Postoperative monitoring and early warning system and method for patient

Abstract

The invention relates to the technical field of postoperative monitoring, in particular to a postoperative monitoring and early warning system and method for a patient, and the system comprises a server side, a user side and a medical care side. The server comprises a data acquisition module, an anomaly judgment module, a dimension calling module, a processing module, an analysis module and an early warning module; wherein the processing module is used for calling habit information of a patient and daily data of the patient from a current time period to a previous time period, and clustering the daily data of the patient corresponding to each time period to form a daily data set of the patient corresponding to each abnormal cause dimension; the analysis module is used for sequentially carrying out disease cause analysis on data in the patient daily data set according to the patient daily data set corresponding to each abnormal disease cause dimension to form an abnormal disease cause result; and the early warning module is used for sending an alarm signal to the user side and the medical care side when a corresponding abnormal cause result is formed or each abnormal detection index data in the current postoperative index data set of the patient is identified.

Description

Technical Field

[0001] The present invention relates to the technical field of postoperative monitoring, and in particular to a monitoring and early warning system and method for patients after surgery. Background Art

[0002] With the development of medical technology, the success rate of surgery has been significantly improved, but postoperative care and monitoring still face many challenges. The occurrence of postoperative complications not only affects the patient's recovery speed, but may also threaten life safety. Therefore, how to effectively monitor the patient's condition after surgery and promptly detect and deal with potential problems has become the key to improving the quality of postoperative care.

[0003] Existing postoperative monitoring systems mostly focus on real-time monitoring of physiological parameters, such as heart rate and blood pressure. Although these systems can help medical staff understand the health status of patients to a certain extent, they have limitations in early warning of postoperative conditions. On the one hand, a single physiological parameter cannot fully reflect the health status of patients; on the other hand, traditional monitoring systems lack consideration of personalized factors such as patients' living habits and medical history, and cannot accurately analyze the causes of abnormal postoperative results in patients, resulting in inaccurate early warnings and prone to false alarms or missed reports. Summary of the invention

[0004] One of the purposes of the present invention is to provide a monitoring and early warning system and method for patients after surgery, which can quickly locate the cause of abnormalities in patients after surgery and issue early warnings in time, thereby improving the safety of patients after surgery.

[0005] In order to achieve the above-mentioned purpose, a patient postoperative monitoring and early warning system is provided, including a service end, a user end and a medical end;

[0006] The server includes:

[0007] A data collection module is used to collect various test index data corresponding to the postoperative patients in the current time period to form a corresponding postoperative index data set of the current patients;

[0008] The abnormality judgment module is used to identify each abnormal detection index data in the current patient's postoperative index data set based on the abnormal threshold value corresponding to each detection index data set preset according to the current patient's postoperative index data set;

[0009] A dimension retrieval module is used to retrieve all abnormal cause dimensions corresponding to each detection indicator data when an abnormality occurs from the database according to each abnormal detection indicator data in the identified postoperative indicator data set of the current patient;

[0010] A processing module is used to retrieve the habit information corresponding to the patient and the daily data of the patient corresponding to the current time period to the previous time period, and cluster the daily data of the patient corresponding to each time period based on a preset patient data clustering strategy to form a daily data set of the patient corresponding to each abnormal etiology dimension;

[0011] An analysis module is used to perform etiology analysis on the data in the corresponding patient daily data set in turn according to the patient daily data set corresponding to each abnormal etiology dimension, and form a corresponding abnormal etiology result;

[0012] The early warning module is used to send an alarm signal to the user end and the medical end when the corresponding abnormal cause result is formed or when each abnormal detection indicator data in the postoperative indicator data set of the current patient is identified.

[0013] The technical principle and effect of this scheme: In this scheme, first, the data of various detection indicators of postoperative patients in the current time period are collected, so as to realize the regular data collection of postoperative patients. Then, based on the abnormal threshold value corresponding to each detection indicator data, that is, the normal value, the abnormal data of each detection indicator data is identified, so as to determine the specific abnormal detection indicators of postoperative patients, so as to facilitate subsequent targeted identification and detection.

