Medical scheme analysis matching method for pharmaceutical clinical test

Abstract

The invention relates to the technical field of medical management, and particularly discloses a medical scheme analysis and matching method for a pharmaceutical clinical test, which comprises the following steps: when a monitoring period reaches a preset date, a management and control center remotely notifies an inspection group to perform scheme analysis and data monitoring on a medical scheme of a currently performed clinical test; according to the method, the monitoring period is adjusted, companies lacking experience in the clinical experiment process are helped by more scheme analysis and data supervision, meanwhile, the adjusting process is random, and the method is suitable for organizers with impure organizing purposes. Sufficient reasons can be provided for rejecting subsequent clinical experiments to save medical resources for supervision problems found by suddenness and more times of supervision, and meanwhile, the adjustment process is based on actual data on the basis of random, so that an applicant with rich application experience and a complete medical team can be subjected to an adjustment process. Inspection times can be reduced so as to reduce interference and hospital human resource waste.

Description

Technical Field

[0001] The present invention relates to the field of medical management technology, and in particular to a medical scheme analysis and matching method for pharmaceutical clinical trials. Background Art

[0002] Pharmaceutical clinical trials are a method of verifying the effects of drugs by conducting clinical trials on drugs to achieve scientific monitoring. There are usually four phases of clinical trials, which are carried out step by step to ensure the accuracy of experimental results and the effective use of medical resources. In order to ensure the effectiveness of the experiment and the safety of the patients participating in the experiment, the hospital where the clinical trial is held usually regularly monitors and analyzes the medical plan of the sponsor and the medical data generated. This analysis is both helpful and regulatory.

[0003] The level of the sponsors varies, and there will be problems such as inaccurate medical plans and data recording errors. Obviously, fixed regular monitoring will not only waste the hospital's human resources for good sponsors, but also easily cause monitoring burdens. For low-level sponsors, more regulatory analysis is needed to help the experiment. Therefore, a medical plan analysis matching method for pharmaceutical clinical trials is needed, so that the hospital can dynamically adjust the process of regulatory analysis of different sponsors based on the specific experimental process, allocate medical resources more reasonably, and provide more supervision and assistance for medical plan analysis and medical data analysis to companies with insufficient sponsorship experience, and also put greater regulatory pressure on sponsors with impure sponsorship purposes. Summary of the invention

[0004] The purpose of the present invention is to provide a method for analyzing and matching medical regimens for pharmaceutical clinical trials to solve the following technical problems: How can we dynamically adjust the hospital's regulatory analysis process for different sponsors based on the specific experimental process, so as to allocate more medical resources, provide more supervision and assistance for medical plan analysis and medical data analysis to companies with insufficient sponsorship experience, and put greater regulatory pressure on sponsors with impure sponsorship purposes?

[0005] The purpose of the present invention can be achieved by the following technical solutions: A method for analyzing and matching medical plans for pharmaceutical clinical trials comprises the following steps: Build a control center, set up a monitoring cycle and a personnel database for the current clinical trial sponsor in the control center, and select a number of personnel from the personnel database as the inspection team every working day. It should be noted that the control center is unmanned and can be pre-edited software that runs automatically after startup. The personnel database should be updated frequently, and the inspection team selected daily should be notified in advance to prepare for the inspection; Obtain the medical plan and monitoring data of the clinical trial, and construct multi-dimensional feature data based on the acquired monitoring data. Perform data recognition on the feature data based on each dimension of the feature data to obtain the target data. The feature data is the relevant data uploaded by the sponsor, including patient vital signs data of the clinical process, medication data, and drug production data; Count the number of target data. If the number of target data is greater than the preset value, adjust the monitoring cycle. The monitoring cycle adjustment includes increasing the monitoring cycle and reducing the monitoring cycle. By automatically adjusting the inspection cycle, while ensuring that the amount of data is sufficient, on the one hand, it can avoid the monitoring problems that may be caused by regular monitoring as much as possible, and on the other hand, increasing the inspection cycle can reduce the pressure on the hospital to hire personnel; When the monitoring cycle reaches the preset date, the control center will remotely notify the inspection team to conduct program analysis and data monitoring on the medical plan of the current clinical trial.

