Method and device for predicting group entering capability of clinical test center
By deploying an entry capability prediction device in the clinical trial center, the maximum and minimum values of entry capability are calculated using the virtual patient data set, the problem of poor accuracy in the evaluation of Site entry capability in clinical trials is solved, and more accurate entry capability prediction and data safety compliance are achieved.
Patent Information
- Application Number
- CN202311420437.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2023-10-30
- Publication Date
- 2025-05-09
AI Technical Summary
In clinical trials, Site's entry ability assessment is poor, and the compliance of information retrieval and competitive entry problems have led to the inability of clinical trials to complete the enrollment and visits of subjects as planned.
A prediction method and device for enrollment capabilities of clinical trial centers is adopted. By receiving query parameter packets, analyzing and retrieving the trial results library of the target clinical trial center, a first virtual patient data set and a second virtual patient data set are generated, the maximum and minimum values of enrollment capabilities are calculated, and the prediction report is generated and the transmission is encrypted.
It improves the accuracy and reliability of the prediction of enrollment capabilities of clinical trial centers, ensures data safety and compliance, and reduces enrollment risks and costs.
Smart Images
Figure CN119964740A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of clinical trial project management, and in particular to a method and device for predicting the enrollment capacity of a clinical trial center. Background Art
[0002] From R&D project establishment to drug regulatory approval and marketing, new innovative drugs need to go through a long R&D and testing cycle, high R&D costs, and a failure rate of over 90%. According to public data, the cost of new drug R&D has risen to US$2.6 billion, taking about 13.5 years. Among them, Phase I to III clinical trials with humans as test subjects are a necessary procedure for the approval of new drugs for marketing, and the cost accounts for 70% of the total investment, which takes an average of 8 years. In this process, selecting a research center (drug clinical research institution, hereinafter referred to as "Site") suitable for the clinical trial project to be carried out, and enrolling subjects who meet the project entry criteria as soon as possible, for the sponsor (the organization or institution that initiates the clinical trial project, usually a pharmaceutical company) to control the project rhythm and cost, and improve the efficiency and quality of trial execution, has become one of the key links that the sponsor must consider when designing the clinical trial plan.
[0003] Generally speaking, when selecting a site, the sponsor's primary consideration is whether the selected site has a sufficiently high enrollment capacity so that the clinical trial can complete the subject enrollment target within the planned time; secondly, they will consider whether the selected site has sufficient influence in the field so that the drug can be successfully sold in the hospital after being approved by the drug regulatory authorities.
[0004] In the actual implementation process, the sponsor will send personnel to the site for a formal visit, interview the principal investigator (PI) to understand the source and distribution of patients with the target indication, and ask the PI to estimate the possible number of patients to be enrolled in this clinical project to evaluate whether the site meets the project's enrollment requirements. This process is called a center screening visit (SSV, Site Selection Visit or PSV, Pre-study Visit). However, due to various reasons such as the limited interview time, personal experience, or insufficient understanding of the data of such patients in the hospital, the conclusion of the SSV survey on enrollment capacity is often far from the actual enrollment capacity in the actual project implementation process, resulting in the inability of the clinical trial to complete the enrollment and visit of subjects as planned. Therefore, with the help of the hospital's completed information platform, a systematic search is conducted throughout the hospital to obtain data and information that initially meet the characteristics of the target subjects of the proposed clinical trial project, which will be able to more accurately support the sponsor's site selection decision and significantly reduce the enrollment risks and cost expenditures that may be faced in the later stage.
[0005] Although the SSV enrollment capacity data obtained by information technology is accurate, it will inevitably come into contact with the hospital's sensitive data and patients' personal privacy information, and will inevitably face the constraints of the "Data Security Law" and the "Personal Information Protection Law" and health and medical management regulations at all levels. How to use the hospital's internal health and medical data reasonably and compliantly to serve the clinical trial business has become a new issue and new direction facing clinical trials.