[0014] Then, through these abnormal detection indicator data, we can know the abnormal cause dimensions corresponding to these detection indicator data when abnormalities occur. For example, when some detection indicator data are abnormal, it can be either abnormal cause dimension A or abnormal cause dimension B. In this way, we can determine all abnormal cause dimensions when this abnormality occurs, avoiding omission problems.

[0015] Then, the patient's habit information and the patient's daily data corresponding to the time period from the current time period to the previous time period are retrieved, and based on the preset patient data clustering strategy, the patient's daily data corresponding to each time period are clustered to form a patient's daily data set corresponding to each abnormal cause dimension; thereby clustering the patient's daily data corresponding to the time period from the current time period to the previous time period, and by clustering the patient's daily data in various dimensions, the efficiency and reliability of subsequent abnormal cause result analysis are greatly improved, and the problem of low efficiency in cause result analysis caused by the large amount of data and complex situation faced in the existing cause result analysis is solved.

[0016] Finally, when the corresponding abnormal cause results are formed or the abnormal detection indicator data in the current patient's postoperative indicator data set is identified, an alarm signal is sent to the user end and the medical end. Through the alarm of these two situations, the attention of patients and medical staff is greatly improved, which can help patients and medical staff to control and solve the abnormalities as soon as possible, and can quickly locate the cause of abnormalities in patients after surgery and issue early warnings in time, thereby improving the safety of patients after surgery.

[0017] Furthermore, the preset patient data clustering strategy is:

[0018] S10, sorting the patient's daily data based on the chronological order according to the retrieved patient's daily data between the current time period and the previous time period;

[0019] S11, determining a corresponding division time interval according to the data volume corresponding to the daily data of the patient;

[0020] S12, dividing the sorted daily data of the patient according to the divided time intervals to form a daily data set corresponding to each divided time interval;

[0021] S13, according to the daily data sets corresponding to the divided time intervals, based on the pre-set data extraction strategy, determine the extraction point extraction scheme corresponding to each divided time interval, and extract the data of the corresponding extraction point in the daily data sets corresponding to each divided time interval according to the extraction point extraction scheme corresponding to each divided time interval;

[0022] The pre-set data extraction strategy is:

[0023] According to the daily data set corresponding to a certain divided time interval, the data type corresponding to each data in the daily data set is identified, and the data corresponding to each data type is determined;

[0024] In the corresponding divided time interval, a number of extraction point extraction schemes are randomly generated and a corresponding extraction point set is formed;

[0025] According to the extraction schemes of each extraction point in the extraction point set, based on the preset extraction rationality calculation formula, the extraction rationality corresponding to each extraction point extraction scheme is calculated in turn, and after each calculation is completed, the extraction rationality calculated this time is compared with the preset rationality threshold. If the extraction rationality is greater than the preset rationality threshold, the extraction point extraction scheme corresponding to the extraction rationality is judged to be the extraction rationality scheme corresponding to the divided time interval. Otherwise, the extraction rationality calculation of the next extraction point extraction scheme is continued;

[0026] The calculation formula for the extraction rationality is:

[0027]

[0028] In the formula, H is the rationality of extraction, L i is the representative value of the i-th sampling point in the sampling point extraction scheme, W iq is the weight value of the qth data type where the i-th extraction point is located in the extraction point extraction scheme, T q is the data time interval corresponding to the qth data type, Q is the total number of data types in a certain partitioned time interval, and Z i is the time rationality of the i-th extraction point in the extraction point extraction scheme, t i is the time corresponding to the i-th extraction point in the extraction point extraction scheme; α and β are the corresponding weighting coefficients;

[0029] S14, according to the data extracted from each divided time interval, based on a preset data similarity calculation formula, calculating the data similarity corresponding to the abnormal patient data corresponding to each divided time interval;

[0030] If the extracted data is a numerical value, the preset data similarity calculation formula is:

[0031]

[0032] Where, X ts is the data similarity between the partition time interval t and the partition time interval s, N is the number of extraction points, d(X tj ,X sk ) is the characteristic distance between the data value of the j-th extracted data in the partition time interval t and the data value of the k-th extracted data in the partition time interval s, ∝ is the attenuation factor corresponding to the feature clustering;

[0033] If the extracted data is non-numeric, the preset data similarity calculation formula is:

[0034]