[0006] Through the above technical scheme: a technical scheme for an unmanned supervision center to manage hospital personnel is provided. The present invention obtains the number of target data that may be retained before data cleaning through the identification of feature data. After judging that the number of target data is sufficient for a data analysis, the monitoring cycle is adjusted based on the experimental data uploaded independently by the sponsor in the clinical trial process, so as to provide more program analysis assistance and data supervision assistance to companies that lack experience in the clinical trial process. At the same time, the adjustment process is random. For sponsors with impure application purposes, the suddenness and more regulatory issues discovered by supervision can also provide sufficient reasons to refuse to conduct subsequent clinical trials to save medical resources. At the same time, the adjustment process is based on actual data on a random basis. For sponsors with rich application experience and a complete medical team, the number of monitoring times can be reduced to reduce interference and waste of hospital human resources.

[0007] As a further technical solution of the present invention: the process of performing data recognition on feature data to obtain target data includes: Set standard values ​​for each dimension of feature data based on historical data; Compare each dimension data of the feature data currently being identified with the standard value to obtain the comparison result; Determine whether the current data is the target data based on the comparison result.

[0008] As a further technical solution of the present invention: the process of obtaining the comparison result and judging whether the current data is the target data according to the comparison result includes: Perform data processing on the dimension data of the feature data and set a weight value for each dimension data; Obtaining a fluctuation coefficient based on the processed data and the weight value, wherein the comparison result is the fluctuation coefficient; Set a comparison interval. If the fluctuation coefficient falls within the comparison interval, the current data is judged to be the target data, otherwise it is judged to be not the target data. Obviously, the comparison interval is set based on the historical data generated in the previous phase of the drug clinical trial process. Especially for the first phase clinical trial, since there is no previous phase experiment, the setting of the comparison interval can be based on the experimental data of the same type of drugs or the target data provided by the sponsor requesting the clinical trial.

[0009] As a further technical solution of the present invention: the generation method of the preset value includes: Set the base value and the added value. The default value is the sum of the base value and the added value. Compare the target data quantity with the feature data quantity remaining after data cleaning during the data monitoring process; If the target data quantity is not less than the remaining feature data quantity after data cleaning, the added value remains unchanged; otherwise, the added value is modified upward, and the modification amount depends on the experimental cycle.

[0010] As a further technical solution of the present invention: the process of adjusting the monitoring cycle includes: Generate a cycle score for the current sponsor based on the last monitoring process. The cycle score is the sum of the sponsor's basic score and monitoring score. When the amount of target data is not less than the preset value, the fluctuation values ​​of the target data are sorted according to the acquisition date; Randomly select several data groups containing the same number of target data, and perform regression analysis on the target data in each data group to obtain the corresponding regression line; Input the regression line slope and cycle score into the preset analysis model and output the analysis results.

[0011] As a further technical solution of the present invention: the process of selecting a data set includes: The difference between the maximum sequence number and the minimum sequence number of any of the data groups is the same and the data repetition rate is lower than a critical value; If the data repetition rate is not lower than the critical value, a new set of data is selected to replace the current data set until the repetition rates of all data sets are lower than the critical value.

[0012] Through the above technical scheme: a scheme for selecting data groups is provided. The data groups of the present invention are the basis for obtaining the stability coefficient. In the process of obtaining the stability coefficient, the more available data there are, the better the final data stability will be. The present invention expands the production of multiple data groups from limited target data through the selection of data groups, thereby obtaining data groups with a step-by-step increase in number. The generation of useless data groups is limited through the sequence number rule and the data repetition rate, and the target data can be covered more completely, so that the acquisition of the stability coefficient is more accurate, and the selection of the data group is random, so that the final result has a certain randomness.

[0013] As a further technical solution of the present invention: the analysis model includes: By formula:

[0014] Get the stability factor ,in is the slope of the regression line corresponding to the jth selected data set, is the preset standard slope, m is the number of selected data sets, j is a non-zero integer less than m, Pr is the period score, α is a preset first correction coefficient, and β is a preset second correction coefficient; The obtained stability coefficient is compared with the preset judgment interval, and the monitoring cycle is adjusted according to the judgment result.

