Multi-edge cooperative computing and resource allocation method for online diagnosis of chronic diseases

Abstract

The invention relates to the technical field of computing resource allocation, in particular to a chronic disease online diagnosis-oriented multilateral cooperative computing and resource allocation method, which comprises the following steps of: through multilateral cooperative computing, acquiring a score table set corresponding to each online diagnosis of each target patient in a current treatment time period, the blood pressure data and the electrocardio data in the current treatment time period are acquired; performing score distribution analysis processing based on the local score value sequence corresponding to the target patient; segmenting the local score value sequence, the blood pressure data sequence and the electrocardiogram data sequence corresponding to the target patient; correcting the target comprehensive score value; and according to the correction score value corresponding to each target patient and the distance condition between each target patient and the hospital, determining a resource allocation priority corresponding to each target patient, and performing resource allocation based on the resource allocation priority. According to the invention, online medical resource allocation is realized, and the rationality of online medical resource allocation is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of computing resource allocation, and particularly to a multi-party collaborative computing and resource allocation method for online diagnosis of chronic diseases. Background Art

[0002] The treatment process of chronic disease patients is often relatively long, and during the treatment process, it is often necessary to monitor the development of chronic disease patients. In order to simplify the treatment process and improve the treatment efficiency, online diagnosis is often carried out during the treatment process. Since there are many chronic disease patients and the online medical resources are limited, it is often difficult for hospitals to conduct online diagnoses for all chronic disease patients who send online diagnosis applications in real time. Therefore, at the current moment, there are often many online diagnosis applications from chronic disease patients, that is, at the current moment, there are often many chronic disease patients who need to allocate online medical resources. Among them, online medical resources often include the medical resources required for online diagnosis, which belong to computing resources. Currently, when allocating online medical resources to chronic disease patients who need to allocate online medical resources at the current moment, the commonly used method is: for chronic disease patients who need to allocate online medical resources at the current moment, the online medical resources are often allocated according to the time when these patients send online diagnosis applications, that is, the earlier a patient sends an online diagnosis application, the earlier the online medical resources are allocated to the patient.

[0003] However, when allocating online medical resources based on the online diagnosis application time, the following technical problems often occur: The monitoring data abnormality situations of different chronic disease patients who send online diagnosis applications are often different, and the urgency of resource allocation for patients with different monitoring data abnormality situations is often different. If online medical resources are allocated only based on the online diagnosis application time, it may result in allocating online medical resources to some patients with relatively serious monitoring data abnormality situations relatively late, thus causing the online medical resources not to be allocated to these patients in a timely manner, and further resulting in poor rationality of online medical resource allocation. Summary of the Invention

[0004] In order to solve the technical problem of poor rationality of online medical resource allocation, the present invention proposes a multi-party collaborative computing and resource allocation method for online diagnosis of chronic diseases.

[0005] In a first aspect, the present invention provides a multi-party collaborative computing and resource allocation method for online diagnosis of chronic diseases, and the method includes: Obtain a set of score sheets corresponding to each online diagnosis during the current treatment period of each target patient, as well as the blood pressure data and electrocardiogram data during the current treatment period of the target patient, to obtain a sequence of score sheet sets, a sequence of blood pressure data, and a sequence of electrocardiogram data corresponding to each target patient; Based on the sequence of score sheet sets corresponding to each target patient, construct a sequence of local score values, and perform score distribution analysis processing based on the sequence of local score values corresponding to each target patient to obtain the target comprehensive score value; Segment the sequence of local score values corresponding to each target patient to obtain segments of local score values; According to all the segments of local score values corresponding to each target patient, segment the blood pressure data sequence and the electrocardiogram data sequence corresponding to each target patient to obtain segments of blood pressure data and segments of electrocardiogram data; According to all the segments of local score values, all the segments of blood pressure data, and all the segments of electrocardiogram data corresponding to each target patient, correct the target comprehensive score value corresponding to each target patient to obtain a corrected score value; According to the corrected score value corresponding to each target patient and the distance between each target patient and the hospital, determine the resource allocation priority corresponding to each target patient, and perform resource allocation based on the resource allocation priority.

[0006] Combined with the first aspect above, in a possible implementation manner, the constructing a sequence of local score values based on the sequence of score sheet sets corresponding to each target patient includes: Determine any one target patient as a marked patient, and form a set of score values by the score values corresponding to all the score sheets in each score sheet set in the sequence of score sheet sets corresponding to the marked patient, to obtain the sequence of score value sets corresponding to the marked patient; Determine the mean value of all the score values in each score value set in the sequence of score value sets corresponding to the marked patient as the local score value, to obtain the sequence of local score values corresponding to the marked patient.

[0007] Combined with the first aspect above, in a possible implementation manner, the performing score distribution analysis processing based on the sequence of local score values corresponding to each target patient to obtain the target comprehensive score value includes: Determine any one target patient as a marked patient, and perform linear fitting on the sequence of local score values corresponding to the marked patient to obtain the fitting line corresponding to the marked patient; According to the slope of the fitting line corresponding to the marked patient and the last local score value in the sequence of local score values corresponding to the marked patient, determine the initial comprehensive score value corresponding to the marked patient; According to the number of score sheets in each score sheet set in the sequence of score sheet sets corresponding to the marked patient, determine the score consistency corresponding to the marked patient; According to the initial comprehensive score value and the score consistency corresponding to the marked patient, determine the target comprehensive score value corresponding to the marked patient, where both the initial comprehensive score value and the score consistency are positively correlated with the target comprehensive score value.

[0008] Combined with the above first aspect, in a possible implementation manner, determining the initial comprehensive score value corresponding to the marked patient according to the slope of the fitting line corresponding to the marked patient and the last local score value in the local score value sequence corresponding to the marked patient includes: Determining the normalized value of the slope of the fitting line corresponding to the marked patient as the target slope corresponding to the marked patient; Determining the last local score value in the local score value sequence corresponding to the marked patient as the recent score value corresponding to the marked patient; Normalizing the product of the target slope and the recent score value corresponding to the marked patient to obtain the initial comprehensive score value corresponding to the marked patient.

[0009] Combined with the above first aspect, in a possible implementation manner, determining the scoring consistency corresponding to the marked patient according to the number of scoring tables in each scoring table set sequence corresponding to the marked patient includes: Determining the number of scoring tables in each scoring table set in the scoring table set sequence corresponding to the marked patient as the target representative number to obtain the target representative number sequence corresponding to the marked patient; Rounding the mean of all modes in the target representative number sequence corresponding to the marked patient to obtain the reference number corresponding to the marked patient; Determining any scoring table in the scoring table set sequence corresponding to the marked patient as the marked scoring table, and screening out the scoring tables of the same type as the marked scoring table from the scoring table set sequence corresponding to the marked patient to form the scoring table group corresponding to the marked scoring table, so as to obtain the scoring table group corresponding to each scoring table; Determining the number of scoring tables in the scoring table group corresponding to each scoring table as the standard number corresponding to each scoring table; Screening out the reference number of the largest standard numbers from the standard numbers corresponding to all scoring tables in the scoring table set sequence corresponding to the marked patient to form the standard number set corresponding to the marked patient; If the standard number corresponding to the scoring table belongs to the standard number set, determining the scoring table as a common scoring table, and determining the scoring table set with common scoring tables as the common scoring table set; If the standard number corresponding to the scoring table does not belong to the standard number set, determining the scoring table as a special scoring table, and determining the scoring table set in which all scoring tables are special scoring tables as the special scoring table set; Determine the scoring consistency corresponding to the marked patient according to the number of common scoring form sets in the scoring form set sequence corresponding to the marked patient, the differences between different common scoring form sets, and the differences between the common scoring form sets and the special scoring form sets.