[0035] Wherein, F is the maximum value of the number of data corresponding to the same content in the extraction points between the divided time interval t and the divided time interval s;

[0036] S15, according to the data similarity of the abnormal patient data corresponding to each divided time interval, the abnormal patient data corresponding to the divided time interval whose data similarity is less than a preset similarity threshold value are divided into the same category of data;

[0037] If the data similarity X between a certain partition time interval A and the partition time intervals B and C AB and X AC are all less than the preset similarity threshold, but the data similarity X between the divided time intervals B and C is BC If it is greater than the preset similarity threshold, then compare X AB and X AC The size between;

[0038] If X AB <X AC , then the abnormal patient data corresponding to the divided time interval A and the divided time interval B are divided into the same type of data;

[0039] If X AB >X AC , then the abnormal patient data corresponding to the divided time interval A and the divided time interval C are divided into the same type of data;

[0040] If X AB =X AC , a random number P from 1 to 1 is randomly generated. When P is greater than the preset random value, the abnormal patient data corresponding to the divided time interval A and the divided time interval B are divided into the same type of data. Otherwise, the abnormal patient data corresponding to the divided time interval A and the divided time interval C are divided into the same type of data.

[0041] S16. After completing the classification of all the abnormal patient data corresponding to the divided time intervals, randomly select a patient abnormal data corresponding to a divided time interval from each data of the same category, and select the data extracted from the corresponding divided time interval, and identify the dimension corresponding to the extracted data, and identify the abnormal cause dimension corresponding to the same category of data corresponding to the extracted data.

[0042] Beneficial effect: In this solution, the daily data of patients over a long period of time is first divided according to time intervals. For example, the difference between the current time period and the previous time period is 24 hours, and the time interval is 2 hours. The daily data of patients can be divided into data of 12 time intervals to form a corresponding daily data set.

[0043] After that, through the data collection strategy, the extraction scheme of the extraction point is determined from the daily data set in each divided time interval, so as to realize the extraction of data in each divided time interval. In the specific extraction, the first step is to randomly generate multiple extraction point extraction schemes, and calculate the schemes in turn through the extraction rationality calculation formula until the calculated extraction rationality meets the requirements. The second step is to extract the data of each divided time interval according to the determined extraction point extraction scheme, and cluster the data of each divided time interval through the data similarity between the data of each divided time interval. When calculating the data similarity, different data similarity calculation formulas are selected according to the different types of extracted data for reasonable calculation, which greatly improves the rationality and reliability of data similarity calculation.

[0044] The third step is to divide the same type of data according to the data similarity corresponding to each divided time interval. In this division process, when there is a division time interval with the same data similarity as the other two division time intervals, the division between the three division time intervals is randomly determined by setting random numbers, thereby improving the randomness of the division while ensuring the rationality of the division.

[0045] The last step is to identify the dimensions corresponding to the same type of data, so as to associate each data of the same type with its corresponding abnormal cause dimension.

[0046] Furthermore, the retrieving from the database all abnormal cause dimensions corresponding to each detection indicator data when an abnormality occurs includes all abnormal cause dimensions corresponding to when each detection indicator data occurs abnormally alone and when abnormalities occur simultaneously between each detection indicator data.

[0047] Beneficial effects: In this solution, the comprehensiveness and reliability of abnormal cause dimension retrieval are greatly improved by selecting abnormal cause dimension corresponding to various forms of abnormality of each detection indicator.

[0048] Furthermore, the server also includes:

[0049] The first reporting module is used to generate a medical solution report corresponding to the corresponding medical staff according to the abnormal cause results;

[0050] The second reporting module is used to formulate a user maintenance report that the corresponding patient needs to complete when executing the medical care solution report based on the abnormal cause results formed and the medical care solution report to be executed;

[0051] The sending module is used to send the medical solution report and the user maintenance report to the medical end and the user end respectively.

[0052] Beneficial effects: In this solution, through the production of medical solution reports, solution reports for specific medical staff can be automatically generated based on the results of abnormal causes. This not only reduces the time for medical staff to manually write reports, but also ensures the professionalism and accuracy of the report content, which helps to improve the quality and efficiency of diagnosis and treatment.