[0006] In addition, the distribution of domestic drug clinical trials is extremely uneven, with the vast majority of clinical trials concentrated on leading sites and PIs. According to statistics from the CDE official website, the cumulative number of clinical trials undertaken by the top five sites in 2020 accounted for 35% of the total number of clinical trials that year. It is common for projects with the same indication to be clustered in one research center for clinical trials, and there is competition among projects for the same type of subjects. This has also led to a large gap between the site enrollment capacity during the survey and the site enrollment situation during actual implementation, which has brought great uncertainty to the scheduled execution of clinical trial projects.
[0007] In view of the poor accuracy of manual surveys, compliance of information retrieval and competition for enrollment problems faced by the above-mentioned survey on the real enrollment capacity of sites in clinical trials, the present invention proposes a method and device for predicting the real enrollment capacity of sites required by the sponsor, which can not only ensure patient privacy and hospital data security, but also utilize the advantages of information technology to meet data compliance requirements in the hospital. Summary of the invention
[0008] In response to the above technical problems, the present invention provides a method and device for predicting the enrollment capacity of a clinical trial center, which can quickly complete the prediction of the actual enrollment capacity of the site required by the sponsor while ensuring the safe and compliant use of data.
[0009] The present invention provides a method for predicting the enrollment capacity of a clinical trial center, the method comprising: Receiving a query parameter packet from a querying party, parsing the query parameter packet to obtain a first type of query parameter and a second type of query parameter; Respectively searching the first type of query parameter and the second type of query parameter in the trial result database of the target clinical trial center to obtain the corresponding first virtual patient data set and second virtual patient data set; generating a first prediction result according to the first virtual patient data set, wherein the first prediction result corresponds to a maximum value of the enrollment capacity of the target clinical trial center; generating a second prediction result according to a difference set between the first virtual patient data set and the second virtual patient data set, wherein the second prediction result corresponds to a minimum value of the enrollment capacity of the target clinical trial center; generating a prediction report of the target clinical trial center's enrollment capacity according to the first prediction result and the second preset result; and The forecast report is encrypted and sent to the querying party.
[0010] In one embodiment, the prediction report of the target clinical trial center's enrollment capacity displays the prediction results in text and chart format, and the prediction results at least include the enrollment capacity rankings of departments matching the target clinical project.
[0011] In one embodiment, the enrollment capacity ranking display of the departments matching the target clinical project includes: the maximum enrollment value and the degree of competitiveness, wherein the degree of competitiveness is the difference between the maximum number of patients corresponding to the first prediction result and the minimum number of patients corresponding to the second prediction result / the percentage interval of the maximum number of patients corresponding to the first prediction result.
[0012] In one embodiment, it also includes: according to the first type of query parameters input by the query party, the first type of query parameters include a first query element; according to the first query element, a second type of query parameters are generated; the first type of query parameters and the second type of query parameters are packaged and encrypted as the query parameter package.
[0013] In one embodiment, the method further includes: detecting a group of parameters to be confirmed in a public database according to the first type of query elements; and selecting at least one of the group of parameters to be confirmed as the second type of query parameters.
[0014] In one embodiment, when two or more parameters are selected from the set of parameters to be confirmed, two or more corresponding sub-query parameters are generated according to the two or more parameters, and the two or more sub-query parameters serve as the second type of query parameters.
[0015] In one embodiment, the two or more sub-query parameters are retrieved in the trial results library of the target clinical trial center, and two or more sets of virtual patient data subsets are generated accordingly; the second prediction result is generated according to the difference between the first virtual patient data set and the collection of all virtual patient subsets of the two or more sets of virtual patient data subsets.
[0016] In one embodiment, it also includes: extracting a first temporary search result from the trial results library of the target clinical trial center according to the data request range corresponding to the first query parameter, and caching the first temporary search result in a cache unit; retrieving a first patient data set in the cache unit according to the first query parameter; encrypting the identity information, department information and attending physician information of each patient in the first patient data set to obtain virtual identity information, virtual department information and virtual attending physician information of each patient, and using the obtained virtual identity information, virtual department information and virtual attending physician information of each patient as the first virtual patient data set; and clearing the first temporary search result in the cache unit.
