Vaccination outpatient service mobile supervision management method and system

Abstract

The invention relates to the technical field of outpatient service supervision management, and particularly discloses a vaccination outpatient service mobile supervision management method and system, which ensure that the prevention supervision rectification rate of each level of vaccination outpatient service department is fed back in time, supervision information is generated, and an intelligent and reasonable supervision process is completed. Comprising the following steps: collecting registration supervision expert archive information and reservation cancelling supervision expert archive information of various levels of historical vaccination outpatient service management mechanisms in a preset time period; performing multi-dimensional data analysis according to the registration supervision expert archive information and the reservation cancelling supervision expert archive information to obtain a supervision rectification rate parameter, and sending supervision information to each level of vaccination outpatient service management mechanism; counting a supervision rectification rate parameter set of each level of historical vaccination outpatient service management institution, and inputting the supervision rectification rate parameter set into the GRU recurrent neural network model for training to obtain an vaccination supervision model; and acquiring a supervision rectification rate parameter set of each inoculation outpatient service in real time, inputting the supervision rectification rate parameter set into the inoculation supervision model for analysis, and generating supervision reminding according to the reminding state coefficient.

Description

Technical Field

[0001] The present invention relates to the technical field of outpatient supervision management, and in particular to a method and system for mobile supervision management of an immunization outpatient clinic. Background Art

[0002] The mobile supervision and management of vaccination clinics involves many department management processes, and the management process also requires the ability to integrate online and offline data information; therefore, various places have actively promoted the informatization construction of vaccination clinics, and through the construction of an informatization management platform for the entire vaccination process, it has realized functions such as online appointments, vaccination information inquiries, and self-service reminders, in order to improve the work efficiency and service quality of the outpatient department, and also provide a data basis for mobile supervision and management.

[0003] The existing mobile supervision and management system for immunization clinics needs to use mobile terminal devices to confirm user vaccination information and install relevant supervision management applications to guide vaccination units to carry out the supervision and management process; however, since the supervision work needs to be implemented at the grassroots level, the paper-based statistical method is prone to cause non-standard assessment items or missing relevant assessment items during the supervision and assessment process, and there are problems such as large storage volume of archival materials and extremely inconvenient access to historical archival materials, which in turn affects the efficiency of supervision and rectification and reduces the quality of outpatient services.

[0004] Therefore, it is necessary to use big data and cloud computing technology to analyze and mine the relevant data of outpatient supervision management, further strengthen the supervision and rectification feedback mechanism, improve monitoring efficiency, and then formulate more scientific and intelligent supervision plans and strategies. Summary of the invention

[0005] The purpose of the present invention is to provide a method and system for mobile supervision and management of vaccination clinics to solve the following technical problems:

[0006] How to ensure timely feedback on the prevention supervision rectification rate of immunization clinics at all levels and generate supervision information to complete an intelligent and reasonable supervision process.

[0007] The purpose of the present invention can be achieved by the following technical solutions:

[0008] A mobile supervision and management method for vaccination clinics, the method comprising:

[0009] S1. Collect the registered supervisory expert file information of the vaccination clinic management agencies at all levels during the preset time period, and count the number of supervisory experts and expert supervision dates based on the registered supervisory expert file information;

[0010] S2. Collect the appointment cancellation supervision expert file information of the historical vaccination outpatient management institutions at all levels during the preset time period, and count the number of unappointed supervision experts and the date of the last supervision of the experts according to the appointment cancellation supervision expert file information;

[0011] S3. Conduct multi-dimensional data analysis based on the registered supervision expert file information and the appointment cancellation supervision expert file information to obtain the supervision rectification rate parameter, judge the level of the supervision rectification rate, and send supervision information to the vaccination outpatient management institutions at all levels;

[0012] S4. Count the set of supervision rectification rate parameters of the historical vaccination outpatient management institutions at all levels and input them into the GRU recurrent neural network model for training to obtain the vaccination supervision model;

[0013] S5. Real-time obtain the set of supervision rectification rate parameters of each vaccination outpatient, input them into the vaccination supervision model for analysis, obtain the reminder status coefficient of each vaccination outpatient, and generate a supervision reminder according to the size of the reminder status coefficient.