[0053] Based on the abnormal cause results and the planned medical solutions, user maintenance reports can be customized for patients. This personalized guidance program helps patients better understand their health status and manage themselves according to the guidance, thereby promoting treatment effectiveness.

[0054] Further, the first reporting module includes:

[0055] An evaluation module is used to retrieve the surgery-related data corresponding to the patient according to the abnormal cause result, evaluate the impact of the abnormal cause result on the patient, and retrieve the medical care solution report corresponding to the impact according to the impact;

[0056] The feedback update module is used to feed back the medical solution report to the medical end, receive reasonable feedback information of the measures fed back by the medical staff through the medical end, and update the medical solution report according to the reasonable feedback information of the measures.

[0057] Beneficial effects: According to the abnormal cause results, important data such as surgery related to the patient are retrieved to evaluate the possible impact of the abnormal cause on the patient. Then, based on the assessed impact, the most appropriate medical solution report is selected. This process ensures that each patient's treatment plan is carefully designed based on their specific situation, improving the targetedness and effectiveness of the treatment. By combining the patient's historical data, the treatment effect can be predicted more accurately, reducing misdiagnosis and unnecessary treatment steps, thereby improving patient safety and satisfaction.

[0058] The generated medical solution report is sent to the medical staff, and the feedback from the medical staff on the rationality of the measures is collected. The original solution report is adjusted or optimized based on this feedback. This mechanism establishes an effective two-way communication channel, enabling the system to quickly respond to changes and needs in clinical practice. The experience and intuition of medical staff are valuable resources. Through their feedback, the system can continuously learn and improve, and provide solutions that better meet actual needs. In addition, this also promotes knowledge sharing and collaboration within the medical team, which is conducive to improving the overall level of medical services.

[0059] The present invention also provides a postoperative monitoring and early warning method for patients, using the above-mentioned postoperative monitoring and early warning system for patients, comprising the following steps:

[0060] S1. Collect various test index data corresponding to the postoperative patients in the current time period to form a corresponding postoperative index data set of the current patients;

[0061] S2. According to the postoperative indicator data set of the current patient, based on the preset abnormal thresholds corresponding to the various detection indicator data, identify the various abnormal detection indicator data in the postoperative indicator data set of the current patient;

[0062] S3. According to the identified abnormal detection index data in the postoperative index data set of the current patient, all abnormal cause dimensions corresponding to each detection index data when the abnormality occurs are retrieved from the database;

[0063] S4, retrieve the habit information corresponding to the patient and the daily data of the patient corresponding to the current time period to the previous time period, and cluster the daily data of the patient corresponding to each time period based on the preset patient data clustering strategy to form a daily data set of the patient corresponding to each abnormal cause dimension;

[0064] S5. According to the patient daily data sets corresponding to each abnormal etiology dimension, the data in the corresponding patient daily data sets are analyzed for etiology in turn to form corresponding abnormal etiology results;

[0065] S6. When the corresponding abnormal cause result is formed or each abnormal detection index data in the postoperative index data set of the current patient is identified, an alarm signal is sent to the user end and the medical end. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] Figure 1 This is a logic block diagram of a monitoring and early warning system for patients after surgery in Embodiment 1 of the present invention;

[0067] Figure 2 The present invention is a flowchart of a postoperative monitoring and early warning method for patients in the first embodiment of the present invention. DETAILED DESCRIPTION

[0068] The following is further described in detail through specific implementation methods:

[0069] Embodiment 1

[0070] A postoperative monitoring and early warning system for patients, basically Figure 1 As shown, it includes a service end, a user end and a medical end;

[0071] The server includes:

[0072] A data collection module is used to collect various test index data corresponding to the postoperative patients in the current time period to form a corresponding postoperative index data set of the current patients;

[0073] The abnormality judgment module is used to identify each abnormal detection index data in the current patient's postoperative index data set based on the abnormal threshold value corresponding to each detection index data set preset according to the current patient's postoperative index data set;

[0074] The dimension retrieval module is used to retrieve from the database all abnormal cause dimensions corresponding to each detection indicator data when the abnormality occurs, based on each abnormal detection indicator data in the postoperative indicator data set of the current patient identified; the said retrieval from the database of all abnormal cause dimensions corresponding to each detection indicator data when the abnormality occurs includes all abnormal cause dimensions corresponding to each detection indicator data when the abnormality occurs individually and when abnormalities occur simultaneously between each detection indicator data.