[0010] Combined with the first aspect above, in a possible implementation manner, segmenting the local scoring value sequence corresponding to each target patient to obtain local scoring value segments includes: Segment the local scoring value sequence corresponding to each target patient through the APCA algorithm to obtain multiple local scoring value segments corresponding to each target patient.

[0011] Combined with the first aspect above, in a possible implementation manner, segmenting the blood pressure data sequence and the electrocardiogram data sequence corresponding to each target patient according to all the local scoring value segments corresponding to each target patient to obtain blood pressure data segments and electrocardiogram data segments includes: Determine any one target patient as the marked patient, and determine the time period corresponding to each local scoring value segment of the marked patient as the local time period; In the blood pressure data sequence of the marked patient, the blood pressure data corresponding to the collection moments belonging to the same local time period constitute the blood pressure data segment corresponding to this local time period, and multiple blood pressure data segments corresponding to the marked patient are obtained; In the electrocardiogram data sequence of the marked patient, the electrocardiogram data corresponding to the collection moments belonging to the same local time period constitute the electrocardiogram data segment, and multiple electrocardiogram data segments corresponding to the marked patient are obtained.

[0012] Combined with the first aspect above, in a possible implementation manner, correcting the target comprehensive scoring value corresponding to each target patient according to all the local scoring value segments, all the blood pressure data segments, and all the electrocardiogram data segments corresponding to each target patient to obtain the corrected scoring value includes: Determine any one target patient as the marked patient, and determine the mean value of all the local scoring values within each local scoring value segment of the marked patient as the overall scoring index, and obtain the overall scoring index sequence corresponding to the marked patient; Perform abnormal analysis and processing on each blood pressure data segment to obtain the target blood pressure abnormal factor corresponding to the local time period corresponding to each blood pressure data segment; Similarly, perform abnormal analysis and processing on each electrocardiogram data segment to obtain the target electrocardiogram abnormal factor corresponding to the local time period corresponding to each electrocardiogram data segment; Determine the product of the target blood pressure abnormal factor and the target electrocardiogram abnormal factor corresponding to each local time period as the target overall abnormal factor corresponding to each local time period; Construct the target overall abnormal factor sequence corresponding to the marked patient from all local time periods corresponding to the marked patient, where the target overall abnormal factor sequence is composed of the target overall abnormal factors corresponding to all local time periods corresponding to the marked patient; Normalize the Pearson correlation coefficient between the target overall abnormal factor sequence corresponding to the marked patient and the overall scoring index sequence to obtain the target similarity corresponding to the marked patient; Normalize the product of the target similarity corresponding to the marked patient and the target comprehensive score value to obtain the corrected score value corresponding to the marked patient.

[0013] Combined with the first aspect above, in a possible implementation manner, the performing abnormal analysis processing on each blood pressure data segment to obtain the target blood pressure abnormal factor corresponding to the local time period corresponding to each blood pressure data segment includes: Perform blood pressure abnormality detection on each blood pressure data segment to obtain abnormal blood pressure data, and form sub - segments of abnormal blood pressure data from the continuous abnormal blood pressure data within each blood pressure data segment; Cluster all sub - segments of abnormal blood pressure data within each blood pressure data segment to obtain clusters of abnormal blood pressure, and select the cluster of abnormal blood pressure with the largest total duration corresponding to each blood pressure data segment from all clusters of abnormal blood pressure corresponding to each blood pressure data segment as the target cluster of abnormal blood pressure corresponding to each blood pressure data segment; Determine the target blood pressure abnormal factor corresponding to the local time period corresponding to each blood pressure data segment according to the total number of all abnormal blood pressure data in each blood pressure data segment and the number of abnormal blood pressure data in its corresponding target cluster of abnormal blood pressure.

[0014] Combined with the first aspect above, in a possible implementation manner, the determining the resource allocation priority corresponding to each target patient according to the corrected score value corresponding to each target patient and the distance between each target patient and the hospital includes: Determine any one target patient as the marked patient, classify each hospital to obtain the target level corresponding to each hospital; Select the hospital closest to the address of the marked patient from all hospitals corresponding to each target level as the reference hospital of the marked patient at each target level; Determine the distance between the reference hospital of the marked patient at each target level and the address of the marked patient as the marked distance of the marked patient at each target level; Determine the richness of offline medical resources corresponding to the marked patient according to the marked distances of the marked patient at all target levels, where there is a negative correlation between the marked distance and the richness of offline medical resources; Based on the richness of offline medical resources corresponding to the marked patient and the corrected score value, determine the resource allocation priority corresponding to the marked patient, where the richness of offline medical resources is negatively correlated with the resource allocation priority, and the corrected score value is positively correlated with the resource allocation priority.

[0015] In a second aspect, the present invention provides a multi - party collaborative computing and resource allocation system for online diagnosis of chronic diseases. The system includes: A data acquisition module, configured to acquire a set of score sheets corresponding to each online diagnosis during the current treatment period of each target patient, as well as the blood pressure data and electrocardiogram data during the current treatment period of each target patient, to obtain a sequence of score sheet sets, a blood pressure data sequence, and an electrocardiogram data sequence corresponding to each target patient; A construction and processing module, configured to construct a local score value sequence based on the sequence of score sheet sets corresponding to each target patient, and perform score distribution analysis and processing based on the local score value sequence corresponding to each target patient to obtain a target comprehensive score value; A score value segmentation module, configured to segment the local score value sequence corresponding to each target patient to obtain local score value segments; A data sequence segmentation module, configured to segment the blood pressure data sequence and the electrocardiogram data sequence corresponding to each target patient according to all local score value segments corresponding to each target patient to obtain blood pressure data segments and electrocardiogram data segments; A score correction module, configured to correct the target comprehensive score value corresponding to each target patient according to all local score value segments, all blood pressure data segments, and all electrocardiogram data segments corresponding to each target patient to obtain a corrected score value; A determination and resource allocation module, configured to determine the resource allocation priority corresponding to each target patient according to the corrected score value corresponding to each target patient and the distance between each target patient and the hospital, and perform resource allocation based on the resource allocation priority.

[0016] In a third aspect, a server is provided, including a memory and a processor. The memory is used to store executable program code, and the processor is used to call and run the executable program code from the memory, so that the device executes the method in the first aspect or any possible implementation manner of the first aspect.

[0017] In a fourth aspect, a computer program product is provided. The computer program product includes: computer program code, which when running on a computer, causes the computer to execute the method in the first aspect or any possible implementation manner of the first aspect.

[0018] In a fifth aspect, a computer-readable storage medium is provided. The computer-readable storage medium stores computer program code, and when the computer program code runs on a computer, the computer is caused to execute the method in the first aspect or any possible implementation manner of the first aspect as described above.