[0014] Preferably, the process of multi-dimensional data analysis in S3 is as follows:

[0015] Screen the number of supervision experts and the date of expert supervision in the registered supervision expert file information and record them as the first input information;

[0016] Screen the number of unappointed supervision experts and the date of the last supervision of the experts in the appointment cancellation supervision expert file information and record them as the second input information;

[0017] Perform information merging processing on the first input information and the second input information to respectively confirm the number of people who were supervised last time and are supervised this time, the number of people who were supervised last time but not this time, and the number of problems in the last supervision and mark them;

[0018] Obtain the total number of people in the registered supervision expert file information and the total number of supervision times, and use the marked number of people who were supervised last time and are supervised this time as the first positive example sample and the number of people who were supervised last time but not this time as the first negative example sample;

[0019] Input the first positive example sample and the first negative example sample into the BP neural network model for training, construct an analysis model, and establish a feedback learning mechanism for the analysis model.

[0020] Preferably, the judgment process of the supervision rectification rate is as follows:

[0021] Calculate the change of the target cycle within the preset time period based on the feedback learning mechanism of the analysis model to obtain the supervision rectification rate parameter:

[0022]

[0023] Among them, is the supervision rectification rate parameter; is the target cycle within the preset time period; is the number of rectification personnel who were supervised last time and have been supervised this time within the duration; is the number of rectification personnel who were supervised last time but not supervised this time within the duration; is the number of rectification personnel who were not supervised either last time or this time within the duration, regardless of whether they were supervised before; is the total number of registered expert file information; is the weight coefficient of the regional registration number within the duration, and 0 < < 1; is the number of supervision times.

[0024] Preferably, compare the supervision and rectification rate parameter with the standard threshold interval of the preset supervision and rectification rate parameter and is the minimum value of the standard supervision and rectification rate parameter, is the maximum value of the standard supervision and rectification rate parameter:

[0025] If > , it is determined that the supervision and rectification rate is high;

[0026] If ≤ ≤ , it is determined that the supervision and rectification rate is medium;

[0027] If < , it is determined that the supervision and rectification rate is low.

[0028] Preferably, the method of statistically collecting the supervision and rectification rate parameter sets of historical vaccination outpatient management institutions at all levels in S4 and inputting them into the GRU recurrent neural network model for training to obtain the vaccination supervision model is as follows:

[0029] Obtain positive and negative samples: Use the set of supervision and rectification rate parameters with a high supervision and rectification rate as the second positive example sample and the sets of supervision and rectification rate parameters with a medium and low supervision and rectification rate as the second negative example samples;

[0030] Input the second positive example sample and the second negative example sample into the GRU recurrent neural network model for training and learning to obtain the vaccination supervision model.

[0031] Preferably, the reminder status coefficient:

[0032] is calculated by the formula to obtain the reminder status coefficient of the ;

[0033] Among them, is the target cycle within the preset time period; is the previous cycle period of two adjacent cycles of the vaccination clinic within the preset time period; is the latter cycle period of two adjacent cycles of the vaccination clinic within the preset time period; is the first preset weight coefficient, is the second preset weight coefficient; and , are both greater than 0; is the real-time supervision and rectification rate parameter of the -level vaccination clinic.

[0034] Preferably, the method for generating supervision reminders according to the reminder status coefficient is as follows:

[0035] Compare the reminder status coefficient with the preset reminder status coefficient threshold to compare their magnitudes:

[0036] If ≤ , it is determined that the completion degree of the supervision of the vaccination work in the current-level vaccination clinic meets the requirements;

[0037] If > , it is determined that the completion degree of the supervision of the vaccination work in the current-level vaccination clinic does not meet the requirements, and a supervision reminder is generated.

[0038] It also includes:

[0039] Register the evaluation information after supervision and rectification and the reasons for canceling supervision, extract the information related to the effect of supervision and rectification, and match the corresponding-level vaccination clinics to send supervision and rectification notices.

[0040] A mobile supervision management system for vaccination clinics is used to implement a mobile supervision management method for vaccination clinics. The system includes:

[0041] A collection unit, which is used to collect the registered supervision expert file information of historical vaccination clinic management agencies at all levels within the preset time period, and count the number of supervision experts and the expert supervision dates according to the registered supervision expert file information; and is used to collect the reserved cancellation supervision expert file information of historical vaccination clinic management agencies at all levels within the preset time period, and count the number of unreserved supervision experts and the expert's last supervision date;

[0042] An analysis unit, which is used to perform multi-dimensional data analysis on the registered supervision expert file information and the reserved cancellation supervision expert file information to obtain the supervision and rectification rate parameter, judge the level of the supervision and rectification rate, and send supervision information to vaccination clinic management agencies at all levels;

[0043] A model construction unit, which is used to count the supervision and rectification rate parameter sets of historical vaccination outpatient management institutions at all levels and input them into a GRU recurrent neural network model for training to obtain a vaccination supervision model;

[0044] A supervision unit, which is used to obtain the supervision and rectification rate parameter sets of vaccination outpatient clinics at all levels in real time, input them into the vaccination supervision model for analysis, obtain the reminder status coefficients of vaccination outpatient clinics at all levels, and generate supervision reminders according to the magnitudes of the reminder status coefficients.