[0075] A processing module is used to retrieve the habit information corresponding to the patient and the daily data of the patient corresponding to the current time period to the previous time period, and cluster the daily data of the patient corresponding to each time period based on a preset patient data clustering strategy to form a daily data set of the patient corresponding to each abnormal etiology dimension;

[0076] The preset patient data clustering strategy is:

[0077] S10, sorting the patient's daily data based on the chronological order according to the retrieved patient's daily data between the current time period and the previous time period;

[0078] S11, determining a corresponding division time interval according to the data volume corresponding to the daily data of the patient;

[0079] S12, dividing the sorted daily data of the patient according to the divided time intervals to form a daily data set corresponding to each divided time interval;

[0080] S13, according to the daily data sets corresponding to the divided time intervals, based on the pre-set data extraction strategy, determine the extraction point extraction scheme corresponding to each divided time interval, and extract the data of the corresponding extraction point in the daily data sets corresponding to each divided time interval according to the extraction point extraction scheme corresponding to each divided time interval;

[0081] The pre-set data extraction strategy is:

[0082] According to the daily data set corresponding to a certain divided time interval, the data type corresponding to each data in the daily data set is identified, and the data corresponding to each data type is determined; for example, in the same divided time interval, the patient's daily data set includes not only the patient's monitoring video information, but also the patient's diet plan data, and the patient's exercise data. These data are not necessarily continuous. For example, the monitoring video information may only be in the first half of the divided time interval, or may be set at intervals in the divided time interval.

[0083] In the corresponding divided time interval, a number of extraction point extraction schemes are randomly generated and a corresponding extraction point set is formed;

[0084] According to the extraction schemes of each extraction point in the extraction point set, based on the preset extraction rationality calculation formula, the extraction rationality corresponding to each extraction point extraction scheme is calculated in turn, and after each calculation is completed, the extraction rationality calculated this time is compared with the preset rationality threshold. If the extraction rationality is greater than the preset rationality threshold, the extraction point extraction scheme corresponding to the extraction rationality is judged to be the extraction rationality scheme corresponding to the divided time interval. Otherwise, the extraction rationality calculation of the next extraction point extraction scheme is continued;

[0085] The calculation formula for the extraction rationality is:

[0086]

[0087] In the formula, H is the rationality of extraction, L i is the representative value of the i-th sampling point in the sampling point extraction scheme, W iq is the weight value of the qth data type where the i-th extraction point is located in the extraction point extraction scheme, T q is the data time interval corresponding to the qth data type, Q is the total number of data types in a certain partitioned time interval, and Z i is the time rationality of the i-th extraction point in the extraction point extraction scheme, t i is the time corresponding to the i-th extraction point in the extraction point extraction scheme; α and β are the corresponding weighting coefficients;

[0088] S14, according to the data extracted from each divided time interval, based on a preset data similarity calculation formula, calculating the data similarity corresponding to the abnormal patient data corresponding to each divided time interval;

[0089] If the extracted data is a numerical value, the preset data similarity calculation formula is:

[0090]

[0091] Where, Xts is the data similarity between the partition time interval t and the partition time interval s, N is the number of extraction points, d(X tj ,X sk ) is the characteristic distance between the data value of the j-th extracted data in the partition time interval t and the data value of the k-th extracted data in the partition time interval s, ∝ is the attenuation factor corresponding to the feature clustering;

[0092] If the extracted data is non-numeric, the preset data similarity calculation formula is:

[0093]

[0094] Wherein, F is the maximum value of the number of data corresponding to the same content in the extraction points between the divided time interval t and the divided time interval s;

[0095] S15, according to the data similarity of the abnormal patient data corresponding to each divided time interval, the abnormal patient data corresponding to the divided time interval whose data similarity is less than a preset similarity threshold value are divided into the same category of data;

[0096] If the data similarity X between a certain partition time interval A and the partition time intervals B and C AB and X AC are all less than the preset similarity threshold, but the data similarity X between the divided time intervals B and C is BC If it is greater than the preset similarity threshold, then compare X AB and X AC The size between;

[0097] If X AB <X AC , then the abnormal patient data corresponding to the divided time interval A and the divided time interval B are divided into the same type of data;

[0098] If X AB >X AC , then the abnormal patient data corresponding to the divided time interval A and the divided time interval C are divided into the same type of data;

[0099] If X AB =X AC , a random number P from 1 to 1 is randomly generated. When P is greater than the preset random value, the abnormal patient data corresponding to the divided time interval A and the divided time interval B are divided into the same type of data. Otherwise, the abnormal patient data corresponding to the divided time interval A and the divided time interval C are divided into the same type of data.

