology data have large difference in comprehensive verification and cannot achieve a better comprehensive verification effect, and provides a pathology data monitoring system based on the Internet of Things according to respective unique data generation periods of different pathology items. Accurately analyzing and determining the corresponding execution frequency; the change characteristics of different pathological data are fully considered in the process, for example, some pathological indexes possibly fluctuate violently in a short time, and the other pathological indexes change slowly. Through deep mining of past historical monitoring data and application of a complex feature confirmation method including standard feature calculation, multi-order combination process execution and the like, the optimal execution frequency with relatively stable data change of each pathological item can be accurately locked.

Description

Technical Field

[0001] The present invention relates to the technical field of pathological data monitoring, and specifically to an Internet of Things-based pathological data monitoring system. Background Art

[0002] In the medical field, accurate and timely pathological data plays a decisive role in disease diagnosis, treatment plan formulation, and patient recovery process tracking. Traditional pathological data monitoring methods mostly rely on manual regular collection and recording, which not only consumes a large amount of manpower and material resources, but also is extremely prone to errors due to human factors; for example, in some large hospitals, medical staff need to frequently shuttle between various wards to manually measure and record basic pathological data of patients such as body temperature and blood pressure. The data measured multiple times a day is recorded in a paper medical record, and then manually entered into the computer system. The whole process is cumbersome and time-consuming. Once the medical staff is busy, the data collection time may be delayed, and the patient's condition changes cannot be reflected in real time.

[0003] An application with the publication number CN103371801A discloses a pathological data monitoring system using Internet of Things technology, including: a user terminal device, including a collection module for collecting the physiological data of a user by wearing on the user; a communication module for sending the collected physiological data to a data processing platform via a wireless communication network; the data processing platform includes the following modules: a maintenance module for recording information of at least one remote terminal specified by the user; an analysis and processing module for analyzing and processing the physiological data of the user collected; an alarm module for generating alarm data when preset conditions are met; a notification module for sending the alarm data to the remote terminal specified by the user via the wireless communication network, thereby realizing a pathological data monitoring system using Internet of Things technology.

[0004] Based on the pathological data monitoring process of the Internet of Things, the monitoring data frequencies associated with different data items are all different, resulting in time differences in the overall evaluation of the monitored data. There are large differences in the comprehensive verification of several groups of different pathological data, and a better comprehensive verification effect cannot be achieved, and the overall comprehensive evaluation accuracy is relatively low. Summary of the Invention

[0005] Aiming at the deficiencies of the prior art, the present invention provides an Internet of Things-based pathological data monitoring system, which solves the problem that there are large differences in the comprehensive verification of several groups of different pathological data and a better comprehensive verification effect cannot be achieved.

[0006] To achieve the above objectives, the present invention is realized through the following technical solutions: An Internet of Things-based pathological data monitoring system, including:

[0007] The associated frequency confirmation end performs frequency analysis on the periods of different data associated with different pathological items. According to the change difference characteristics between different pathological items in different frequency periods, the execution frequency of the corresponding pathological item is confirmed. Then, based on the execution frequencies associated with different pathological items, the acquisition frequencies for the synchronous execution of multiple pathological items are confirmed and transmitted to the execution end. The specific method is as follows:

[0008] Calibrate the periods of different data associated with different pathological items as Z i , where i represents different pathological items. Based on the historical monitoring data associated with different pathological items in the past stage, confirm the different monitoring data associated with different pathological items in different data generation periods and calibrate them as SJ i-k , for the same pathological item, select at least thirty groups of monitoring data belonging to the same pathological item as a data set;

[0009] Based on the different data sets associated with different pathological items, confirm the characteristics of different data sets and select the execution frequencies associated with the corresponding data sets:

[0010] Confirm the standard characteristics: Perform variance processing on several groups of monitoring data associated in the data set to confirm the first process characteristics, and use: Standard characteristics = First process characteristics × C1 + 0 × C2, where both C1 and C2 are preset fixed coefficient factors, to confirm the standard characteristics associated with the corresponding data set;