[0017] In one embodiment, it also includes: extracting a second temporary search result from the trial results library of the target clinical trial center according to the data request range corresponding to the second query parameter, and caching the second temporary search result in a cache unit; retrieving a second patient data set in the cache unit according to the second query parameter; encrypting the identity information, department information and attending physician information of each patient in the second patient data set to obtain virtual identity information, virtual department information information and virtual attending physician information of each patient, and using the obtained virtual identity information, virtual department information information and virtual attending physician information of each patient as the second virtual patient data set; and clearing the second temporary search result in the cache unit, wherein, when the second query parameter includes multiple sub-query parameters, each sub-query parameter is executed in sequence according to the above steps.
[0018] The present invention also provides a prediction device for the enrollment capacity of a clinical trial center, the prediction device comprising: a receiving module, used to accept a query parameter package from a query party, and parse the query parameter package to obtain a first type of query parameter and a second type of query parameter; a data retrieval module, used to sequentially retrieve the first type of query parameter and the second type of query parameter in the trial results library of the target clinical trial center to obtain the corresponding first virtual patient data set and second virtual patient data set; a first data analysis module, used to generate a first prediction result based on the first virtual patient data set, the first prediction result corresponding to the maximum value of the enrollment capacity of the target clinical trial center; a second data analysis module, used to generate a second prediction result based on the difference set of the first virtual patient data set and the second virtual patient data set, the second prediction result corresponding to the minimum value of the enrollment capacity of the target clinical trial center; a report generation module, used to generate a prediction report of the enrollment capacity of the target clinical trial center based on the first prediction result and the second preset result; and a transmitter, used to send the encrypted prediction report to the query party.
[0019] In one embodiment, it also includes: a query parameter package generation module, which is used to generate second-category query parameters based on first-category query parameters input by the query party, wherein the first-category query parameters include first query elements; and to generate the query parameter package after packaging and encrypting the first-category query parameters and the second-category query parameters.
[0020] In one embodiment, the query parameter package generation module also includes: a second query parameter generation module, which includes a query module and a selection module, the query module is used to detect and obtain a group of parameters to be confirmed in a public database based on the first type of query elements; the selection module is used to select at least one of the group of parameters to be confirmed as the second type of query parameter.
[0021] In one embodiment, the selection module is further used to generate corresponding two or more sub-query parameters based on the two or more parameters when selecting two or more parameters from the group of parameters to be confirmed, and the two or more sub-query parameters serve as the second type of query parameters.
[0022] In one embodiment, the data retrieval module is also used to retrieve the two or more sub-query parameters in the trial results library of the target clinical trial center, and generate two or more groups of virtual patient data subsets accordingly; the second data analysis module is also used to generate the second prediction result based on the difference between the first virtual patient data set and the collection of all virtual patient subsets of the two or more groups of virtual patient data subsets.
[0023] In one embodiment, the selection module is also used to compare each parameter in the set of parameters to be confirmed with the preset parameters of the target project and use the parameters that meet the preset parameters as the second category query parameters; wherein, when two or more parameters in the set of parameters to be confirmed meet the preset parameters, two or more corresponding sub-query parameters are generated based on the two or more parameters, and the two or more sub-query parameters are sorted according to the preset priority as the second category query parameters and displayed on an operation interface.
[0024] Compared with the prior art, the present invention provides a prediction method and prediction device for the enrollment capacity of a clinical trial center. Compared with the enrollment capacity assessment in the traditional SSV survey, the present invention is based on the analysis of real patient visit data within the site, and the conclusions obtained are more accurate and reliable than the traditional focus group or questionnaire survey method. In addition, due to the use of the privacy computing model, the data and patient information within the site do not leave the site, and ciphertext encryption and calculation are used in the entire process of data transmission, storage, and processing, which has advantages in the rigor of personal information protection and the security of data compliance.