[0100] S16. After completing the classification of all the abnormal patient data corresponding to the divided time intervals, randomly select a patient abnormal data corresponding to a divided time interval from each data of the same category, and select the data extracted from the corresponding divided time interval, and identify the dimension corresponding to the extracted data, and identify the abnormal cause dimension corresponding to the same category of data corresponding to the extracted data.

[0101] An analysis module is used to perform etiology analysis on the data in the corresponding patient daily data set in turn according to the patient daily data set corresponding to each abnormal etiology dimension, and form a corresponding abnormal etiology result;

[0102] The early warning module is used to send an alarm signal to the user end and the medical end when the corresponding abnormal cause result is formed or when each abnormal detection indicator data in the postoperative indicator data set of the current patient is identified.

[0103] The server also includes:

[0104] The first reporting module is used to generate a medical solution report corresponding to the corresponding medical staff according to the abnormal cause results;

[0105] The first reporting module comprises:

[0106] An evaluation module is used to retrieve the surgery-related data corresponding to the patient according to the abnormal cause result, evaluate the impact of the abnormal cause result on the patient, and retrieve the medical care solution report corresponding to the impact according to the impact;

[0107] The feedback update module is used to feed back the medical solution report to the medical end, receive reasonable feedback information of the measures fed back by the medical staff through the medical end, and update the medical solution report according to the reasonable feedback information of the measures.

[0108] The second reporting module is used to formulate a user maintenance report that the corresponding patient needs to complete when executing the medical care solution report based on the abnormal cause results formed and the medical care solution report to be executed;

[0109] The sending module is used to send the medical solution report and the user maintenance report to the medical end and the user end respectively.

[0110] like Figure 2 As shown, this embodiment also discloses a postoperative monitoring and early warning method for patients, using the above-mentioned postoperative monitoring and early warning system for patients, including the following steps:

[0111] S1. Collect various test index data corresponding to the postoperative patients in the current time period to form a corresponding postoperative index data set of the current patients;

[0112] S2. According to the postoperative indicator data set of the current patient, based on the preset abnormal thresholds corresponding to the various detection indicator data, identify the various abnormal detection indicator data in the postoperative indicator data set of the current patient;

[0113] S3. According to the identified abnormal detection index data in the postoperative index data set of the current patient, all abnormal cause dimensions corresponding to each detection index data when the abnormality occurs are retrieved from the database;

[0114] S4, retrieve the habit information corresponding to the patient and the daily data of the patient corresponding to the current time period to the previous time period, and cluster the daily data of the patient corresponding to each time period based on the preset patient data clustering strategy to form a daily data set of the patient corresponding to each abnormal cause dimension;

[0115] S5. According to the patient daily data sets corresponding to each abnormal etiology dimension, the data in the corresponding patient daily data sets are analyzed for etiology in turn to form corresponding abnormal etiology results;

[0116] S6. When the corresponding abnormal cause result is formed or each abnormal detection index data in the postoperative index data set of the current patient is identified, an alarm signal is sent to the user end and the medical end.

[0117] The above is only an embodiment of the present invention. The common sense such as the known specific structure and characteristics in the scheme is described too much here. The ordinary technicians in the relevant field know all the common technical knowledge in the technical field of the invention before the application date or priority date, can know all the existing technologies in the field, and have the ability to apply the conventional experimental means before that date. The ordinary technicians in the relevant field can improve and implement this scheme in combination with their own abilities under the enlightenment given by this application. Some typical known structures or known methods should not become obstacles for ordinary technicians in the relevant field to implement this application. It should be pointed out that for those skilled in the art, without departing from the structure of the present invention, several deformations and improvements can be made, which should also be regarded as the protection scope of the present invention, which will not affect the effect of the implementation of the present invention and the practicality of the patent. The protection scope required by this application shall be based on the content of its claims, and the specific implementation methods and other records in the specification can be used to interpret the content of the claims.