[0011] Execute the second-order combination process: Starting from the first monitoring data in the data set, select several groups of adjacent data columns from the data set in the form of two-by-two selection. The same group of monitoring data cannot appear in different adjacent data columns. Perform mean processing on the monitoring data associated in the adjacent data columns to confirm the data column characteristics of the adjacent data columns. Perform variance processing on several data column characteristics in this data set to confirm the first process characteristics of the second-order combination process. Then confirm the change difference between adjacent data columns, where the change difference = |Previous group of monitoring data - Next group of monitoring data|, and both the previous group of monitoring data and the next group of monitoring data belong to adjacent data columns. Perform mean processing on several groups of change differences associated with several adjacent data columns to confirm the second process characteristics of the second-order combination process. Use: Combined characteristics = First process characteristics × C1 + Second process characteristics × C2 to confirm the process characteristics belonging to this second-order combination process, where both C1 and C2 are preset fixed coefficient factors;

[0012] Execute the third-order combination process: starting from the first monitored data in the data set, select several groups of adjacent data columns from the data set in the form of two-by-two selection, and use the same processing method as the second-order combination process to confirm the first process feature of the third-order combination process, and then confirm the change difference between adjacent data columns. The change difference = |the maximum monitored data in the adjacent data columns - the minimum monitored data in the adjacent data columns|, and perform a mean process on several groups of change differences associated with several adjacent data columns to confirm the second process feature of the second-order combination process, and then confirm the process feature associated with this third-order combination process;

[0013] And so on, then execute the fourth-order combination process and the fifth-order combination process respectively, and confirm the process features associated with the corresponding combination processes. For different data sets associated with different pathological items, use the same processing method to confirm the different process features associated with the corresponding combination processes of the corresponding pathological items, and lock the execution frequency from different process features;

[0014] The specific method of locking the execution frequency from different process features is:

[0015] Based on several groups of process features associated with different pathological items, select a qualified process: mark the combination process that satisfies |standard feature - process feature| ≤ Y1 as a qualified process, where Y1 is a preset value:

[0016] If there is no qualified process, then use the original data generation period Z of the pathological item i as the confirmed execution frequency;

[0017] If there is only one group of qualified processes, then confirm the combination method of this qualified process. If it is a second-order combination process, the combination value is 2. If it is a third-order combination process, the combination value is 3. Use: execution frequency = Z i × combination value to confirm the execution frequency associated with this pathological item;

[0018] If there are multiple groups of qualified processes, then select the qualified process associated with the minimum process feature, and confirm its execution frequency based on the combination method of the selected qualified process;

[0019] The specific method of confirming the acquisition frequency of synchronous execution of multiple pathological items according to the execution frequencies associated with different pathological items is:

[0020] Mark the execution frequencies confirmed for different pathological items as P i , where i represents different pathological items. From several groups of execution frequencies, select the execution frequency associated with P i min as the pending frequency;

[0021] Confirm whether there is a multiple relationship between other execution frequencies and the pending frequency. Use: other execution frequency ÷ pending frequency = eigenvalue. If the eigenvalue ≥ 2 and is an integer, then use the other execution frequency that meets this evaluation condition as the acquisition frequency for the corresponding pathological item. For other execution frequencies that do not meet this evaluation condition, use the pending frequency as the acquisition frequency for the corresponding pathological item;

[0022] The execution end calibrates the acquisition times associated with different pathological items to the same time, and based on the confirmed different acquisition frequencies, executes the data monitoring and acquisition process for different pathological items;

[0023] The data monitoring center monitors different pathological items based on the different acquisition frequencies associated with different pathological items, and generates a data change curve based on the monitored data of different pathological items;

[0024] The data analysis end confirms the change trends of different data change curves based on the different data change curves associated with different pathological items, and based on the specific change characteristics of the change trends, locks the abnormal index segments, and displays the locked abnormal index segments through the display end. The specific method is as follows:

[0025] For a single group of pathological items, confirm the change trend of its data change curve in real time, and confirm the standard median line based on the associated change trend: Based on the monitoring data generated in real time within the data change curve, perform mean processing on several groups of monitoring data, and confirm the standard median line perpendicular to the Y-axis and parallel to the X-axis in the two-dimensional coordinate system. The value associated with the standard straight line is the data mean;