[0045] Advantages of the present invention:

[0046] (1) The present invention obtains supervision and rectification rate parameters through multi-dimensional data analysis, judges the high or low of the supervision and rectification rate, sends supervision information to vaccination outpatient management institutions at all levels, and confirms the information of the outpatient management institutions that send supervision information; through predictive analysis of the high or low of the supervision and rectification rate, timely send supervision information to vaccination outpatient management institutions at all levels to ensure that outpatient departments at all levels generate supervision and rectification rate parameters based on historical vaccination data as a reference.

[0047] (2) The present invention counts the supervision and rectification rate parameter sets of historical vaccination outpatient management institutions at all levels according to the sent supervision information and inputs them into a GRU recurrent neural network model for training to obtain a vaccination supervision model; analyzes the dynamic changes of the supervision and rectification rate parameters in adjacent periods to obtain reminder status coefficients, and determines the real-time stage of the vaccination work during this period according to the reminder status coefficients; furthermore, makes corresponding urging feedback on the vaccination work in a timely manner according to the output of the magnitudes of the reminder status coefficients.

[0048] Of course, it is not necessary for any product implementing the present invention to achieve all the above-described advantages simultaneously. Description of the Drawings

[0049] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for describing the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention, and those of ordinary skill in the art can obtain other drawings based on these drawings without creative efforts.

[0050] Figure 1 It is a flowchart of the steps of a mobile supervision and management method for vaccination outpatient clinics of the present invention;

[0051] Figure 2 It is a unit diagram of a mobile supervision and management system for vaccination outpatient clinics of the present invention. Detailed Embodiments

[0052] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0053] Please refer to Figure 1 as shown in the figure, the present invention is a mobile supervision and management method for vaccination clinics, and the method includes:

[0054] S1. Collect the registration supervision expert file information of the historical vaccination clinic management institutions at all levels in a preset time period, and count the number of supervision experts and the expert supervision dates according to the registration supervision expert file information;

[0055] S2. Collect the appointment cancellation supervision expert file information of the historical vaccination clinic management institutions at all levels in a preset time period, and count the number of non-appointed supervision experts and the expert's last supervision date according to the appointment cancellation supervision expert file information;

[0056] S3. Perform multi-dimensional data analysis based on the registration supervision expert file information and the appointment cancellation supervision expert file information to obtain the supervision rectification rate parameter, judge the level of the supervision rectification rate, and send supervision information to the vaccination clinic management institutions at all levels;

[0057] S4. Count the set of supervision rectification rate parameters of the historical vaccination clinic management institutions at all levels and input them into the GRU recurrent neural network model for training to obtain the vaccination supervision model;

[0058] S5. Real-time obtain the set of supervision rectification rate parameters of each vaccination clinic and input them into the vaccination supervision model for analysis, obtain the reminder status coefficient of each vaccination clinic, and generate a supervision reminder according to the size of the reminder status coefficient.

[0059] In the above technical solution, by collecting the login information of supervision experts on mobile terminal devices such as mobile phones and tablets, and obtaining the supervision data of the information supervision management department of vaccination clinics through the mobile information receiving end, it mainly realizes the accurate acquisition of the vaccination situation of historical vaccination departments at all levels during the supervision process by collecting the supervision registration information of historical vaccination clinic management institutions at all levels within a preset time period and screening the supervision file information of supervised registration vaccinations.

[0060] Specifically, first, collect the registration supervision expert file information of the historical vaccination outpatient management institutions at all levels during the preset time period, and count the number of supervision experts and the expert supervision dates according to the registration supervision expert file information; and collect the appointment cancellation supervision expert file information of the historical vaccination outpatient management institutions at all levels during the preset time period, and count the number of non-appointed supervision experts and the expert's last supervision date according to the appointment cancellation supervision expert file information. Moreover, since there are various situations where experts are registered but not supervised, further data screening and processing are required to ensure the realization of the multi-dimensional data analysis process based on the registration supervision expert file information and the appointment cancellation supervision expert file information. The multi-dimensional data analysis technology can screen the data related to the supervision vaccination status. In this embodiment, the supervision rectification rate parameter is calculated and obtained through multi-dimensional data analysis, the level of the supervision rectification rate is judged, and the supervision information is sent to the vaccination outpatient management institutions at all levels, and the information of the outpatient management institutions that send the supervision information is confirmed; through the predictive analysis of the level of the supervision rectification rate, the supervision information is sent to the vaccination outpatient management institutions at all levels in a timely manner to ensure that each outpatient department generates the supervision rectification rate parameter based on the historical vaccination data as a reference; and then, based on machine learning and data model selection, a further mining and analysis process is carried out on the supervision rectification rate parameter.