Claims

1. A postoperative monitoring and early warning system for patients, characterized in that: Including service side, user side and medical side; The server includes: A data collection module is used to collect various test index data corresponding to the postoperative patients in the current time period to form a corresponding postoperative index data set of the current patients; The abnormality judgment module is used to identify each abnormal detection index data in the current patient's postoperative index data set based on the abnormal threshold value corresponding to each detection index data set preset according to the current patient's postoperative index data set; A dimension retrieval module is used to retrieve all abnormal cause dimensions corresponding to each detection indicator data when an abnormality occurs from the database according to each abnormal detection indicator data in the identified postoperative indicator data set of the current patient; A processing module is used to retrieve the habit information corresponding to the patient and the daily data of the patient corresponding to the current time period to the previous time period, and cluster the daily data of the patient corresponding to each time period based on a preset patient data clustering strategy to form a daily data set of the patient corresponding to each abnormal etiology dimension; An analysis module is used to perform etiology analysis on the data in the corresponding patient daily data set in turn according to the patient daily data set corresponding to each abnormal etiology dimension, and form a corresponding abnormal etiology result; The early warning module is used to send an alarm signal to the user end and the medical end when the corresponding abnormal cause result is formed or when each abnormal detection indicator data in the postoperative indicator data set of the current patient is identified.

2. A postoperative monitoring and early warning system for patients according to claim 1, characterized in that: The preset patient data clustering strategy is: S10, sorting the patient's daily data based on the chronological order according to the retrieved patient's daily data between the current time period and the previous time period; S11, determining a corresponding division time interval according to the data volume corresponding to the daily data of the patient; S12, dividing the sorted daily data of the patient according to the divided time intervals to form a daily data set corresponding to each divided time interval; S13, according to the daily data sets corresponding to the divided time intervals, based on the pre-set data extraction strategy, determine the extraction point extraction scheme corresponding to each divided time interval, and extract the data of the corresponding extraction point in the daily data sets corresponding to each divided time interval according to the extraction point extraction scheme corresponding to each divided time interval; The pre-set data extraction strategy is: According to the daily data set corresponding to a certain divided time interval, the data type corresponding to each data in the daily data set is identified, and the data corresponding to each data type is determined; In the corresponding divided time interval, a number of extraction point extraction schemes are randomly generated and a corresponding extraction point set is formed; According to the extraction schemes of each extraction point in the extraction point set, based on the preset extraction rationality calculation formula, the extraction rationality corresponding to each extraction point extraction scheme is calculated in turn, and after each calculation is completed, the extraction rationality calculated this time is compared with the preset rationality threshold. If the extraction rationality is greater than the preset rationality threshold, the extraction point extraction scheme corresponding to the extraction rationality is judged to be the extraction rationality scheme corresponding to the divided time interval. Otherwise, the extraction rationality calculation of the next extraction point extraction scheme is continued; The calculation formula for the extraction rationality is: In the formula, H is the rationality of extraction, L i is the representative value of the i-th sampling point in the sampling point extraction scheme, W iq is the weight value of the qth data type where the i-th extraction point is located in the extraction point extraction scheme, T q is the data time interval corresponding to the qth data type, Q is the total number of data types in a certain partitioned time interval, and Z i is the time rationality of the i-th extraction point in the extraction point extraction scheme, t i is the time corresponding to the i-th extraction point in the extraction point extraction scheme; α and β are the corresponding weighting coefficients; S14, according to the data extracted from each divided time interval, based on a preset data similarity calculation formula, calculating the data similarity corresponding to the abnormal patient data corresponding to each divided time interval; If the extracted data is a numerical value, the preset data similarity calculation formula is: Where, X ts is the data similarity between the partition time interval t and the partition time interval s, N is the number of extraction points, d(X tj ,X sk ) is the characteristic distance between the data value of the j-th extracted data in the partition time interval t and the data value of the k-th extracted data in the partition time interval s, ∝ is the attenuation factor corresponding to the characteristic clustering; If the extracted data is non-numeric, the preset data similarity calculation formula is: Wherein, F is the maximum value of the number of data corresponding to the same content in the extraction points between the divided time interval t and the divided time interval s; S15, according to the data similarity of the abnormal patient data corresponding to each divided time interval, the abnormal patient data corresponding to the divided time interval whose data similarity is less than a preset similarity threshold value are divided into the same category of data; If the data similarity X between a certain partition time interval A and the partition time intervals B and C AB and X AC are all less than the preset similarity threshold, but the data similarity X between the divided time intervals B and C is BC If it is greater than the preset similarity threshold, then compare X AB and X AC The size between; If X AB <X AC , then the abnormal patient data corresponding to the divided time interval A and the divided time interval B are divided into the same type of data; If X AB >X AC , then the abnormal patient data corresponding to the divided time interval A and the divided time interval C are divided into the same type of data; If X AB =X AC , a random number P from 1 to 1 is randomly generated. When P is greater than the preset random value, the abnormal patient data corresponding to the divided time interval A and the divided time interval B are divided into the same type of data. Otherwise, the abnormal patient data corresponding to the divided time interval A and the divided time interval C are divided into the same type of data. S16. After completing the classification of all the abnormal patient data corresponding to the divided time intervals, randomly select a patient abnormal data corresponding to a divided time interval from each data of the same category, and select the data extracted from the corresponding divided time interval, and identify the dimension corresponding to the extracted data, and identify the abnormal cause dimension corresponding to the same category of data corresponding to the extracted data.