[0019] The present invention has the following beneficial effects: The multi-party collaborative computing and resource allocation method for chronic disease online diagnosis of the present invention realizes online medical resource allocation, solves the technical problem of poor rationality of online medical resource allocation, and improves the rationality of online medical resource allocation. Specifically, when the present invention performs online medical resource allocation, it comprehensively considers a set of score tables representing abnormal monitoring data under the diagnosis time corresponding to different online diagnoses, and analyzes the score table set sequence, blood pressure data sequence, and electrocardiogram data sequence, so as to adaptively quantify the resource allocation priority corresponding to each target patient, and finally realizes resource allocation based on the resource allocation priority, and to a certain extent improves the rationality of online medical resource allocation. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0021] Figure 1 is a flowchart of a multi-party collaborative computing and resource allocation method for chronic disease online diagnosis of the present invention; Figure 2 is a schematic diagram of the composition structure of a multi-party collaborative computing and resource allocation system for chronic disease online diagnosis of the present invention; Figure 3 is a schematic diagram of the structure of a computer device of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0022] In order to further elaborate on the technical means and effects adopted by the present invention to achieve the intended invention purpose, the following, in combination with the accompanying drawings and preferred embodiments, details the specific implementation manners, structures, features, and effects of the technical solutions proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, the specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0023] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the technical field to which this invention belongs.

[0024] The multi - party collaborative computing and resource allocation method for chronic disease online diagnosis is a method that comprehensively uses technical means such as cloud computing, edge computing, and the Internet of Things to realize the health management service for chronic disease patients. Its aim is to improve the efficiency and accuracy of chronic disease online diagnosis through multi - party collaborative computing and optimize resource allocation at the same time. Among them, multi - party collaborative computing mainly realizes the interconnection and efficient sharing of information by integrating data resources from different platforms. The data used in this invention often comes from different platforms, and to a certain extent, multi - party collaborative computing is achieved.

[0025] Reference Figure 1 , which shows the flow of some embodiments of a multi - party collaborative computing and resource allocation method for chronic disease online diagnosis according to the present invention. The multi - party collaborative computing and resource allocation method for chronic disease online diagnosis includes the following steps: Step S1, obtain the set of score sheets corresponding to each online diagnosis of each target patient during their current treatment period, as well as their blood pressure data and electrocardiogram data during their current treatment period, to obtain a sequence of score sheet sets, a sequence of blood pressure data, and a sequence of electrocardiogram data corresponding to each target patient.

[0026] Among them, the target patient can be a chronic disease patient who has sent an online diagnosis application at the current moment and has not been allocated online medical resources. For example, the chronic disease patient can be a patient with coronary heart disease. The current treatment time period of the target patient can be the time period corresponding to the treatment process of the target patient with the current moment as the end moment, that is, it can represent the process of the target patient undergoing chronic disease treatment. Online diagnosis can be that a doctor makes a disease diagnosis in an online manner. For example, online diagnosis can be achieved through video calls between the patient and the doctor. In actual situations, the more severe the patient's condition, the more abnormal the corresponding monitoring data tends to be. Since the scoring table is mainly used to evaluate the severity of the patient's condition and the prognosis risk, the scoring table can, to a certain extent, represent the abnormal situation of the monitoring data. Therefore, the set of scoring tables corresponding to the online diagnosis can include the scoring tables used by the doctor during the online diagnosis, which can represent the abnormal situation of the monitoring data at the diagnosis time corresponding to the online diagnosis. For example, the scoring table can be, but is not limited to: GRACE (Global Registry of Acute Coronary Events) scoring table, TIMI (Thrombolysis In Myocardial Infarction) scoring table, and SYNTAX (Synergy between PCI with TAXUS and Cardiac Surgery) scoring table. Blood pressure data can represent the blood pressure situation of the patient, and it can be normalized data. For example, the blood pressure data can be represented by the normalized value of the sum of the systolic blood pressure and the diastolic blood pressure of the patient at the same moment. Electrocardiogram data can represent the cardiac activity situation, and it usually exists in the form of numerical values. For example, the electrocardiogram data can be represented by the normalized value of the accumulated value of the P wave (atrial depolarization wave) value, QRS wave (ventricular depolarization wave) value, and T wave (ventricular repolarization wave) value of the patient at the same moment. The P wave value, QRS wave value, and T wave value can be numerical values obtained by converting the P wave, QRS wave, and T wave respectively. The P wave represents the depolarization process of the atrium and is usually related to atrial contraction. The QRS wave represents the ventricular depolarization process, which is the most significant part of the electrocardiogram and is usually related to ventricular contraction. The T wave represents the ventricular repolarization process and is usually related to ventricular relaxation.

[0027] As an example, this step may include the following steps: First step, obtain the set of scoring tables corresponding to each online diagnosis of each target patient during their current treatment time period, and obtain the sequence of sets of scoring tables corresponding to each target patient.

[0028] For example, during the current treatment period of a target patient, the score sheets used in the completed online diagnoses of the target patient can be collected to form a set of score sheets corresponding to the online diagnosis, and the sets of score sheets corresponding to all the online diagnoses of the target patient during the current treatment period are used to form a sequence of sets of score sheets. Among them, the sequence of sets of score sheets is a time series.

[0029] It should be noted that the same chronic disease patient may receive online diagnoses from different doctors in different hospitals during the long-term treatment process, and there are often certain differences in the types and quantities of score sheets used by different doctors during online diagnoses. For example, some doctors may use two score sheets during online diagnoses, while some doctors may use three score sheets during online diagnoses. Even if different doctors use the same number of score sheets, the types of their score sheets may also be different. Therefore, obtaining the sets of score sheets corresponding to different online diagnoses can facilitate subsequent data integration and analysis, and thus facilitate subsequent online medical resource allocation.

[0030] Step 2: Obtain the blood pressure data and electrocardiogram data of each target patient during their current treatment period to obtain a blood pressure data sequence and an electrocardiogram data sequence corresponding to each target patient.

[0031] For example, the blood pressure data and electrocardiogram data of the target patient during their current treatment period can be collected through a blood pressure monitoring bracelet and a smart watch respectively, and all the blood pressure data collected during the current treatment period are used to form a blood pressure data sequence, and all the electrocardiogram data collected during the current treatment period are used to form an electrocardiogram data sequence. Among them, the blood pressure data sequence and the electrocardiogram data sequence are time series.

[0032] Step S2: Based on the sequence of sets of score sheets corresponding to each target patient, construct a sequence of local score values, and perform score distribution analysis processing based on the sequence of local score values corresponding to each target patient to obtain a target comprehensive score value.

[0033] As an example, this step may include the following steps: First step: Determine any one target patient as a marked patient, and form a set of score values by using the score values corresponding to all the score sheets in each set of score sheets in the sequence of sets of score sheets corresponding to the marked patient, so as to obtain a sequence of sets of score values corresponding to the marked patient.

[0034] Among them, the score value corresponding to the score sheet can be a value calculated through the score sheet. The larger the value, the more serious the patient's condition may be, and the more abnormal the patient's monitoring data may be.

[0035] It should be noted that the sets of score values in the sequence of sets of score values can correspond one by one to the sets of score sheets in the sequence of sets of score sheets.

[0036] In the second step, the mean value of all the score values in each score value set in the score value set sequence corresponding to the above-marked patients is determined as the local score value, and a local score value sequence corresponding to the above-marked patients is obtained.

[0037] It should be noted that the local score values in the local score value sequence can correspond one by one to the score value sets in the score value set sequence.

[0038] In the third step, any one target patient is determined as the marked patient, and the local score value sequence corresponding to the above-marked patient is linearly fitted to obtain a fitted line corresponding to the above-marked patient.