[0026] Calibrate the monitoring data generated in real time as JC q and confirm JC q The vertical difference value from the standard median line associated with the previous moment, and compare the confirmed vertical difference value with the preset value Y2: If the vertical difference value > Y2, then calibrate the corresponding monitoring data as abnormal data. Otherwise, do not perform any calibration. Here, q represents different moments, and Y2 is the preset value;

[0027] Confirm the data segments of continuously occurring abnormal data, calibrate the confirmed data segments as abnormal numerical segments, identify and confirm the abnormal numerical segments of each data change curve, and display the monitoring data associated with each abnormal numerical segment through the display end;

[0028] Confirm the abnormal time periods associated with each abnormal numerical segment, calibrate the abnormal numerical segments with the same time period as abnormal index segments, and display the confirmed abnormal index segments through the display end.

[0029] Preferably, the data generation periods associated with different pathological items are confirmed by a data period confirmation end, and the confirmed data generation periods are transmitted into an associated frequency confirmation end, where different pathological items are associated with different data generation periods.

[0030] Preferably, the specific manner in which the data monitoring center generates a data change curve is as follows:

[0031] Based on the different monitoring data associated with different pathological items, according to the chronological relationship, the different monitoring data associated with different moments are confirmed and sorted. Based on several groups of monitoring data sorted before and after, a data change curve belonging to the corresponding pathological item is generated. The horizontal axis of this curve is the time line, and its vertical axis is the monitoring data.

[0032] The present invention provides a pathological data monitoring system based on the Internet of Things. Compared with the prior art, it has the following beneficial effects:

[0033] The present invention can accurately analyze and determine the corresponding execution frequency according to the unique data generation periods of different pathological items. This process fully considers the characteristics of different pathological data changes. For example, some pathological indicators may fluctuate violently in a short period of time, while others change relatively slowly. By deeply mining the past historical monitoring data and using complex feature confirmation methods, including calculating standard features and performing multi-stage combination processes, the best execution frequency with relatively stable data changes for each pathological item can be accurately locked. Compared with the traditional unified acquisition frequency setting, this personalized frequency setting method greatly improves the pertinence and effectiveness of data acquisition, avoids data omission or redundancy caused by improper acquisition frequency, ensures that the collected data can accurately reflect the pathological changes, and provides a high-quality data basis for subsequent data analysis and diagnosis;

[0034] The system cleverly selects the minimum frequency from the execution frequencies of different pathological items as the pending frequency, and reasonably determines the acquisition frequency of each pathological item by judging the multiple relationship between other execution frequencies and the pending frequency. This method can effectively coordinate the data acquisition of different pathological items in time, enabling the data acquisition of multiple pathological items to not only meet their respective frequency requirements but also achieve synchronization to a certain extent, facilitating subsequent data integration and analysis;

[0035] The data monitoring center can monitor the pathological data in real time and accurately according to the acquisition frequencies of different pathological items, and generate intuitive data change curves. These curves have the time as the horizontal axis and the monitoring data as the vertical axis, clearly showing the dynamic change process of the pathological data over time; the data analysis end, based on these curves, confirms the change trend through scientific methods, accurately locks the abnormal index segments, and visually presents them to medical staff through the display end. Description of the Drawings

[0036] Figure 1 This is a schematic diagram of the principle framework of the present invention. Specific embodiments

[0037] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0038] The first embodiment

[0039] Please refer to Figure 1 , this application provides a pathological data monitoring system based on the Internet of Things, including a data cycle confirmation end, an associated frequency confirmation end, an execution end, a data monitoring center, a data analysis end, and a display end. Among them, the data cycle confirmation end, the associated frequency confirmation end, and the execution end are electrically connected in sequence from the output node to the input node, and the execution end, the data monitoring center, the data analysis end, and the display end are electrically connected in sequence from the output node to the input node;

[0040] Among them, the data cycle confirmation end confirms the data generation cycle associated with different pathological items and transmits the confirmed data generation cycle to the associated frequency confirmation end. Among them, different pathological items are associated with different data generation cycles, and their data generation cycles are all preset cycles, which are formulated by relevant operators according to experience, and the data generation cycles associated with different pathological items are all different;