[0061] Furthermore, the mining and analysis process is specifically to count the set of supervision rectification rate parameters of the historical vaccination outpatient management institutions at all levels according to the sent supervision information and input it into the GRU recurrent neural network model for training to obtain the vaccination supervision model; optimize the supervision rectification information to realize the accurate input process of monitoring the supervision vaccination data; finally, obtain the set of supervision rectification rate parameters of each vaccination outpatient in real time and input it into the vaccination supervision model for analysis to realize the real-time data detection process, so as to obtain the reminder status coefficient of each vaccination outpatient, and generate supervision reminders in real time according to the size of the reminder status coefficient to ensure that each outpatient department receives accurate supervision message prompts and optimize the supervision process.

[0062] As an implementation manner of the present invention, the process of multi-dimensional data analysis in S3 is as follows:

[0063] The number of supervision experts and the expert supervision dates in the screened registration supervision expert file information are recorded as the first input information;

[0064] The number of non-appointed supervision experts and the expert's last supervision date in the screened appointment cancellation supervision expert file information are recorded as the second input information;

[0065] The first input information and the second input information are merged and processed to respectively confirm the number of people who were supervised last time and are supervised this time, the number of people who were supervised last time but not supervised this time, and the number of problems in the last supervision and mark them;

[0066] Obtain the total number of registered supervision expert file information and the total number of supervisions, and use the number of people who were supervised last time and have been supervised this time as the first positive example sample, and the number of people who were supervised last time but not supervised this time as the first negative example sample;

[0067] Input the first positive example sample and the first negative example sample into the BP neural network model for training, construct an analysis model, and establish a feedback learning mechanism for the analysis model.

[0068] In the above technical solution, the specific process of setting multi-dimensional data analysis is to screen the number of supervision experts and the expert supervision dates in the registered supervision expert file information as the first input information; and screen the number of non-reserved supervision experts and the expert's last supervision date in the reserved cancellation supervision expert file information as the second input information; determine the number of currently registered and supervised experts through the first input data, and confirm the number of currently registered but not participating in supervision this time through the second input information; according to the first input information and the second input information, perform data insertion and table merging and other statistical knowledge, and four situations can be confirmed: supervised last time and supervised this time, supervised last time but not supervised this time, not supervised both last time and this time, not supervised last time but supervised this time. Extract conditional information through the input data, screen and confirm the number of people who were supervised last time and have been supervised this time, the number of people who were supervised last time but not supervised this time, and the number of supervision problem times last time and mark them.

[0069] Then, obtain the total number of registered supervision expert file information and the total number of supervisions, extract the total number of supervision experts data and the total number of supervision times data from the supervision expert file information; then determine the sample information, use the number of people who were supervised last time and have been supervised this time as the first positive example sample and the number of people who were supervised last time but not supervised this time as the first negative example sample. After the obtained sample information is confirmed, the machine model such as the BP neural network model completes the training process. By inputting the first positive example sample and the first negative example sample into the BP neural network model for training, construct an analysis model, and establish a feedback learning mechanism for the analysis model, complete the construction of the analysis model and perform the preliminary training process according to the above historical data set.

[0070] As an implementation manner of the present invention, the judgment process of the supervision rectification rate is as follows:

[0071] Calculate the supervision rectification rate parameter based on the feedback learning mechanism of the analysis model for the target cycle change within a preset time period:

[0072]

[0073] Wherein, is the supervision rectification rate parameter; is the target cycle within the preset time period; is The number of rectified people who were last supervised within the time period and have been supervised this time; is The number of rectified people who were last supervised within the time period but have not been supervised this time; is The number of rectified people who have not been supervised either last time or this time within the time period, regardless of whether they were supervised before; is the total number of people registered in the expert file; is The weight coefficient of the regional registration number within the time period, and 0 < < 1; is the number of supervision times.