[0039] For example, with the serial number of the local score value in the local score value sequence as the abscissa and the local score value in the local score value sequence corresponding to the marked patient as the ordinate, a scatter plot is constructed, and the scatter plot is linearly fitted to obtain a fitted line corresponding to the marked patient.

[0040] In the fourth step, according to the slope of the fitted line corresponding to the above-marked patient and the last local score value in the local score value sequence corresponding to the above-marked patient, determining the initial comprehensive score value corresponding to the above-marked patient may include the following sub-steps: In the first sub-step, the normalized value of the slope of the fitted line corresponding to the above-marked patient is determined as the target slope corresponding to the above-marked patient.

[0041] In the second sub-step, the last local score value in the local score value sequence corresponding to the above-marked patient is determined as the recent score value corresponding to the above-marked patient.

[0042] It should be noted that the last local score value in the local score value sequence corresponding to the marked patient can often characterize the abnormal situation of the monitoring data obtained during the marked patient's most recent online diagnosis.

[0043] In the third sub-step, the product of the target slope and the recent score value corresponding to the above-marked patient is normalized to obtain the initial comprehensive score value corresponding to the above-marked patient.

[0044] For example, the formula for determining the initial comprehensive score value corresponding to the marked patient can be: ; where P is the initial comprehensive score value corresponding to the marked patient. is the normalization function. k is the target slope corresponding to the marked patient. A is the recent score value corresponding to the marked patient.

[0045] It should be noted that when k is larger, it often indicates that the score value representing the disease risk during the current treatment period of the marked patient shows an increasing trend, often indicating that the condition of the marked patient is more likely to deteriorate, often indicating that the abnormal degree of the monitoring data of the marked patient who sends an online diagnosis application is relatively larger, and often indicating that it is more necessary to allocate online medical resources to the marked patient more quickly. When A is larger, it often indicates that the score value representing the disease risk situation of the most recent online diagnosis during the current treatment period of the marked patient is higher, often indicating that the condition of the marked patient is more likely to deteriorate at the current stage, and often indicating that the abnormal degree of the monitoring data of the marked patient who sends an online diagnosis application is relatively larger. Therefore, when P is larger, it often indicates that it is more necessary to allocate online medical resources to the marked patient more quickly.

[0046] The fifth step, determining the scoring consistency corresponding to the above-mentioned marked patient according to the number of scoring tables in each scoring table set in the scoring table set sequence corresponding to the above-mentioned marked patient may include the following sub-steps: The first sub-step is to determine the number of scoring tables in each scoring table set in the scoring table set sequence corresponding to the above-mentioned marked patient as the target representative number, and obtain the target representative number sequence corresponding to the above-mentioned marked patient.

[0047] Among them, the target representative numbers in the target representative number sequence can correspond one by one to the scoring table sets in the scoring table set sequence.

[0048] The second sub-step is to round the average value of all the modes in the target representative number sequence corresponding to the above-mentioned marked patient to obtain the reference number corresponding to the above-mentioned marked patient.

[0049] It should be noted that there are often multiple or one mode in a set of data. Therefore, there may be multiple or one mode in the target representative number sequence. Therefore, the present invention embodiment performs an average value calculation when using the modes in the target representative number sequence.

[0050] The third sub-step is to determine any one scoring table in the scoring table set sequence corresponding to the above-mentioned marked patient as the marked scoring table, and screen out the scoring tables with the same type as the above-mentioned marked scoring table from the scoring table set sequence corresponding to the above-mentioned marked patient to form the scoring table group corresponding to the above-mentioned marked scoring table, so as to obtain the scoring table group corresponding to each scoring table.

[0051] For example, if the set sequence of score sheets corresponding to a marked patient includes 4 sets of score sheets, and these 4 sets of score sheets are, in sequence: {the first GRACE score sheet, the first TIMI score sheet, the first SYNTAX score sheet}, {the second GRACE score sheet, the second TIMI score sheet}, {the third TIMI score sheet, the third SYNTAX score sheet}, {the fourth GRACE score sheet, the fourth TIMI score sheet, the fourth SYNTAX score sheet}, and the marked score sheet is the second GRACE score sheet, then the score sheet group corresponding to the second GRACE score sheet may include: all the score sheet sets in this set sequence that contain the GRACE score sheet. Therefore, the score sheet group corresponding to the second GRACE score sheet may be {the first GRACE score sheet, the second GRACE score sheet, the fourth GRACE score sheet}.

[0052] The fourth sub-step is to determine the number of score sheets in the score sheet group corresponding to each score sheet as the standard number corresponding to each score sheet.

[0053] It should be noted that in the set sequence of score sheets corresponding to a marked patient, the standard numbers corresponding to the same type of score sheets are often the same. Among the standard numbers corresponding to all the score sheets in the set sequence of score sheets corresponding to a marked patient, the types of standard numbers often correspond one by one to the types of score sheets, that is, one type of score sheet often corresponds to one type of standard number.

[0054] The fifth sub-step is to screen out the largest standard numbers among the standard numbers corresponding to all the score sheets in the above-mentioned set sequence of score sheets corresponding to a marked patient to form the set of standard numbers corresponding to the above-mentioned marked patient.

[0055] For example, the different standard numbers corresponding to all the score sheets in the set sequence of score sheets corresponding to a marked patient can be sorted in descending order, and the sequence obtained at this time is recorded as the simplified score sequence. The first reference number of standard numbers in the simplified score sequence form the set of standard numbers.

[0056] The sixth sub-step is that if the standard number corresponding to a score sheet belongs to the above-mentioned set of standard numbers, then the score sheet is determined as a commonly used score sheet, and the score sheet set in which there is a commonly used score sheet is determined as the set of commonly used score sheets.

[0057] Among them, there are commonly used score sheets in the set of commonly used score sheets.

[0058] It should be noted that the score sheets in the set of commonly used score sheets are often the score sheets that have been used in most online diagnoses.

[0059] The seventh sub-step: If the standard quantity corresponding to the score table does not belong to the above standard quantity set, then determine the score table as a special score table, and determine the score table set in which all the score tables are special score tables as the special score table set.

[0060] Among them, there is no common score table in the special score table set.

[0061] It should be noted that the score tables in the special score table set are often score tables that have been used less frequently in online diagnosis.

[0062] The eighth sub-step: Determine the scoring consistency corresponding to the above-marked patient according to the number of common score table sets in the score table set sequence corresponding to the above-marked patient, the differences between different common score table sets, and the differences between the common score table set and the special score table set.

[0063] For example, the formula for determining the scoring consistency corresponding to the marked patient can be: ; ; ; Among them, L is the scoring consistency corresponding to the marked patient. S is the number of score table sets in the score table set sequence corresponding to the marked patient. n is the number of common score table sets in the score table set sequence corresponding to the marked patient. is the natural exponential function. B represents the average value of the differences between different common score table sets in the score table set sequence corresponding to the marked patient. C represents the average value of the differences between the common score table set and the special score table set in the score table set sequence corresponding to the marked patient. i is the serial number of the common score table set in the score table set sequence corresponding to the marked patient. represents the difference between the i-th common score table set and the (i + 1)-th common score table set. is the number of common score tables in the union of the i-th common score table set and the (i + 1)-th common score table set in the score table set sequence corresponding to the marked patient. is the number of common score tables in the intersection of the i-th common score table set and the (i + 1)-th common score table set in the score table set sequence corresponding to the marked patient. N is the number of special score table sets in the score table set sequence corresponding to the marked patient. j is the serial number of the special score table set in the score table set sequence corresponding to the marked patient. represents the difference between the i-th common score table set and the j-th special score table set. is the number of score tables in the union of the i-th common score table set and the j-th special score table set in the score table set sequence corresponding to the marked patient. is the number of score tables in the intersection of the \(i\)-th common score table set and the \(j\)-th special score table set in the sequence of score table sets corresponding to the marked patient.