[0041] Among them, the associated frequency confirmation end performs frequency analysis on the different data generation cycles associated with different pathological items, confirms the execution frequency of the corresponding pathological item according to the change difference characteristics between different pathological items in different frequency cycles, and then confirms the acquisition frequency of multiple pathological items to be executed synchronously according to the execution frequency associated with different pathological items and transmits it to the execution end. Specifically, the data acquisition cycles associated with different data items are all different. Then, in the actual application analysis process, there will be a large difference between data. To avoid such a situation, the acquisition frequency in the data monitoring process is adjusted to a unified frequency for convenient unified acquisition. Subsequently, the acquisition frequencies associated with multiple pathological items may be the same or different;

[0042] The specific method for confirming the execution frequency of the corresponding pathological item is:

[0043] Calibrate the different data generation cycles associated with different pathological items as Z i, where i represents different pathological items. Based on the historical monitoring data associated with different pathological items in previous stages, different monitoring data associated with different pathological items in different data generation cycles are confirmed and labeled as SJ. i-k , for the same pathological item, at least thirty groups of monitoring data belonging to the same pathological item are selected as a data set;

[0044] Based on the different data sets associated with different pathological items, the features of different data sets are confirmed, and the execution frequencies associated with the corresponding data sets are selected:

[0045] Confirm the standard features: Perform variance processing on several groups of monitoring data associated with the data set to confirm the first process features, and use: Standard features = First process features × C1 + 0 × C2, where both C1 and C2 are preset fixed coefficient factors, and their specific values are determined by the operator according to experience, to confirm the standard features associated with the corresponding data set;

[0046] Execute the second-order combination process: Starting from the first monitoring data in the data set, in the form of selecting two by two, several groups of adjacent data columns are selected from the data set. The same group of monitoring data cannot appear in different adjacent data columns. For example, if a data set is designated as {X1, X2, X3, X4,..., Xn}, then when selecting data columns in the form of two by two, the associated adjacent data columns can be expressed as X1 - X2, X3 - X4, X5 - X6, etc. Perform mean processing on the monitoring data associated with the adjacent data columns to confirm the data column features of the adjacent data columns. Perform variance processing on several data column features within this data set to confirm the first process features of the second-order combination process, and then confirm the change difference between adjacent data columns. The change difference = |previous group of monitoring data - next group of monitoring data|, where the previous group of monitoring data and the next group of monitoring data both belong to adjacent data columns. Perform mean processing on several groups of change differences associated with several adjacent data columns to confirm the second process features of the second-order combination process. Use: Combined features = First process features × C1 + Second process features × C2 to confirm the process features belonging to this second-order combination process, where both C1 and C2 are preset fixed coefficient factors, and their specific values are determined by the operator according to experience;

[0047] Execute the third-order combination process: starting from the first monitored data in the dataset, select several groups of adjacent data columns from the dataset in the form of selecting two by two, and adopt the same processing method as the second-order combination process to confirm the first process feature of the third-order combination process. Then confirm the change difference between adjacent data columns, and the change difference = |the maximum monitored data in the adjacent data columns - the minimum monitored data in the adjacent data columns|. And perform mean processing on several groups of change differences associated with several adjacent data columns to confirm the second process feature of the second-order combination process, and then confirm the process feature associated with this third-order combination process;

[0048] And so on, then execute the fourth-order combination process and the fifth-order combination process respectively, and confirm the process features associated with the corresponding combination processes. For different datasets associated with different pathological items, adopt the same processing method to confirm the different process features associated with the corresponding combination processes of the corresponding pathological items;

[0049] Based on several groups of process features associated with different pathological items, select a qualified process: Mark the combination process that satisfies |standard feature - process feature| ≤ Y1 as a qualified process, where Y1 is a preset value, and its specific value is determined by the operator according to experience:

[0050] If there is no qualified process, then use the original data generation period Z of the pathological item i as the confirmed execution frequency;

[0051] If there is only one group of qualified processes, then confirm the combination method of this qualified process. If it is a second-order combination process, the combination value is 2. If it is a third-order combination process, the combination value is 3. Adopt: execution frequency = Z i × combination value to confirm the execution frequency associated with this pathological item;