3. A postoperative monitoring and early warning system for patients according to claim 2, characterized in that: The retrieving from the database all abnormal cause dimensions corresponding to each detection indicator data when an abnormality occurs includes all abnormal cause dimensions corresponding to when each detection indicator data occurs abnormally alone and when abnormalities occur simultaneously between each detection indicator data.

4. A postoperative monitoring and early warning system for patients according to claim 3, characterized in that: The server also includes: The first reporting module is used to generate a medical solution report corresponding to the corresponding medical staff according to the abnormal cause results; The second reporting module is used to formulate a user maintenance report that the corresponding patient needs to complete when executing the medical care solution report based on the abnormal cause results formed and the medical care solution report to be executed; The sending module is used to send the medical care solution report and the user maintenance report to the medical care end and the user end respectively.

5. A postoperative monitoring and early warning system for patients according to claim 4, characterized in that: The first reporting module comprises: An evaluation module is used to retrieve the surgery-related data corresponding to the patient according to the abnormal cause result, evaluate the impact of the abnormal cause result on the patient, and retrieve the medical care solution report corresponding to the impact according to the impact; The feedback update module is used to feed back the medical solution report to the medical end, receive reasonable feedback information of the measures fed back by the medical staff through the medical end, and update the medical solution report according to the reasonable feedback information of the measures.

6. A postoperative monitoring and early warning method for a patient, using a postoperative monitoring and early warning system for a patient according to any one of claims 1 to 5, characterized in that: The following steps are involved: S1. Collect various test index data corresponding to the postoperative patients in the current time period to form a corresponding postoperative index data set of the current patients; S2. According to the postoperative indicator data set of the current patient, based on the preset abnormal thresholds corresponding to the various detection indicator data, identify the various abnormal detection indicator data in the postoperative indicator data set of the current patient; S3. According to the identified abnormal detection index data in the postoperative index data set of the current patient, all abnormal cause dimensions corresponding to each detection index data when the abnormality occurs are retrieved from the database; S4, retrieve the habit information corresponding to the patient and the daily data of the patient corresponding to the current time period to the previous time period, and cluster the daily data of the patient corresponding to each time period based on the preset patient data clustering strategy to form a daily data set of the patient corresponding to each abnormal cause dimension; S5. According to the patient daily data sets corresponding to each abnormal etiology dimension, the data in the corresponding patient daily data sets are analyzed for etiology in turn to form corresponding abnormal etiology results; S6. When the corresponding abnormal cause result is formed or each abnormal detection index data in the postoperative index data set of the current patient is identified, an alarm signal is sent to the user end and the medical end.

Images