[0025] 1) Strict measures to protect patient personal information to avoid the risk of patient privacy leakage. In the data retrieval of the prediction and evaluation of enrollment capacity, the present invention is based only on the patient's ciphertext code information, and no personal information of the patient appears, avoiding the risk of patient privacy leakage; in the enrollment capacity evaluation link, all data processing is completed on the privacy computing device within the site, and the evaluation results are only transmitted in the form of statistical reports. The original data and intermediate data are all in ciphertext form and do not leave the site to ensure data security and compliance.
[0026] 2) Accuracy and reliability of enrollment capacity assessment. The present invention conducts enrollment capacity assessment based on real patient visit data within the site. Compared with the expert discussion or questionnaire survey method used in the traditional SSV survey process, it has certain advantages in data accuracy. In addition, in terms of the analysis of competitive projects, the traditional evaluation method based on the judgment of the sponsor's evaluators on the content of the on-site discussion and personal experience has been changed. Through mathematical processing of real data, a more reliable and reasonable evaluation conclusion is obtained, reducing the risk of evaluation errors.
[0027] 3) Convenience of system deployment of prediction device. The present invention adopts lightweight interface development, does not need to change the existing business processes and business systems of the hospital, and has the characteristics of simple deployment and strong promotion. Through standardized components and hardware and software integrated equipment, it can be quickly deployed in the hospital where the clinical trial project is carried out to complete the enrollment capacity assessment of the clinical trial project. BRIEF DESCRIPTION OF THE DRAWINGS
[0028] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0029] Figure 1 A flow chart of a method for predicting the enrollment capacity of a clinical trial center provided by the present invention; Figure 2 A flow chart of generating the second type of query parameters provided by the present invention; Figure 3 A query flow chart of the first virtual patient data set provided by the present invention; Figure 4 A query flow chart of a second virtual patient data set provided by the present invention; Figure 5 A functional module diagram of a device for predicting the enrollment capacity of a clinical trial center provided by the present invention; Figure 6This is a functional module diagram of the data retrieval module provided by the present invention. Implementation
[0030] In order to provide a further understanding of the purpose, structure, features, and functions of the present invention, the following detailed description is given in conjunction with the embodiments.
[0031] The terms "first", "second", "third", "fourth", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein, for example. In addition, the terms "including" and "corresponding to" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.
[0032] The purpose of the present invention is to provide a prediction method and prediction device that can predict the actual enrollment capacity of a clinical trial center. The prediction device is deployed locally in the clinical trial center, and the execution of the prediction method based on the prediction device is completed within the clinical trial center, and only the results generated by the execution of the prediction method are fed back to the querying party (or the sponsor). The prediction device and prediction method of the present invention use privacy computing technology to obtain basic data such as the trend and distribution of target subjects according to the classification of indications and subject characteristics, and generate a method for evaluating the enrollment capacity of the clinical trial center after analyzing and processing the basic data, so that the sponsor can obtain relatively real enrollment capacity information of the clinical trial center when screening the clinical trial center, and support the screening of the clinical trial center.
[0033] like Figure 1 As shown, the present invention first provides a method 100 for predicting the enrollment capacity of a clinical trial center, comprising: Step S1, receiving a query parameter packet from a querying party, parsing the query parameter packet to obtain a first type of query parameter and a second type of query parameter; Step S2, respectively searching the first type of query parameter and the second type of query parameter in the trial result database of the target clinical trial center to obtain the corresponding first virtual patient data set and second virtual patient data set; Step S3, generating a first prediction result according to the first virtual patient data set, where the first prediction result corresponds to the maximum value of the target clinical trial center's enrollment capacity; Step S4, generating a second prediction result according to the difference between the first virtual patient data set and the second virtual patient data set, the second prediction result corresponding to the minimum value of the target clinical trial center's enrollment capacity; Step S5, generating a prediction report of the target clinical trial center's enrollment capacity according to the first prediction result and the second preset result; and Step S6: Encrypt the forecast report and send it to the inquiring party.