[0052] If there are multiple groups of qualified processes, then select the qualified process associated with the minimum process feature, and based on the combination method of the selected qualified process, confirm its execution frequency;

[0053] Specifically, for different pathological items, different execution frequencies need to be confirmed. Subsequently, according to this execution frequency, subsequent monitoring and collection of pathological items are carried out. Because for different execution frequencies in the specific execution process, the data change states associated with the pathological items are all in a relatively small change state, then the best execution frequency can be locked to facilitate the specific collection of monitoring data in the future;

[0054] Among them, the specific method for confirming the collection frequency of multiple pathological items executed synchronously based on the execution frequencies associated with different pathological items is:

[0055] Mark the execution frequencies confirmed for different pathological items as P i, where i represents different pathological items, and from several sets of execution frequencies, the execution frequency associated with P i min is selected as the pending frequency;

[0056] Confirm whether there is a multiple relationship between other execution frequencies and the pending frequency. Use: other execution frequency ÷ pending frequency = eigenvalue. If the eigenvalue ≥ 2 and is an integer, then use the other execution frequencies that meet this evaluation condition as the acquisition frequencies for the corresponding pathological items. For other execution frequencies that do not meet this evaluation condition, use the pending frequency as the acquisition frequency for the corresponding pathological items;

[0057] Specifically, if the determined execution frequencies associated with different pathological items are: 1, 3, 2.5, and 5 respectively, then the confirmed pending frequency is 1. Therefore, the execution frequencies with an integer multiple relationship are 3 or 5, so 3 or 5 is the acquisition frequency associated with the corresponding pathological item. However, the execution frequency of 2.5 has no multiple relationship, so the execution frequency of 2.5 is adjusted to the pending frequency of 1. Thus, the acquisition frequencies corresponding to different pathological items are all specifically confirmed.

[0058] Among them, the execution end calibrates the acquisition times associated with different pathological items to the same time, and based on the confirmed different acquisition frequencies, executes the data monitoring and acquisition process for different pathological items.

[0059] Second Embodiment

[0060] Among them, the data monitoring center monitors different pathological items based on the different acquisition frequencies associated with different pathological items, generates a data change curve based on the monitored data of different pathological items, and transmits the generated data change curve into the data analysis end. The generation method of the data change curve is as follows:

[0061] Based on the different monitored data associated with different pathological items, according to the chronological relationship, confirm the different monitored data associated with different times and sort them. Based on several groups of monitored data sorted before and after, generate a data change curve belonging to the corresponding pathological item. The horizontal axis of this curve is the time line, and its vertical axis is the monitored data.

[0062] Among them, the data analysis end confirms the change trends of different data change curves based on the different data change curves associated with different pathological items, and locks the abnormal index segments according to the specific change characteristics of the change trends, and displays the locked abnormal index segments through the display end. The specific method of locking is as follows:

[0063] For a single set of pathological items, the change trend of its data change curve is confirmed in real time, and the standard median line is confirmed based on the associated change trend confirmation criteria: based on the monitoring data generated in real time within the data change curve, several groups of monitoring data are averaged, and a standard median line perpendicular to the Y-axis and parallel to the X-axis is confirmed in a two-dimensional coordinate system. The value associated with the standard straight line is the data average value;

[0064] Calibrate the monitoring data generated in real time as JC q , and confirm JC q The vertical difference value between JC and the standard median line associated with the previous moment (the vertical difference value is the vertical distance between the corresponding data and the standard straight line of the previous moment, and the vertical distance is the value associated with the Y-axis), and compare the confirmed vertical difference value with the preset value Y2: if the vertical difference value > Y2, then calibrate the corresponding monitoring data as abnormal data; otherwise, do not perform any calibration. Here, q represents different moments, Y2 is the preset value, and its specific value is determined by the operator according to experience;

[0065] Confirm the data segment of continuously occurring abnormal data, calibrate the confirmed data segment as an abnormal value segment, identify and confirm the abnormal value segment of each data change curve, and display the monitoring data associated with each abnormal value segment through the display terminal;

[0066] Confirm the abnormal time period associated with each abnormal value segment, calibrate the abnormal value segments with the same time period as an abnormal index segment, and display the confirmed abnormal index segment through the display terminal. The same time period can be understood as follows: if there are two abnormal value segments, and their abnormal time periods are 1-5 or 2-6, then the generated same time period is 2-5, and the abnormal value segment associated with 2-5 is the associated abnormal index segment. The abnormal index segment represents that there are two or more groups of abnormalities, so attention needs to be paid. Therefore, specific display is required for such abnormal index segments.