[0064] It should be noted that when is larger, it often indicates that the set of common score tables corresponding to the marked patient is relatively larger, and it often indicates that the types of score tables used in different online diagnoses of the marked patient are relatively more consistent. When \(B\) is smaller, it often indicates that the different common score table sets corresponding to the marked patient are more similar, and it often indicates that the contents in the different common score tables used in the online diagnosis of the marked patient are relatively more similar, and it often indicates that the common disease analysis directions of the marked patient are relatively more consistent. When \(C\) is smaller, it often indicates that the set of common score tables corresponding to the marked patient is more similar to the set of special score tables, and it often indicates that the contents in the common score tables and special score tables used in the online diagnosis of the marked patient are relatively more similar, and it often indicates that the disease analysis directions reflected by the common score tables and special score tables of the marked patient are relatively more consistent. Therefore, when \(L\) is larger, it often indicates that the types of score tables used in different online diagnoses of the marked patient are relatively more unified, and it often indicates that the scoring criteria used in different online diagnoses of the marked patient are more similar, and it often indicates that the sequence of score table sets corresponding to the marked patient is more meaningful for reference.

[0065] Step 6: Determine the target comprehensive score value corresponding to the above-mentioned marked patient according to the initial comprehensive score value and score consistency corresponding to the above-mentioned marked patient.

[0066] Among them, both the initial comprehensive score value and the score consistency can have a positive correlation with the target comprehensive score value.

[0067] For example, the formula for determining the target comprehensive score value corresponding to the marked patient can be: ; where is the target comprehensive score value corresponding to the marked patient. is the normalization function. \(P\) is the initial comprehensive score value corresponding to the marked patient. \(L\) is the score consistency corresponding to the marked patient.

[0068] It should be noted that when \(P\) is larger, it often indicates that it is more necessary to allocate online medical resources to the marked patient faster. \(L\) can be used as the weight of \(P\), that is, \(L\) can be used to correct \(P\). When \(L\) is larger, it often indicates that the types of score tables used in different online diagnoses of the marked patient are relatively more unified, and it often indicates that the scoring criteria used in different online diagnoses of the marked patient are more similar, and it often indicates that the sequence of score table sets corresponding to the marked patient is more meaningful for reference. Therefore, when is larger, it often indicates that it is more necessary to allocate online medical resources to the marked patient faster.

[0069] Step S3: Segment the local score value sequence corresponding to each target patient to obtain local score value segments.

[0070] As an example, the APCA (Adaptive Piecewise Constant Approximation) algorithm can be used to segment the local score value sequence corresponding to each target patient to obtain multiple local score value segments corresponding to each target patient. Among them, the local score value segment is a segment of the local score value sequence.

[0071] It should be noted that the elements within the same segment obtained by segmenting through the APCA algorithm are often relatively similar.

[0072] Step S4: Segment the blood pressure data sequence and the electrocardiogram data sequence corresponding to each target patient according to all the local score value segments corresponding to each target patient to obtain blood pressure data segments and electrocardiogram data segments.

[0073] As an example, this step may include the following steps: First step: Determine any one of the target patients as the marked patient, and determine the time period corresponding to each local score value segment of the above-mentioned marked patient as the local time period.

[0074] It should be noted that the local score values in the local score value sequence can correspond one by one to the score table sets in the score table set sequence. The score table sets in the score table set sequence can correspond one by one to the online diagnosis. Therefore, the local score values in the local score value sequence can be related to the online diagnosis.

[0075] Among them, the start time of the time period corresponding to the local score value segment can be the start time of the online diagnosis corresponding to the first local score value in the local score value segment. The end time of the time period corresponding to the local score value segment can be the end time of the online diagnosis corresponding to the last local score value in the local score value segment.

[0076] Second step: In the blood pressure data sequence of the above-mentioned marked patient, the blood pressure data corresponding to the acquisition times belonging to the same local time period form the blood pressure data segment corresponding to this local time period, and multiple blood pressure data segments corresponding to the above-mentioned marked patient are obtained.

[0077] Third step: In the electrocardiogram data sequence of the above-mentioned marked patient, the electrocardiogram data corresponding to the acquisition times belonging to the same local time period form the electrocardiogram data segment, and multiple electrocardiogram data segments corresponding to the above-mentioned marked patient are obtained.

[0078] Step S5: Based on all local score value segments, all blood pressure data segments, and all electrocardiogram data segments corresponding to each target patient, correct the target comprehensive score value corresponding to each target patient to obtain a corrected score value.

[0079] As an example, this step may include the following steps: First step: Designate any one of the target patients as a marked patient, and determine the mean value of all local score values within each local score value segment corresponding to the marked patient as the overall score index, obtaining the overall score index sequence corresponding to the marked patient.

[0080] Second step: Conduct abnormal analysis processing on each blood pressure data segment to obtain the target blood pressure abnormality factor corresponding to the local time period corresponding to each blood pressure data segment, which may include the following sub-steps: First sub-step: Detect blood pressure abnormalities for each blood pressure data segment to obtain abnormal blood pressure data, and form sub-segments of abnormal blood pressure data from the continuous abnormal blood pressure data within each blood pressure data segment.

[0081] It should be noted that high blood pressure can affect blood vessel tolerance, which may lead to the onset of chronic diseases. For example, high blood pressure often leads to the onset of coronary heart disease. Therefore, usually, abnormal blood pressure data can be screened based on a threshold, and blood pressure data exceeding the threshold is considered abnormal blood pressure data.

[0082] Second sub-step: Cluster all sub-segments of abnormal blood pressure data within each blood pressure data segment to obtain clusters of abnormal blood pressure, and select the cluster of abnormal blood pressure with the largest corresponding total duration from all clusters of abnormal blood pressure corresponding to each blood pressure data segment as the target cluster of abnormal blood pressure corresponding to each blood pressure data segment.

[0083] Among them, the total duration corresponding to the cluster of abnormal blood pressure may be equal to: the sum of the durations corresponding to all sub-segments of abnormal blood pressure within the cluster of abnormal blood pressure.

[0084] For example, the duration between the start times corresponding to every two sub-segments of abnormal blood pressure within the blood pressure data segment can be used as the clustering distance, and all sub-segments of abnormal blood pressure within the blood pressure data segment can be clustered through the K-means clustering algorithm, and the clustering clusters obtained at this time are used as clusters of abnormal blood pressure. Among them, the value of K in the K-means clustering algorithm can be obtained through the elbow method.

[0085] Third sub-step: Determine the target blood pressure abnormality factor corresponding to the local time period corresponding to each blood pressure data segment according to the total number of all abnormal blood pressure data in each blood pressure data segment and the number of abnormal blood pressure data in its corresponding target cluster of abnormal blood pressure.

[0086] For example, the formula for determining the target blood pressure abnormality factor corresponding to the local time period corresponding to the blood pressure data segment can be: ; where Q is the target blood pressure abnormality factor corresponding to the local time period corresponding to the blood pressure data segment. is the total number of abnormal blood pressure data in the blood pressure data segment. q is the number of blood pressure data in the blood pressure data segment. is the number of abnormal blood pressure data in the target abnormal blood pressure cluster corresponding to the blood pressure data segment.