[0015] As a further technical solution of the present invention: the judgment result includes: If the stability coefficient falls into the judgment interval , then from the adjustment interval An integer is randomly selected to adjust the monitoring period; If the stability factor is less than , it is determined that the experimental data of the current cycle does not meet expectations, the first adjustment value is obtained and the monitoring cycle is adjusted; If the stability factor is greater than , then it is judged that the experimental data of the current cycle meets expectations, the second adjustment value is obtained and the monitoring cycle is adjusted.

[0016] Through the above technical scheme: a scheme for adjusting the monitoring period is provided. The monitoring period is adjusted in a variety of ways including adjusting the interval, the first adjustment value and the second adjustment value. The various adjustment methods can not only ensure the unknown nature of the monitoring date, but also to a certain extent ensure that the sponsor who does a better job in the experimental process, data reporting and record keeping based on the experimental data can reduce the number of monitoring times, which can reduce interference to the sponsor and save the hospital's human resources. For the sponsor who makes many mistakes in the experimental process, data reporting and record keeping, the number of monitoring times will be increased, thereby urging the sponsor to record the experimental data more carefully and accurately to ensure the accuracy of the clinical experiment.

[0017] As a further technical solution of the present invention: the first adjustment value is the sum of a first base value and a first correction value, and the second adjustment value is the sum of a second base value and a second correction value.

[0018] Beneficial effects of the present invention: (1) The present invention obtains the number of target data that may be retained before data cleaning through the identification of characteristic data. After determining that the number of target data is sufficient for a data analysis, the present invention adjusts the monitoring cycle based on the experimental data uploaded by the sponsor in the clinical trial process, thereby providing more program analysis assistance and data supervision assistance to companies that lack experience in the clinical trial process. At the same time, the adjustment process is random, and for sponsors with impure sponsor purposes, the suddenness and more regulatory issues discovered by supervision can provide sufficient reasons to refuse to conduct subsequent clinical trials to save medical resources. At the same time, the adjustment process is based on actual data on a random basis, and for sponsors with rich sponsorship experience and a complete medical team, the number of monitoring times can be reduced to reduce interference and waste of hospital human resources.

[0019] (2) The data group of the present invention is the basis for obtaining the stability coefficient. In the process of obtaining the stability coefficient, the more available data there are, the better the final data stability will be. The present invention expands the production of multiple data groups from limited target data through the selection of data groups, thereby obtaining data groups with a step-by-step increase in number. The generation of useless data groups is limited by the sequence number rule and the data repetition rate, and the target data can be covered more completely, so that the acquisition of the stability coefficient is more accurate. In addition, the selection of data groups is carried out randomly, so that the final result has a certain degree of randomness.

[0020] (3) The present invention provides a scheme for adjusting the monitoring period. The monitoring period is adjusted in a variety of ways, including adjusting the interval, the first adjustment value and the second adjustment value. The various adjustment methods can not only ensure the unknown nature of the monitoring date, but also to a certain extent ensure that the sponsor who does a better job of the experimental process, data reporting and record keeping based on the experimental data can reduce the number of monitoring times, which can reduce interference to the sponsor and save the hospital's human resources. For the sponsor who makes many mistakes in the experimental process, data reporting and record keeping, the number of monitoring times will be increased, thereby urging the sponsor to record the experimental data more carefully and accurately, and ensuring the accuracy of the clinical experiment. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] The present invention will be further described below in conjunction with the accompanying drawings.

[0022] Figure 1 It is a flow chart of the overall steps of the present invention; Figure 2 It is a flow chart of the process steps of acquiring target data of the present invention. DETAILED DESCRIPTION