[0067] Some of the data in the above formula are numerically calculated after removing their dimensions, and the content not described in detail in this specification belongs to the prior art well-known to those skilled in the art.

[0068] The above embodiments are only used to illustrate the technical method of the present invention and not to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical method of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical method of the present invention.

Claims

1. The pathological data monitoring system based on the Internet of Things is characterized in that, Including: An associated frequency confirmation terminal that performs frequency analysis on the data generation cycles of different data associated with different pathological items. According to the change difference characteristics between different frequency cycles of different pathological items, it confirms the execution frequency of the corresponding pathological item. Then, based on the execution frequencies associated with different pathological items, it confirms the acquisition frequency for the synchronous execution of multiple pathological items and transmits it to the execution terminal; An execution terminal that calibrates the acquisition times associated with different pathological items to the same time and executes the data monitoring and acquisition processes of different pathological items according to the confirmed different acquisition frequencies; A data monitoring center that monitors data for different pathological items according to the different acquisition frequencies associated with different pathological items and generates data change curves based on the monitored data of different pathological items; A data analysis terminal that confirms the change trends of different data change curves according to the different data change curves associated with different pathological items. Based on the specific change characteristics of the change trends, it locks the abnormal index segments and displays the locked abnormal index segments through a display terminal.

2. The pathological data monitoring system based on the Internet of Things according to claim 1, wherein, The data generation cycles associated with different pathological items are confirmed by a data cycle confirmation terminal, and the confirmed data generation cycles are transmitted to the associated frequency confirmation terminal, where different pathological items are associated with different data generation cycles.

3. The pathological data monitoring system based on the Internet of Things according to claim 1, characterized in that, The specific method for the associated frequency confirmation terminal to confirm the execution frequency of the corresponding pathological item is as follows: Calibrate the different data generation cycles associated with different pathological items as Z i , where i represents different pathological items, and based on the historical monitoring data associated with different pathological items in the past stage, confirm the different monitoring data associated with different pathological items in different data generation cycles, and calibrate them as SJ i-k , for the same pathological item, select at least thirty groups of monitoring data belonging to the same pathological item as a data set; Based on different data sets associated with different pathological items, it performs feature confirmation on different data sets and selects the execution frequency associated with the corresponding data set: Confirm the standard feature: Perform variance processing on several groups of monitored data associated with the data set to confirm the first process feature, and use: standard feature = first process feature × C1 + 0 × C2, where C1 and C2 are both preset fixed coefficient factors, to confirm the standard feature associated with the corresponding data set; Execute the second-order combination process: Starting from the first monitored data of the data set, in the form of selecting two by two, select several groups of adjacent data columns from the data set. The same group of monitored data cannot appear in different adjacent data columns. Perform mean processing on the monitored data associated with the adjacent data columns to confirm the data column features of the adjacent data columns. Perform variance processing on several data column features within this data set to confirm the first process feature of the second-order combination process. Then confirm the change difference between adjacent data columns, where the change difference = |previous group of monitored data - next group of monitored data|, and the previous group of monitored data and the next group of monitored data both belong to adjacent data columns. Perform mean processing on several groups of change differences associated with several adjacent data columns to confirm the second process feature of the second-order combination process. Use: combined feature = first process feature × C1 + second process feature × C2 to confirm the process feature belonging to this second-order combination process, where C1 and C2 are both preset fixed coefficient factors; Execute the third-order combination process: starting from the first monitored data in the dataset, select several groups of adjacent data columns from the dataset in the form of selecting two by two, and adopt the same processing method as the second-order combination process to confirm the first process feature of the third-order combination process. Then, confirm the change difference between adjacent data columns, where the change difference = |the maximum monitored data in the adjacent data column - the minimum monitored data in the adjacent data column|, and perform mean processing on several groups of change differences associated with several adjacent data columns to confirm the second process feature of the second-order combination process. Then, confirm the process feature associated with this third-order combination process; And so on, then execute the fourth-order combination process and the fifth-order combination process respectively, and confirm the process features associated with the corresponding combination processes. For different datasets associated with different pathological items, adopt the same processing method to confirm the different process features associated with the corresponding combination processes of the corresponding pathological items, and lock the execution frequency from different process features.