[0087] It should be noted that when is larger, it often indicates that there are relatively more abnormal blood pressure data in the blood pressure data segment, and it often indicates that the blood pressure data segment is relatively more abnormal. When is larger, it often indicates that the abnormal blood pressure data in the blood pressure data segment is relatively more concentrated, and it often indicates that the degree of persistence of blood pressure abnormality in the blood pressure data segment is relatively larger. Therefore, when Q is larger, it often indicates that the blood pressure data segment is relatively more abnormal.

[0088] Thirdly, similarly, perform abnormal analysis and processing on each electrocardiogram data segment to obtain the target electrocardiogram abnormality factor corresponding to the local time period corresponding to each electrocardiogram data segment.

[0089] It should be noted that the method for obtaining the target electrocardiogram abnormality factor corresponding to the local time period can be the same as the method for obtaining the target blood pressure abnormality factor corresponding to the local time period. Specifically, the electrocardiogram data segment can be regarded as the blood pressure data segment, and at this time, the obtained target blood pressure abnormality factor is the target electrocardiogram abnormality factor.

[0090] Fourthly, determine the product of the target blood pressure abnormality factor and the target electrocardiogram abnormality factor corresponding to each local time period as the target overall abnormality factor corresponding to each local time period.

[0091] For example, the formula for determining the target overall abnormality factor corresponding to the local time period can be: ; where is the target overall abnormality factor corresponding to the a-th local time period of the marked patient. a is the serial number of the local time period of the marked patient. is the target blood pressure abnormality factor corresponding to the a-th local time period of the marked patient. is the target electrocardiogram abnormality factor corresponding to the a-th local time period of the marked patient.

[0092] It should be noted that when is larger, it often indicates that the blood pressure data in the a-th local time period of the marked patient is relatively more abnormal. When is larger, it often indicates that the electrocardiogram data in the a-th local time period of the marked patient is relatively more abnormal. Therefore, when The larger it is, it often indicates that both the blood pressure data and the electrocardiogram data of the marked patient in the corresponding a-th local time period are relatively more abnormal, and it often indicates that the physical monitoring data of the marked patient in the corresponding a-th local time period is relatively more abnormal.

[0093] Step 5: Combine the target overall abnormality factors corresponding to all local time periods of the above-mentioned marked patient to form the target overall abnormality factor sequence corresponding to the above-mentioned marked patient.

[0094] Step 6: Normalize the Pearson correlation coefficient between the target overall abnormality factor sequence and the overall scoring index sequence corresponding to the above-mentioned marked patient to obtain the target similarity corresponding to the above-mentioned marked patient.

[0095] It should be noted that the target overall abnormality factor sequence corresponding to the marked patient can characterize the physical abnormality of the marked patient during his current treatment time period. The overall scoring index sequence corresponding to the marked patient can characterize the disease risk assessment of the marked patient during his current treatment time period. Therefore, when the target similarity corresponding to the marked patient is larger, it often indicates that the target overall abnormality factor sequence and the overall scoring index sequence corresponding to the marked patient are more similar, and it often indicates that during the current treatment time period of the marked patient, the changes in the scoring results of different online diagnoses are more consistent with the changes in the patient's monitoring data, and it often indicates that the scoring tables used in different online diagnoses are more reliable, and it often indicates that the previously calculated target comprehensive score value can better characterize the urgency of online medical resource allocation.

[0096] Step 7: Normalize the product of the target similarity and the target comprehensive score value corresponding to the above-mentioned marked patient to obtain the corrected score value corresponding to the above-mentioned marked patient.

[0097] For example, the formula for determining the corrected score value corresponding to the marked patient can be: ; where is the corrected score value corresponding to the marked patient. is the normalization function. G is the target similarity corresponding to the marked patient. is the target comprehensive score value corresponding to the marked patient.

[0098] It should be noted that when is larger, it often indicates that it is more necessary to allocate online medical resources to the marked patient more quickly. G can be used as Weight. When G is larger, it often indicates that the target overall abnormal factor sequence corresponding to the marked patient and the overall scoring index sequence are more similar, often indicating that during the current treatment period of the marked patient, the changes in the scoring results of different online diagnoses are more consistent with the changes in the patient's monitoring data, often indicating that the scoring tables used in different online diagnoses are more reliable, and often indicating that the previously calculated target comprehensive score value can better represent the urgency of online medical resource allocation. Therefore, when is larger, it often indicates that it is more necessary to allocate online medical resources to the marked patient more quickly.

[0099] Step S6: Determine the resource allocation priority corresponding to each target patient according to the corrected score value corresponding to each target patient and the distance between each target patient and the hospital, and perform resource allocation based on the resource allocation priority.

[0100] As an example, this step may include the following steps: First step: Determine any one target patient as the marked patient, classify each hospital to obtain the target level corresponding to each hospital.

[0101] Among them, the target level corresponding to the hospital can represent the richness of medical resources related to chronic diseases in the hospital. The larger its value, the more abundant the medical resources related to chronic diseases, and the target level can be represented by a numerical value.

[0102] For example, hospitals can be classified into tertiary hospitals, secondary hospitals, and primary hospitals. The target level corresponding to tertiary hospitals can be greater than the target level corresponding to secondary hospitals, and the target level corresponding to secondary hospitals can be greater than the target level corresponding to primary hospitals. For example, the target level corresponding to tertiary hospitals can be set to 3, the target level corresponding to secondary hospitals can be set to 2, and the target level corresponding to primary hospitals can be set to 1. Among them, tertiary hospitals, also known as Class-III Grade-A hospitals, are hospitals with the strongest medical service capabilities, usually having strong scientific research, teaching, and treatment capabilities, and being able to provide highly specialized and complex medical services. Secondary hospitals have relatively strong medical service capabilities, but are smaller in scale and handle cases with lower complexity compared to tertiary hospitals. Primary hospitals are basic hospitals mainly responsible for the diagnosis and treatment of common and frequently-occurring diseases at the grass-roots level, and their technical strength is relatively weak.

[0103] Second step: Select the hospital closest to the address of the above-mentioned marked patient from all hospitals corresponding to each target level as the reference hospital for the above-mentioned marked patient at each target level.

[0104] Among them, when any one target level is determined as the marked level, the hospitals corresponding to the marked level can be the hospitals with the corresponding target level being the marked level.

[0105] In the third step, the distance between the reference hospital of the above-marked patient at each target level and the address of the above-marked patient is determined as the marked distance of the above-marked patient at each target level.

[0106] In the fourth step, according to the marked distances of the above-marked patient at all target levels, the richness of offline medical resources corresponding to the above-marked patient is determined.

[0107] Among them, the marked distance can have a negative correlation with the richness of offline medical resources.

[0108] For example, the formula for determining the richness of offline medical resources corresponding to the marked patient can be: ; where D is the richness of offline medical resources corresponding to the marked patient. F is the total number of types of target levels. h is the type serial number of the target level. is the h-th type of target level. is the marked distance of the marked patient at the h-th type of target level.

[0109] It should be noted that when is larger, it often indicates that the medical resources of the corresponding hospital are more abundant. When is smaller, it often indicates that the distance between the marked patient and the corresponding hospital is smaller. Therefore, when D is larger, it often indicates that the medical resources near the residence of the marked patient are relatively more abundant.