[0034] In this embodiment, the prediction report displays the prediction results in text and chart format, and the prediction results include at least the enrollment capacity ranking of the departments matching the target clinical project. Among them, the enrollment capacity ranking of the departments matching the target clinical project includes: the maximum enrollment value and the competitiveness, and the competitiveness is the difference between the maximum number of patients corresponding to the first prediction result and the minimum number of patients corresponding to the second prediction result / the percentage interval of the maximum number of patients corresponding to the first prediction result. It can be understood that the competitiveness can be divided into multiple levels according to 0-100% as needed, for example: each interval of 10 percentages is one level, divided into ten levels, wherein the greater the competitiveness, the lower the actual enrollment capacity of the clinical trial center.
[0035] Specifically, the prediction report generated according to the first prediction result representing the maximum enrollment capacity and the second prediction result representing the minimum enrollment capacity specifically includes: (1) Basic information of the clinical trial center, which displays the center's bed information, annual outpatient visits, and inpatient composition in the form of numbers and icons; (2) Information on the target clinical trial project, which includes the composition of the clinical trial projects under development of the clinical trial center that are publicly recorded in external databases, including information on the clinical trial project, the corresponding indications of the project, the principal investigator of the clinical trial center and the department where the principal investigator is located, and the total number of patients recruited by the project; (3) The enrollment capacity prediction results are presented in the form of charts based on the highest enrollment capacity report form and the lowest enrollment capacity report form, which are the core contents of the enrollment capacity assessment report for this project; (4) Ranking of experimental departments: Based on the data in the highest enrollment capacity report form and the lowest enrollment capacity report form, the ranking of experimental departments is generated. The specific contents include: department name, number of patients (using the corresponding data of the department in the highest enrollment capacity report form), and competitiveness (using the difference between the corresponding data of the department in the highest enrollment capacity report form and the corresponding data of the department in the lowest enrollment capacity report form / the corresponding data of the department in the highest enrollment capacity report form, and the classification is based on percentage, such as 0-10% is level 1, 10%-20% is level 2, and so on).
[0036] Finally, the above-mentioned project enrollment capacity prediction report is submitted to the sponsor as a reference for selecting clinical trial centers for use by the sponsor.
[0037] Continue to refer to Figure 2 , step S1: also includes: Step S11, according to the first type of query parameters input by the query party, the first type of query parameters include the first query element; Step S12: according to the first type of query elements, a set of parameters to be confirmed is detected in a public database; Step S13: selecting at least one of a group of parameters to be confirmed as a second type of query parameter; Step S14: Pack and encrypt the first type of query parameters and the second type of query parameters to form the query parameter package.
[0038] Specifically, first, the sponsor provides the indications of the target clinical trial project and the registered professional information of the proposed clinical trial to be conducted in the designated clinical trial center, where the indications and the registered professional information of the proposed clinical trial are used as the first type of query parameters. Among them, the main original research information in the registered professional information of the proposed clinical trial is used as the first query element.
[0039] Next, according to the first query element, the prediction device (such as Figure 5 The clinical trial data already filed on the CDE official website are retrieved, and a list of clinical trial projects under investigation that may compete with the target clinical trial project is generated. The list of clinical trial projects under investigation is used to be displayed on the operation interface of the sponsor. The sponsor can select at least one clinical trial project under investigation as a competing project from the list of clinical trial projects under investigation.
[0040] Next, after the sponsor selects at least one ongoing clinical trial project as a competing project on the operation interface, the indications and registered professional information of the target clinical trial project are used as the first query parameter; the indications of the competing project and the information of the principal investigator and the corresponding department published by CED are used as the second query parameter. Among them, when the sponsor selects 2 or more ongoing clinical trial projects as competing projects on the operation interface, the second query parameter includes 2 or more sub-query parameters, and each sub-query parameter includes the indications of each competing project and the information of the principal investigator and the corresponding department published by CED.
[0041] Finally, the prediction device further packages the first query parameter and the second query parameter, encrypts them into a query parameter package, and sends the query parameter package to a data retrieval module deployed locally in the clinical trial center, and completes the retrieval of the first virtual patient data set and the second virtual patient data set in the data retrieval module.