4. The pathological data monitoring system based on the Internet of Things according to claim 3, characterized in that, The specific method for the associated frequency confirmation end to lock the execution frequency from different process features is as follows: Based on several groups of process features associated with different pathological items, select a qualified process: mark the combination process that satisfies |standard feature - process feature| ≤ Y1 as a qualified process, where Y1 is a preset value: If the compliance process does not exist, the original data generation cycle Z of the pathological item is used i as the confirmed execution frequency; If there is only one set of compliance processes, confirm the combination method of this compliance process. If it is a second-order combined process, the combination value is 2; if it is a third-order combined process, the combination value is 3. Adopt: Execution frequency = Z i × The combination value to confirm the execution frequency associated with this pathological item; If there are multiple groups of qualified processes, select the qualified process associated with the minimum process feature, and confirm its execution frequency based on the combination method of the selected qualified process.

5. The pathological data monitoring system based on the Internet of Things according to claim 4, characterized in that, The specific method for the associated frequency confirmation end to confirm the acquisition frequency of synchronous execution of multiple pathological items based on the execution frequencies associated with different pathological items is as follows: Calibrate the execution frequencies confirmed by different pathological items as P i , where i represents different pathological items, and select the execution frequency associated with P i min as the frequency to be determined; Confirm whether there is a multiple relationship between other execution frequencies and the pending frequency, using: other execution frequency ÷ pending frequency = eigenvalue. If the eigenvalue ≥ 2 and is an integer, then use the other execution frequency that satisfies this evaluation condition as the acquisition frequency of the corresponding pathological item. For other execution frequencies that do not satisfy this evaluation condition, use the pending frequency as the acquisition frequency of the corresponding pathological item.

6. The pathological data monitoring system based on the Internet of Things according to claim 1, wherein, The specific method for the data monitoring center to generate a data change curve is as follows: Based on different monitored data associated with different pathological items, confirm the different monitored data associated with different times according to the time sequence relationship, and sort them. Based on several groups of monitored data sorted before and after, generate a data change curve belonging to the corresponding pathological item. The horizontal axis of this curve is the time line, and its vertical axis is the monitored data.

7. The Internet of Things-based pathological data monitoring system according to claim 1, wherein The specific method for the data analysis end to analyze the change trend of different data change curves is as follows: For a single group of pathological items, confirm the change trend of its data change curve in real time, and confirm the standard median line based on the associated change trend: based on the monitored data generated in real time within the data change curve, perform mean processing on several groups of monitored data, and confirm a standard median line perpendicular to the Y-axis and parallel to the X-axis in the two-dimensional coordinate system. The value associated with the standard median line is the data mean; Calibrate the monitoring data generated in real time as JC q , and confirm JC q The vertical difference value from the standard median line associated with the previous moment, and compare the confirmed vertical difference value with the preset value Y2: If the vertical difference value > Y2, calibrate the corresponding monitoring data as abnormal data, otherwise, do not perform any calibration, where q represents different moments, and Y2 is the preset value; Confirm the data segment of continuously occurring abnormal data, mark the confirmed data segment as an abnormal value segment, identify and confirm the abnormal value segment of each data change curve, and display the monitored data associated with each abnormal value segment through the display end.

8. The pathological data monitoring system based on the Internet of Things according to claim 7, wherein The data analysis terminal confirms the abnormal time periods associated with each abnormal numerical segment, calibrates the abnormal numerical segments that exist in the same time period as abnormal index segments, and displays the confirmed abnormal index segments through a display terminal.

Images