[0110] In the fifth step, according to the richness of offline medical resources corresponding to the above-marked patient and the corrected score value, the resource allocation priority corresponding to the above-marked patient is determined.

[0111] Among them, the richness of offline medical resources can have a negative correlation with the resource allocation priority. The corrected score value can have a positive correlation with the resource allocation priority.

[0112] For example, the formula for determining the resource allocation priority corresponding to the marked patient can be: ; where is the resource allocation priority corresponding to the marked patient. is the normalization function. is the corrected score value corresponding to the marked patient. D is the richness of offline medical resources corresponding to the marked patient.

[0113] It should be noted that when is larger, it often indicates that it is more necessary to allocate online medical resources to the marked patient more quickly. can be used as weight. When D is larger, it often indicates that the medical resources near the marked patient's residence are relatively more abundant, and it often indicates that the feasibility of the marked patient choosing to go to an offline hospital for examination is relatively strong. Therefore, when is larger, it often indicates that it is more necessary to allocate online medical resources to the marked patient more quickly.

[0114] Sixth step, allocate resources based on the resource allocation priority.

[0115] For example, based on the resource allocation priorities corresponding to all target patients, all target patients can be sorted in descending order, and online diagnosis resources can be allocated to the target patients in turn according to the sorting of the target patients.

[0116] It should be noted that when the resource allocation priority is larger, it is more necessary to arrange online diagnosis for the target patient in a timely manner.

[0117] Refer to Figure 2 , based on the same inventive concept as the above method embodiment, the present invention provides a multi-party collaborative computing and resource allocation system for online diagnosis of chronic diseases. The system includes a memory, a processor, and a computer program stored on the memory and executable on the processor. When the above computer program is executed by the processor, it implements the steps of a multi-party collaborative computing and resource allocation method for online diagnosis of chronic diseases, which may specifically include: A data acquisition module 201, configured to acquire a set of score sheets corresponding to each online diagnosis of each target patient during its current treatment period, as well as the blood pressure data and electrocardiogram data during its current treatment period, to obtain a sequence of score sheet sets, a blood pressure data sequence, and an electrocardiogram data sequence corresponding to each target patient; A construction and processing module 202, configured to construct a local score value sequence based on the sequence of score sheet sets corresponding to each target patient, and perform score distribution analysis and processing based on the local score value sequence corresponding to each target patient to obtain a target comprehensive score value; A score value segmentation module 203, configured to segment the local score value sequence corresponding to each target patient to obtain local score value segments; A data sequence segmentation module 204, configured to segment the blood pressure data sequence and the electrocardiogram data sequence corresponding to each target patient according to all local score value segments corresponding to each target patient to obtain blood pressure data segments and electrocardiogram data segments; A score correction module 205, configured to correct the target comprehensive score value corresponding to each target patient according to all local score value segments, all blood pressure data segments, and all electrocardiogram data segments corresponding to each target patient to obtain a corrected score value; A determination and resource allocation module 206, configured to determine the resource allocation priority corresponding to each target patient according to the corrected score value corresponding to each target patient and the distance between each target patient and the hospital, and perform resource allocation based on the resource allocation priority.

[0118] Figure 3 It is a schematic structural diagram of a computer device provided by an embodiment of the present invention. Exemplarily, as Figure 3 shown, the computer device 300 includes: a memory 301, a processor 302, and a computer program 303 stored in the memory 301 and running on the processor 302. When the processor 302 executes the computer program 303, the computer device can execute any one of the above-described multi-party collaborative computing and resource allocation methods for online diagnosis of chronic diseases.

[0119] Based on the same inventive concept as the above method embodiment, the present invention provides a server, including a memory and a processor. The memory is used to store executable program code, and the processor is used to call and run the executable program code from the memory, so that the device executes any one of the above-described multi-party collaborative computing and resource allocation methods for online diagnosis of chronic diseases.

[0120] Based on the same inventive concept as the above method embodiment, the present invention provides a computer program product, which includes: computer program code. When the computer program code runs on a computer, the computer executes any one of the above-described multi-party collaborative computing and resource allocation methods for online diagnosis of chronic diseases.

[0121] Based on the same inventive concept as the above method embodiment, the present invention provides a computer-readable storage medium, which stores computer program code. When the computer program code runs on a computer, the computer executes any one of the above-described multi-party collaborative computing and resource allocation methods for online diagnosis of chronic diseases.

[0122] In summary, when the present invention performs online medical resource allocation, it comprehensively considers a set of score tables representing abnormal monitoring data in the diagnosis time corresponding to different online diagnoses, analyzes the sequence of score table sets, the sequence of blood pressure data, and the sequence of electrocardiogram data, thereby adaptively quantifying the resource allocation priority corresponding to each target patient, and finally realizing resource allocation based on the resource allocation priority, and to a certain extent improving the rationality of online medical resource allocation.

[0123] 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 foregoing embodiments, those of ordinary skill in the art should understand that: they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the various embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A multi - party collaborative computing and resource allocation method for online diagnosis of chronic diseases, characterized in that Including the following steps: Obtain the set of score sheets corresponding to each online diagnosis during the current treatment period of each target patient, as well as the blood pressure data and electrocardiogram data during the current treatment period of each target patient, to obtain a sequence of score sheet sets, a sequence of blood pressure data, and a sequence of electrocardiogram data corresponding to each target patient; Based on the sequence of score sheet sets corresponding to each target patient, construct a sequence of local score values, and perform score distribution analysis processing based on the sequence of local score values corresponding to each target patient to obtain a target comprehensive score value; Segment the sequence of local score values corresponding to each target patient to obtain segments of local score values; According to all segments of local score values corresponding to each target patient, segment the sequence of blood pressure data and the sequence of electrocardiogram data corresponding to each target patient to obtain segments of blood pressure data and segments of electrocardiogram data; According to all segments of local score values, all segments of blood pressure data, and all segments of electrocardiogram data corresponding to each target patient, correct the target comprehensive score value corresponding to each target patient to obtain a corrected score value; According to the corrected score value corresponding to each target patient and the distance between each target patient and the hospital, determine the resource allocation priority corresponding to each target patient, and perform resource allocation based on the resource allocation priority.

2. The multilateral collaborative computing and resource allocation method for chronic disease online diagnosis according to claim 1, wherein The constructing a sequence of local score values based on the sequence of score sheet sets corresponding to each target patient includes: Determine any one target patient as a marked patient, and form a set of score values by taking the score values corresponding to all score sheets in each score sheet set in the sequence of score sheet sets corresponding to the marked patient, to obtain a sequence of sets of score values corresponding to the marked patient; Determine the mean value of all score values in each set of score values in the sequence of sets of score values corresponding to the marked patient as the local score value, to obtain a sequence of local score values corresponding to the marked patient.

3. A multi - party collaborative computing and resource allocation method for chronic disease online diagnosis according to claim 1, characterized in that, The performing score distribution analysis processing based on the sequence of local score values corresponding to each target patient to obtain a target comprehensive score value includes: Determine any one target patient as a marked patient, and perform linear fitting on the sequence of local score values corresponding to the marked patient to obtain a fitting line corresponding to the marked patient; According to the slope of the fitting line corresponding to the marked patient and the last local score value in the sequence of local score values corresponding to the marked patient, determine the initial comprehensive score value corresponding to the marked patient; According to the number of score sheets in each score sheet set in the sequence of score sheet sets corresponding to the marked patient, determine the score consistency corresponding to the marked patient; According to the initial comprehensive score value and the score consistency corresponding to the marked patient, determine the target comprehensive score value corresponding to the marked patient, where both the initial comprehensive score value and the score consistency are positively correlated with the target comprehensive score value.