[0074] In the above technical solution, by defining a preset time period, several time intervals within this time period are evenly collected as target cycles, and the supervision and rectification rate parameter is obtained through calculation according to the change of the target cycle; the specific process of judging the size of the supervision and rectification rate parameter is obtained through calculation by the feedback learning mechanism, and the specific calculation process is through the formula Calculate to obtain the supervision and rectification rate parameter ; Since the data range related to the supervision and rectification rate includes: the number of people supervised and rectified and this vaccination content (obtained through hospital diagnosis and CDC statistics), and because the supervision and rectification process has periodicity, seasonality and regionality due to the vaccination disease situation, therefore, only by statistically confirming the fixed number of people supervised and rectified cannot accurately reflect the current epidemic prevention effect, and it is also necessary to add the judgment process of the number of people before and after vaccination supervision; specifically, the analysis process is to judge the weight of this number Under the state of, combine and statistically count the number of rectifications for single supervision 、multiple supervision and non-supervision and other situations of the number of rectifications, and in this formula represents the degree of rectification for the last supervision but not this time; The larger it is, the lower the supervision and rectification degree, and the higher the supervision and rectification efficiency; Indicates the rectification ratio for the last supervision and this supervision, and The smaller it is, the larger the relative rectification ratio The value of, indicating the higher the supervision and rectification rate. In addition, the weight coefficient is determined according to the ratio of the number of people supervised and rectified (including those supervised more than once) to the total number of registrations, and is a value that can be statistically determined. Therefore, judging the supervision and rectification rate is through statistically counting the change of the number of people supervised and rectified, that is The size of; actually, by judging that when The overall value is larger, and 0 < < 1, then the supervision and rectification rate parameter It will increase as the number of people supervised for rectification increases; is the number of supervision times, and > 0.

[0075] As an implementation manner of the present invention, the supervision and rectification rate parameter is compared with the standard threshold interval of the preset supervision and rectification rate parameter and is the minimum value of the standard supervision and rectification rate parameter, is the maximum value of the standard supervision and rectification rate parameter:

[0076] If > , it is determined that the supervision and rectification rate is high;

[0077] If ≤ ≤ , it is determined that the supervision and rectification rate is medium;

[0078] If < , it is determined that the supervision and rectification rate is low.

[0079] In the above technical solution, when the supervision and rectification rate parameter is greater than the maximum value of the preset interval, it indicates that the supervision and rectification rate is high. On the contrary, it indicates that the supervision and rectification rate is general or low. During the period of strengthening supervision and rectification, this result needs to be promoted for vaccination. The vaccination institutions should be supervised and the further expansion of the scope of this epidemic should be reduced, thereby reducing the impact on people's normal life.

[0080] As an implementation manner of the present invention, the method for statistically collecting the supervision and rectification rate parameter sets of historical vaccination outpatient management institutions at all levels in step S4 and inputting them into the GRU recurrent neural network model for training to obtain the vaccination supervision model is as follows:

[0081] Obtain positive and negative samples: Use the set of supervision and rectification rate parameters with a high supervision and rectification rate as the second positive example sample and the set of supervision and rectification rate parameters with a medium and low supervision and rectification rate as the second negative example sample;

[0082] Input the second positive example sample and the second negative example sample into the GRU recurrent neural network model for training and learning to obtain the vaccination supervision model.

[0083] In the above technical solution, the supervision information obtained is used to count the supervision rectification rate parameter sets of the historical vaccination clinic management agencies at all levels for machine training. The training method first requires the use of a machine model, and uses the deep learning framework of the GRU recurrent neural network model (such as PyTorch, TensorFlow, etc.) to define the structure of a new GRU model, including input layer, GRU layer, output layer and other structures. The trained model (vaccination supervision model) can predict the changing trend of the supervision rectification rate parameters, and the amount of data that GRU needs to calculate is relatively small, and it can quickly provide feedback on the real-time acquired data in a timely manner to optimize the supervision process; specifically, first obtain positive and negative sample information, and use the supervision rectification rate parameter set with a high supervision rectification rate as the second positive sample; and use the supervision rectification rate parameter set with a medium and low supervision rectification rate as the second negative sample; then input the second positive sample and the second negative sample into the GRU recurrent neural network model for training and learning to construct a vaccination supervision model.

[0084] As an implementation mode of the present invention, the reminder state coefficient is:

[0085] By formula Calculate the first Reminder status coefficient for vaccination clinics ;

[0086] in, is the target period within the preset time period; The vaccination clinic is in the previous cycle of two consecutive cycles within the preset time period; The vaccination clinic is in the latter cycle of two consecutive cycles within the preset time period; is the first preset weight coefficient, is a second preset weight coefficient; and , All are greater than 0; For the Real-time supervision and rectification rate parameters of level 1 vaccination clinics.