[0023] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0024] See also Figure 1 and Figure 2 As shown, in one embodiment, a medical regimen analysis and matching method for pharmaceutical clinical trials is provided, comprising: S100. Build a control center, set up a monitoring cycle and a personnel database for the current clinical trial sponsor in the control center, and select a number of personnel from the personnel database as the inspection team every working day. It should be noted that the control center is unmanned and can be pre-edited software that runs automatically after startup. The personnel database should be updated frequently, and the inspection team selected daily should be notified in advance to prepare for the inspection; S200, obtaining the medical plan and monitoring data of the clinical trial, and constructing multi-dimensional feature data based on the acquired monitoring data, and performing data recognition on the feature data based on each dimension of the feature data to obtain target data. The feature data is the relevant data uploaded by the sponsor, including patient vital signs data of the clinical process, medication data, and drug production data; S300, counting the number of target data, if the number of target data is greater than a preset value, adjusting the monitoring cycle, the monitoring cycle adjustment includes increasing the monitoring cycle and reducing the monitoring cycle, by automatically adjusting the inspection cycle, it is possible to avoid monitoring problems that may be caused by regular monitoring as much as possible while ensuring sufficient data volume, and on the other hand, increasing the inspection cycle can reduce the employment pressure of the hospital. Sufficient number of target data is a critical condition for data cleaning. If the number is insufficient, it is basically impossible to analyze the medical data by means of data cleaning, and thus it is impossible to provide a data basis for subsequent experimental processes; S400. When the monitoring cycle reaches the preset date, the control center remotely notifies the inspection team to conduct program analysis and data monitoring on the medical program of the current clinical trial.

[0025] In this embodiment, a technical solution for an unmanned supervision center to manage hospital personnel is provided. The present invention obtains the number of target data that may be retained before data cleaning through the identification of feature data. After judging that the number of target data is sufficient for a data analysis, the monitoring cycle is adjusted based on the experimental data uploaded independently by the sponsor in the clinical trial process, so as to provide more program analysis assistance and data supervision assistance to companies that lack experience in the clinical trial process. At the same time, the adjustment process is random. For sponsors with impure application purposes, the suddenness and more regulatory issues discovered by supervision can also provide sufficient reasons to refuse to conduct subsequent clinical trials to save medical resources. At the same time, the adjustment process is based on actual data on a random basis. For sponsors with rich application experience and a complete medical team, the number of monitoring times can be reduced to reduce interference and waste of hospital human resources.

[0026] The process of performing data recognition on feature data to obtain target data includes: S210, setting standard values ​​for each dimension of feature data based on historical data, wherein the standard values ​​are set according to actual conditions, such as dosage of medication for different patients and treatment plans for different conditions; S220, comparing each dimension data of the feature data currently being identified with the standard value to obtain a comparison result; S230: Determine whether the current data is the target data based on the comparison result.

[0027] The process of obtaining the comparison result and determining whether the current data is the target data according to the comparison result includes: S221, performing data processing on the dimension data of the feature data, and setting a weight value for each dimension data; S222, obtaining a fluctuation coefficient based on the processed data and the weight value, and the comparison result is the fluctuation coefficient; S223. Set a comparison interval. If the fluctuation coefficient falls within the comparison interval, the current data is judged to be the target data, otherwise it is judged to be not the target data. Obviously, the comparison interval is set based on the historical data generated in the previous phase of the drug clinical trial process. Especially for the first phase clinical trial, since there is no previous phase experiment, the setting of the comparison interval can be based on the experimental data of the same type of drugs or the target data provided by the sponsor requesting the clinical trial.

[0028] As an example, by the formula:

[0029] Get the volatility coefficient ,in is the weight value of the i-th dimension data, is the i-th dimension data, is the standard value of the i-th dimension data, where for the dimension data corresponding to the vital sign data, the standard value should be set according to the status of different patients, n is the number of dimensions of the feature data, i is a non-zero integer less than or equal to n, and the process of data processing of the dimension data is as follows: .

[0030] The preset values ​​are generated by: S310, setting a basic value and an added value, wherein the preset value is the sum of the basic value and the added value; S320, comparing the target data quantity with the feature data quantity remaining after data cleaning during the data monitoring process; S330. If the target data quantity is not less than the remaining feature data quantity after data cleaning, the added value remains unchanged; otherwise, the added value is modified upward, and the modification amount is determined according to the experimental cycle.