4. A multi - party collaborative computing and resource allocation method for chronic disease online diagnosis according to claim 3, characterized in that The determining the initial comprehensive score value corresponding to the marked patient according to the slope of the fitting line corresponding to the marked patient and the last local score value in the sequence of local score values corresponding to the marked patient includes: Determine the normalized value of the slope of the fitting line corresponding to the marked patient as the target slope corresponding to the marked patient; Determine the last local score value in the local score value sequence corresponding to the marked patient as the recent score value corresponding to the marked patient; Normalize the product of the target slope and the recent score value corresponding to the marked patient to obtain the initial comprehensive score value corresponding to the marked patient.

5. The multilateral collaborative computing and resource allocation method for chronic disease online diagnosis according to claim 3, characterized in that The determining of the scoring consistency corresponding to the marked patient according to the number of scoring tables in each scoring table set in the scoring table set sequence corresponding to the marked patient includes: Determine the number of scoring tables in each scoring table set in the scoring table set sequence corresponding to the marked patient as the target representative number, and obtain the target representative number sequence corresponding to the marked patient; Round the mean of all modes in the target representative number sequence corresponding to the marked patient to obtain the reference number corresponding to the marked patient; Determine any one scoring table in the scoring table set sequence corresponding to the marked patient as the marked scoring table, and screen out the scoring tables with the same type as the marked scoring table from the scoring table set sequence corresponding to the marked patient to form the scoring table group corresponding to the marked scoring table, so as to obtain the scoring table group corresponding to each scoring table; Determine the number of scoring tables in the scoring table group corresponding to each scoring table as the standard number corresponding to each scoring table; Screen out the reference number of the largest standard numbers from the standard numbers corresponding to all scoring tables in the scoring table set sequence corresponding to the marked patient to form the standard number set corresponding to the marked patient; If the standard number corresponding to the scoring table belongs to the standard number set, determine the scoring table as a common scoring table, and determine the scoring table set with common scoring tables as the common scoring table set; If the standard number corresponding to the scoring table does not belong to the standard number set, determine the scoring table as a special scoring table, and determine the scoring table set in which all scoring tables are special scoring tables as the special scoring table set; Determine the scoring consistency corresponding to the marked patient according to the number of common scoring table sets in the scoring table set sequence corresponding to the marked patient, the differences between different common scoring table sets, and the differences between the common scoring table set and the special scoring table set.

6. The multi - party collaborative computing and resource allocation method for chronic disease online diagnosis according to claim 1, characterized in that The segmenting of the local score value sequence corresponding to each target patient to obtain local score value segments includes: Segment the local score value sequence corresponding to each target patient through the APCA algorithm to obtain multiple local score value segments corresponding to each target patient.

7. A multi - party collaborative computing and resource allocation method for chronic disease online diagnosis according to claim 1, characterized in that, The segmenting of the blood pressure data sequence and the electrocardiogram data sequence corresponding to each target patient according to all local score value segments corresponding to each target patient to obtain blood pressure data segments and electrocardiogram data segments includes: Determine any one target patient as the marked patient, and determine the time period corresponding to each local score value segment corresponding to the marked patient as the local time period; Constitute the blood pressure data corresponding to the acquisition moments belonging to the same local time period in the blood pressure data sequence corresponding to the marked patient as the blood pressure data segment corresponding to this local time period, and obtain multiple blood pressure data segments corresponding to the marked patient; In the ECG data sequence corresponding to the marked patient, the ECG data whose corresponding acquisition times belong to the same local time period are grouped to form ECG data segments, and multiple ECG data segments corresponding to the marked patient are obtained.

8. A multi - party collaborative computing and resource allocation method for chronic disease online diagnosis according to claim 7, characterized in that Based on all local score value segments, all blood pressure data segments, and all ECG data segments corresponding to each target patient, the target comprehensive score value corresponding to each target patient is corrected to obtain a corrected score value, including: Any one of the target patients is determined as the marked patient, and the mean value of all local score values within each local score value segment corresponding to the marked patient is determined as the overall score index, obtaining the overall score index sequence corresponding to the marked patient; Perform abnormal analysis on each blood pressure data segment to obtain the target blood pressure abnormality factor corresponding to the local time period corresponding to each blood pressure data segment; Similarly, perform abnormal analysis on each ECG data segment to obtain the target ECG abnormality factor corresponding to the local time period corresponding to each ECG data segment; The product of the target blood pressure abnormality factor and the target ECG abnormality factor corresponding to each local time period is determined as the target overall abnormality factor corresponding to each local time period; The target overall abnormality factors corresponding to all local time periods corresponding to the marked patient are grouped to form the target overall abnormality factor sequence corresponding to the marked patient; Normalize the Pearson correlation coefficient between the target overall abnormality factor sequence and the overall score index sequence corresponding to the marked patient to obtain the target similarity corresponding to the marked patient; Normalize the product of the target similarity and the target comprehensive score value corresponding to the marked patient to obtain the corrected score value corresponding to the marked patient.

9. A multi - party collaborative computing and resource allocation method for chronic disease online diagnosis according to claim 8, characterized in that, The performing abnormal analysis on each blood pressure data segment to obtain the target blood pressure abnormality factor corresponding to the local time period corresponding to each blood pressure data segment includes: Perform blood pressure abnormality detection on each blood pressure data segment to obtain abnormal blood pressure data, and group the continuous abnormal blood pressure data within each blood pressure data segment to form abnormal blood pressure data sub-segments; Cluster all abnormal blood pressure data sub-segments within each blood pressure data segment to obtain abnormal blood pressure clusters, and select the abnormal blood pressure cluster with the largest total duration corresponding to each blood pressure data segment from all abnormal blood pressure clusters corresponding to each blood pressure data segment as the target abnormal blood pressure cluster corresponding to each blood pressure data segment; Based on the total number of all abnormal blood pressure data in each blood pressure data segment and the number of abnormal blood pressure data in its corresponding target abnormal blood pressure cluster, determine the target blood pressure abnormality factor corresponding to the local time period corresponding to each blood pressure data segment.

10. A multi - party collaborative computing and resource allocation method for chronic disease online diagnosis according to claim 1, characterized in that, Based on the corrected score value corresponding to each target patient and the distance between each target patient and the hospital, determine the resource allocation priority corresponding to each target patient, including: Any one of the target patients is determined as the marked patient, and each hospital is classified to obtain the target level corresponding to each hospital; Select the hospital closest to the address of the marked patient from all hospitals corresponding to each target level as the reference hospital for the marked patient under each target level; Determine the distance between the reference hospital of the marked patient at each target level and the address of the marked patient as the marked distance of the marked patient at each target level; Determine the richness of offline medical resources corresponding to the marked patient according to the marked distances of the marked patient at all target levels, where the marked distance is negatively correlated with the richness of offline medical resources; Determine the resource allocation priority corresponding to the marked patient according to the richness of offline medical resources corresponding to the marked patient and the corrected score value, where the richness of offline medical resources is negatively correlated with the resource allocation priority, and the corrected score value is positively correlated with the resource allocation priority.

Images