[0087] In the above technical solution, after the vaccination supervision model is constructed, the supervision rectification rate parameters obtained in real time are input into the model for testing and analysis. Specifically, the supervision rectification rate parameter set of vaccination clinics at all levels is obtained in real time and input into the vaccination supervision model for testing; after the test is completed, according to the test output results, the change state of the supervision rectification rate parameters of adjacent periods is required to calculate the dynamic data, and the coefficient calculation is used to determine whether the vaccination after the current training meets the requirements, and a supervision reminder is generated for those that do not meet the requirements, which is specifically through the formula Calculate the first Reminder status coefficient for vaccination clinics ; Reminder status coefficient It reflects the periodic completion effect and completion degree of the vaccination work in the vaccination clinics at the current level; therefore, by judging the dynamic changes in adjacent periods, the real-time stage of the vaccination work during this period can be explained; furthermore, according to the output of the reminder status coefficient, corresponding supervision feedback can be given to the vaccination work in a timely manner.

[0088] Among them, it should be explained that the first preset weight coefficient and the second preset weight coefficient are calculated by the CRITIC weight method for the real-time supervision and rectification rate parameter sets of the two periods respectively. The weight value is obtained by multiplying the contrast intensity (represented by the standard deviation of all supervision and rectification rate parameters in each period) and the conflict index (represented by the correlation coefficient of all supervision and rectification rate parameters in each period, usually considering the non-rectification rate after supervision as the correlation coefficient) and then normalizing the result.

[0089] As an implementation manner of the present invention, the method for generating a supervision reminder according to the reminder status coefficient is as follows:

[0090] Compare the reminder status coefficient with the preset reminder status coefficient threshold to determine their magnitudes:

[0091] If ≤ , it is determined that the completion degree of the supervision vaccination work in the vaccination clinic at the current level meets the requirements;

[0092] If > , it is determined that the completion degree of the supervision vaccination work in the vaccination clinic at the current level does not meet the requirements, and a supervision reminder is generated.

[0093] In the above technical solution, as shown in Table 1, the method for generating a supervision reminder includes generating a supervision selection form for the corresponding department, which is updated at regular intervals, generally updated according to periodic changes. According to the dynamic supervision form, the corresponding time period is selected to arrange supervisors to go to vaccination clinics at all levels in a timely manner to conduct supervision vaccination and vaccination publicity. In this application, first, data is input, and the rectification rate parameters of outpatient clinics at all levels are collected in real time. The rectification rate parameters are the rectification-related data obtained daily (cumulative number of problems found, cumulative number of rectifications implemented, problem rectification rate (%), number of outpatient clinics with incomplete rectification, outpatient clinic complete rectification rate (%)); then, the rectification rate data of each outpatient clinic is integrated according to a time window, and the formula can be simplified to: ; where is the number of data points within the cycle period, and the cycle period includes , , and ; is the time interval between data points; is the supervision and rectification rate parameter of the -level vaccination clinic at the th data point; if > , it is determined as "not meeting the requirements", and a supervision reminder is triggered; the standard threshold needs to be adjusted according to historical data (such as determining the normal fluctuation range through statistical methods).

[0094] Table 1 Supervision Reminder for Vaccination Cycle

[0095]

[0096] As an implementation manner of the present invention, it further includes:

[0097] Register the evaluation information after supervision and rectification and the reasons for canceling supervision, extract the information related to the effect of supervision and rectification, and match the corresponding-level vaccination clinics to send supervision and rectification notices.

[0098] In the above technical solution, the information on the reasons for canceling supervision and the evaluation information after supervision and rectification are deconstructed using a language model, the evaluation related to the effect of supervision and rectification is extracted, the number of evaluation information and the vaccinated clinics evaluated are counted, and supervision notices are sent in a timely manner.

[0099] Please refer to Figure 2 shown. The present invention also sets up a mobile supervision and management system for vaccination clinics to implement a mobile supervision and management method for vaccination clinics. The system includes:

[0100] A collection unit for collecting the registered supervision expert file information of historical vaccination clinic management institutions at all levels during a preset time period, counting the number of supervision experts and the expert supervision dates according to the registered supervision expert file information; and for collecting the reserved cancellation supervision expert file information of historical vaccination clinic management institutions at all levels during a preset time period, counting the number of unreserved supervision experts and the expert's last supervision date;

[0101] An analysis unit for performing multi-dimensional data analysis on the registered supervision expert file information and the reserved cancellation supervision expert file information to obtain the supervision and rectification rate parameter, judging the level of the supervision and rectification rate, and sending supervision information to vaccination clinic management institutions at all levels;

[0102] A model construction unit for counting the set of supervision and rectification rate parameters of historical vaccination clinic management institutions at all levels and inputting them into a GRU recurrent neural network model for training to obtain a vaccination supervision model;

[0103] A supervision unit for obtaining the set of supervision and rectification rate parameters of vaccination clinics at all levels in real time, inputting them into the vaccination supervision model for analysis, obtaining the reminder status coefficient of vaccination clinics at all levels, and generating supervision reminders according to the size of the reminder status coefficient.