[0031] The process of adjusting the monitoring cycle includes: S340, based on the last monitoring process, a cycle score is generated for the current sponsor. The cycle score is the sum of the sponsor's basic score and the monitoring score. The monitoring score is based on the sum of the scores of all monitoring events in the inspection process. The inspection event score is selected from the preset event score table based on the monitoring event type. It should be noted that if there is no last monitoring process, such as the first monitoring process of a phase I clinical trial, the monitoring score is 0; S350, when the amount of target data is not less than a preset value, sorting the fluctuation values ​​of the target data according to the acquisition date; S360, randomly selecting a number of data groups containing the same number of target data, and performing regression analysis on the target data in each data group to obtain a corresponding regression line; S370: Input the regression line slope and the cycle score into a preset analysis model and output the analysis results.

[0032] The process of selecting a data set includes: S361: The difference between the maximum sequence number and the minimum sequence number of any data group is the same and the data repetition rate is lower than the critical value. The data repetition rate is the ratio of the number of target data repeated in the current data group compared with other data groups to the total number of the current data group. The critical value is a preset value. S362. If the data repetition rate is not lower than the critical value, a new set of data is selected to replace the current data set until the repetition rates of all data sets are lower than the critical value.

[0033] In this embodiment, a scheme for selecting data groups is provided. The data groups of the present invention are the basis for obtaining the stability coefficient. In the process of obtaining the stability coefficient, the more available data there are, the better the final data stability will be. The present invention expands the production of multiple data groups from limited target data through the selection of data groups, thereby obtaining data groups with a step-by-step increase in number. The generation of useless data groups is limited through the sequence number rule and the data repetition rate, and the target data can be covered more completely, so that the acquisition of the stability coefficient is more accurate. In addition, the selection of data groups is performed randomly, so that the final result has a certain degree of randomness.

[0034] The analytical model includes: By formula:

[0035] Get the stability factor ,in is the slope of the regression line corresponding to the jth selected data set, is the preset standard slope, m is the number of selected data sets, j is a non-zero integer less than m, Pr is the cycle score, α is the preset first correction coefficient, β is the preset second correction coefficient, The correction factor is a preset constant, set based on empirical data; The obtained stability coefficient is compared with the preset judgment interval, and the monitoring cycle is adjusted according to the judgment result.

[0036] The judgment results include: If the stability coefficient falls into the judgment interval , then from the adjustment interval Randomly select an integer between to adjust the monitoring period. a and b respectively according to and The corresponding value is selected from the preset table function. Both a and b are positive numbers. In the process of random selection, the positive and negative lengths of the interval are controlled by the values ​​of a and b, so as to increase the probability of the monitoring cycle for the better sponsor, rather than being completely random. If the stability factor is less than , it is determined that the experimental data of the current cycle does not meet expectations, the first adjustment value is obtained and the monitoring cycle is adjusted; If the stability factor is greater than , then it is judged that the experimental data of the current cycle meets expectations, the second adjustment value is obtained and the monitoring cycle is adjusted.

[0037] In this embodiment, a solution for adjusting the monitoring period is provided. The monitoring period is adjusted in a variety of ways including adjusting the interval, the first adjustment value and the second adjustment value. The various adjustment methods can not only ensure the unknown nature of the monitoring date, but also to a certain extent ensure that the sponsor who does a better job of the experimental process, data reporting and record keeping based on the experimental data can reduce the number of monitoring times, which can reduce interference to the sponsor and save the hospital's human resources. For the sponsor who makes many mistakes in the experimental process, data reporting and record keeping, the number of monitoring times will be increased, thereby urging the sponsor to record the experimental data more carefully and accurately to ensure the accuracy of the clinical experiment.

[0038] The first adjustment value is the sum of the first base value and the first correction value, and the second adjustment value is the sum of the second base value and the second correction value;

[0039] in is the first adjustment value, is the first base value, which is a preset constant and is selected from the preset comparison table based on the data of the stability coefficient. is the first correction value, To determine the continuous stability coefficient is less than In the case of , the number of monitoring cycles involved, the monitoring cycle is the time before two consecutive monitoring days after the start of the experiment, is the basic increase, which is a preset constant. is the second adjustment value, is the second base value, which is a preset constant and is selected from the preset comparison table based on the data of the stability coefficient. is the second correction value, To determine the continuous stability coefficient greater than In the case of is the basic reduction amount, which is a preset constant.