[0104] Tables 2 and 3 reflect the analysis through the mobile supervision and management system for vaccination clinics, and statistically analyze the implementation and rectification of outpatient supervision in different regions over the years; by analyzing the cumulative number of discovered problems and the cumulative data of implemented rectification, information on the rectification rate is obtained to optimize and implement the rectification process.

[0105] Implementation and rectification situation in local areas in Table 2

[0106]

[0107] Table 3 Implementation and rectification situation in different rounds in multiple regions cumulatively

[0108]

[0109] Table 4 Rectification situation of supervision problems

[0110]

[0111] In Table 4 above, the number of problems in the previous supervision is counted according to the third-level headings; under the condition of selecting the time period / task name, it is the sum of the problems found in the previous regular supervision of all supervised units in this area ("sum of the number of 'latest anomalies' problems").

[0112] Number of problems with rectification completed: Only counted when selecting regular supervision; under the condition of selecting the time period / task name, it is the number of problems selected as [rectified] by all supervised units in this area in this regular supervision;

[0113] Rectification rate (%): Number of problems with rectification completed / Number of problems in the previous supervision * 100; Only counted when selecting regular supervision;

[0114] Number of supervision problems: The definition remains unchanged. Under the condition of selecting the time period / task name, it is the number of problems found by all supervised units in this area in this regular supervision (note: including the problems that were found last time and selected as [not rectified] this time);

[0115] Number of problems in the previous supervision, number of supervision problems: The problem lists can be clicked to view respectively.

[0116] Each embodiment in this specification is described in a progressive manner. For the same or similar parts between each embodiment, reference can be made to each other. Each embodiment focuses on the differences from other embodiments. In particular, for the embodiments of devices, equipment, and non-volatile computer storage media, since they are basically similar to the method embodiments, the description is relatively simple, and for the relevant parts, reference can be made to the partial description of the method embodiments.

[0117] The above describes specific embodiments of this specification. Other embodiments are within the scope of the appended documents. In some cases, the actions or steps recited in this application may be performed in a different order than in the embodiments and still achieve the desired results. Additionally, the processes depicted in the figures do not necessarily require the particular order or sequential order shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0118] The above content is only an example and illustration of the concept of the present invention. Those skilled in the art of this technology can make various modifications or supplements to the described specific embodiments or use similar methods for substitution, as long as they do not deviate from the concept of the invention or exceed the scope defined by this application, they should all fall within the protection scope of the present invention.

Claims

1. A mobile supervision and management method for vaccination clinics, characterized in that: The method comprises: S1. Collect the registered supervisory expert file information of the vaccination clinic management agencies at all levels during the preset time period, and count the number of supervisory experts and expert supervision dates based on the registered supervisory expert file information; S2. Collect the appointment cancellation supervision expert file information of the preset time period of the vaccination clinic management agencies at all levels, and count the number of supervision experts who have not made an appointment and the date of the last supervision by the experts based on the appointment cancellation supervision expert file information; S3. Perform multi-dimensional data analysis based on the registered supervisory expert file information and the appointment cancellation supervisory expert file information to obtain the supervisory rectification rate parameters, determine the supervisory rectification rate and send supervisory information to vaccination clinic management agencies at all levels; S4. Statistically calculate the supervision rectification rate parameter set of vaccination clinic management agencies at all levels based on the supervision information, and input it into the GRU recurrent neural network model for training to obtain the vaccination supervision model; S5. Obtain the supervision rectification rate parameter set of vaccination clinics at all levels in real time and input it into the vaccination supervision model for analysis, obtain the reminder status coefficient of vaccination clinics at all levels, and generate supervision reminders according to the size of the reminder status coefficient.

2. A mobile supervision and management method for vaccination clinics according to claim 1, characterized in that: The process of multidimensional data analysis in S3 is as follows: The number of supervisory experts and the expert supervision date of the registered supervisory expert file information are recorded as the first input information; The number of supervisory experts who have not made an appointment and the date of the last supervisory visit of the experts who have cancelled their appointments are recorded as the second input information; The first input information and the second input information are combined to confirm and mark the number of people who were supervised last time and have been supervised this time, the number of people who were supervised last time but not supervised this time, and the number of problems in the last supervision; Obtain the total number of registered supervisory expert file information and the total number of supervisions, and take the number of people who were supervised last time and supervised this time as the first positive sample and the number of people who were supervised last time and not supervised this time as the first negative sample; The first positive sample and the first negative sample are input into the BP neural network model for training, an analysis model is constructed, and a feedback learning mechanism is established for the analysis model.