[0040] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. A method for analyzing and matching medical regimens for pharmaceutical clinical trials, characterized in that: The steps include: Establish a control center, set up a monitoring cycle and a personnel database for the current clinical trial sponsor in the control center, and select several personnel from the personnel database as the inspection team; Obtain the medical plan and monitoring data of the clinical trial, and construct multi-dimensional feature data based on the acquired monitoring data, and perform data recognition on the feature data based on each dimension of the feature data to obtain the target data; Count the number of target data. If the number of target data is greater than the preset value, adjust the monitoring cycle. The adjustment of the monitoring cycle includes increasing the monitoring cycle and reducing the monitoring cycle. When the monitoring cycle reaches the preset date, the control center will remotely notify the inspection team to conduct program analysis and data monitoring on the medical plan of the current clinical trial.

2. A medical plan analysis and matching method for pharmaceutical clinical trials according to claim 1, characterized in that: The process of performing data recognition on feature data to obtain target data includes: Set standard values ​​for each dimension of feature data based on historical data; Compare each dimension data of the feature data currently being identified with the standard value to obtain the comparison result; Determine whether the current data is the target data based on the comparison result.

3. A medical plan analysis and matching method for pharmaceutical clinical trials according to claim 2, characterized in that: The process of obtaining the comparison result and determining whether the current data is the target data according to the comparison result includes: Perform data processing on the dimension data of the feature data and set a weight value for each dimension data; Obtaining a fluctuation coefficient based on the processed data and the weight value, wherein the comparison result is the fluctuation coefficient; Set a comparison interval. If the fluctuation coefficient falls within the comparison interval, the current data is judged to be the target data.

4. A method for analyzing and matching medical regimens for pharmaceutical clinical trials according to claim 1, characterized in that: The generation method of the preset value includes: Set the base value and the added value. The default value is the sum of the base value and the added value. Compare the target data quantity with the feature data quantity remaining after data cleaning during the data monitoring process; If the target data quantity is not less than the remaining feature data quantity after data cleaning, the added value remains unchanged, otherwise the added value is modified upward.

5. The medical plan analysis and matching method for pharmaceutical clinical trials according to claim 1, characterized in that: The process of adjusting the monitoring cycle includes: Generate a cycle score for the current sponsor, which is the sum of the sponsor's basic score and monitoring score; When the amount of target data is not less than the preset value, the fluctuation values ​​of the target data are sorted according to the acquisition date; Randomly select several data groups containing the same number of target data, and perform regression analysis on the target data in each data group to obtain the corresponding regression line; Input the regression line slope and cycle score into the preset analysis model and output the analysis results.

6. A method for analyzing and matching medical regimens for pharmaceutical clinical trials according to claim 5, characterized in that: The process of selecting a data set includes: The difference between the maximum sequence number and the minimum sequence number of any of the data groups is the same and the data repetition rate is lower than a critical value; If the data repetition rate is not lower than the critical value, a new set of data is selected to replace the current data set until the repetition rates of all data sets are lower than the critical value.

7. A method for analyzing and matching medical plans for pharmaceutical clinical trials according to claim 1, characterized in that: The analytical models include: By formula: Get the stability factor ,in is the slope of the regression line corresponding to the jth selected data set, is the preset standard slope, m is the number of selected data sets, j is a non-zero integer less than m, Pr is the cycle score, α is the preset first correction coefficient, and β is the preset second correction coefficient; The obtained stability coefficient is compared with the preset judgment interval, and the monitoring cycle is adjusted according to the judgment result.

8. A method for analyzing and matching medical regimens for pharmaceutical clinical trials according to claim 7, characterized in that: The judgment results include: If the stability coefficient falls into the judgment interval , then from the adjustment interval An integer is randomly selected to adjust the monitoring period; If the stability factor is less than , it is determined that the experimental data of the current cycle does not meet expectations, the first adjustment value is obtained and the monitoring cycle is adjusted; If the stability factor is greater than , then it is judged that the experimental data of the current cycle meets expectations, the second adjustment value is obtained and the monitoring cycle is adjusted.

9. A method for analyzing and matching medical regimens for pharmaceutical clinical trials according to claim 8, characterized in that: The first adjustment value is the sum of a first base value and a first correction value, and the second adjustment value is the sum of a second base value and a second correction value.

Images