3. A mobile supervision and management method for vaccination clinics according to claim 2, characterized in that: The judgment process of the supervision rectification rate is as follows: The feedback learning mechanism based on the analysis model calculates the target cycle changes within the preset time period to obtain the supervision rectification rate parameters: in, It is the parameter for supervising the rectification rate; is the target period within the preset time period; for The number of people who have made rectifications during the last supervision and have been supervised this time; for The number of people who made rectifications during the last supervision but not this time; for The number of people who did not receive supervision last time and this time during the period, and whether they received supervision before is not taken into account; The total number of registered expert file information; for The weight coefficient of the number of registered people in the region within the duration, and 0< <1; The number of supervisions.

4. A mobile supervision and management method for vaccination clinics according to claim 3, characterized in that: The supervision rectification rate parameter The standard threshold interval of the preset supervision correction rate parameter To compare, and is the minimum value of the standard supervision rectification rate parameter, The maximum value of the standard supervision correction rate parameter: like > , then the supervision rectification rate is judged to be high; like ≤ ≤ , then the supervision rectification rate is judged as medium; like < , then it is judged that the supervision rectification rate is low.

5. The mobile supervision and management method for vaccination clinics according to claim 1 is characterized in that: The method of collecting historical supervision rectification rate parameter sets of vaccination clinic management agencies at all levels in S4 and inputting them into the GRU recurrent neural network model for training to obtain the vaccination supervision model is as follows: Obtain positive and negative samples: take the supervision rectification rate parameter set with high supervision rectification rate as the second positive sample and take the supervision rectification rate parameter set with medium and low supervision rectification rate as the second negative sample; The second positive sample and the second negative sample are input into the GRU recurrent neural network model for training and learning to obtain the vaccination supervision model.

6. A mobile supervision and management method for vaccination clinics according to claim 3, characterized in that: The reminder status coefficient: By formula Calculate the first Reminder status coefficient for vaccination clinics ; in, is the target period within the preset time period; The vaccination clinic is in the previous cycle of two consecutive cycles within the preset time period; The vaccination clinic is in the latter cycle of two consecutive cycles within the preset time period; is the first preset weight coefficient, is a second preset weight coefficient; and , All are greater than 0; For the Real-time supervision and rectification rate parameters of level 1 vaccination clinics.

7. A mobile supervision and management method for vaccination clinics according to claim 6, characterized in that: The method of generating a supervisory reminder according to the size of the reminder status coefficient is: Will remind the status coefficient The preset alarm status coefficient threshold To compare the sizes: like ≤ , then it is judged that the completion degree of the current vaccination outpatient supervision work meets the requirements; like > , it is judged that the completion degree of the vaccination work supervised by the current vaccination clinic does not meet the requirements, and a supervision reminder is generated.

8. A method for mobile supervision and management of vaccination clinics according to claim 7, characterized in that: Also includes: Register the evaluation information after supervision and rectification and the reasons for cancellation of supervision, extract information related to the effectiveness of supervision and rectification, and match the vaccination clinics of the corresponding level to send supervision and rectification notifications.

9. A mobile supervision and management system for vaccination clinics, characterized in that: A mobile supervision and management method for a vaccination clinic is used to implement any one of claims 1 to 8, the system comprising: A collection unit is used to collect the historical information of registered supervisory experts in the preset time period of vaccination clinic management agencies at all levels, and count the number of supervisory experts and expert supervision dates based on the registered supervisory expert information; and is used to collect the historical information of appointment cancellation supervisory experts in the preset time period of vaccination clinic management agencies at all levels, and count the number of supervisory experts who have not made an appointment and the expert's last supervision date based on the appointment cancellation supervisory expert information; An analysis unit is used to perform multi-dimensional data analysis based on the registered supervisory expert file information and the appointment cancellation supervisory expert file information to obtain supervisory rectification rate parameters, determine the supervisory rectification rate and send supervisory information to vaccination clinic management agencies at all levels; A model building unit is used to collect statistics on the supervision rectification rate parameter sets of vaccination clinic management agencies at all levels in history and input them into the GRU recurrent neural network model for training to obtain the vaccination supervision model; The supervision unit is used to obtain the supervision rectification rate parameter set of vaccination clinics at all levels in real time and input it into the vaccination supervision model for analysis, obtain the reminder status coefficient of vaccination clinics at all levels, and generate supervision reminders according to the size of the reminder status coefficient.

Images