[0042] Reference Figure 3 and Figure 6 , illustrating the process of retrieving the first virtual patient data set in step S2.
[0043] The data retrieval module 3 deployed in the clinical trial center, S2-1a, the data extraction module 31 extracts the first temporary retrieval result from the trial result library 8 of the target clinical trial center according to the data request range (for example: the indication of the target clinical project) corresponding to the first type of query parameters (for example: the diagnosis and treatment information table corresponding to the indication of the target clinical project), and caches the first temporary retrieval result in the cache unit 32; S2-2a, the retrieval engine 33 retrieves the first patient data set in the cache unit 32 according to the first type of query parameters (for example: the indication of the target clinical project); S2-3a, the encryption unit 34 encrypts the identity information, department information and attending physician information of each patient in the first patient data set to obtain the identity virtual information, department information virtual information and attending physician virtual information of each patient, and uses the obtained identity virtual information, department information virtual information and attending physician virtual information of each patient as the first virtual patient data set; and S2-4a, clears the first temporary retrieval result in the cache unit 32.
[0044] Reference Figure 4 and Figure 6 , illustrating the process of retrieving the second virtual patient data set in step S2.
[0045] The data retrieval module 3 deployed in the clinical trial center, S2-1b, the data extraction module 31 extracts the second temporary retrieval result from the trial result library 8 of the target clinical trial center according to the data request range (for example, the diagnosis and treatment information table corresponding to the indications and main researcher information of the competing project and their department information) corresponding to the second type of query parameters (for example, the indications and main researcher information of the competing project and their department information), and caches the second temporary retrieval result in the cache unit 32; S2-2b, the retrieval engine 33 retrieves the second patient data set in the cache unit 32 according to the second type of query parameters (for example, the indications and main researcher information of the competing project and their department information); S2-3b, the encryption unit 34 encrypts the identity information, department information and attending physician information of each patient in the second patient data set to obtain the identity virtual information, department information virtual information and attending physician virtual information of each patient, and uses the obtained identity virtual information, department information virtual information and attending physician virtual information of each patient as the second virtual patient data set; and S2-4b, clears the second temporary retrieval result in the cache unit 32.
[0046] It should be noted that when there are two or more competing projects, a sub-query parameter is generated for each competing project, and each sub-query parameter includes the indication and the main researcher and his department information of the competing project. Each sub-query parameter is detected in the data retrieval module 3 with reference to the above process, and a corresponding virtual patient data subset is obtained, wherein two or more virtual patient data subsets are used as the second virtual patient data set.
[0047] Continue to refer to Figure 1 In step S3, a first prediction result is generated according to the first virtual patient data set, and the first prediction result corresponds to the maximum value of the enrollment capacity of the target clinical trial center.
[0048] Perform statistical analysis on the situation distribution of the first virtual patient data set. Specifically, statistically analyze the diagnosis and treatment situation distribution map of all patients in the first virtual patient data set within a certain time period. For example, generate a report form of the highest enrollment capacity of the clinical trial center for the indication of the target clinical trial by patient source (outpatient, inpatient) and department distribution (based on the department settings within the hospital, corresponding to the registered specialty of the target clinical trial project).
[0049] When there are competing projects in the clinical trial center, the method further includes: in step S4, a second prediction result is generated according to the difference between the first virtual patient data set and the second virtual patient data set, and the second prediction result corresponds to the minimum value of the target clinical trial center's enrollment capacity.
[0050] Process all patient information corresponding to the second virtual patient data set, specifically including: according to the indications of the competing project and the information of the main researcher and his department, retrieve and analyze all patient information in the second virtual patient data set to generate a set of patients to be excluded. The set of patients to be excluded can be the same as the second virtual patient data set. Further, compare the virtual identity information of each patient in the second virtual patient data set with the virtual identity information in the first virtual patient data set, exclude patients with the same virtual identity information, and obtain a possible patient set. The patients in the possible patient set do not overlap with the patients in the competing project. Analyze the possible patient set and generate a minimum enrollment capacity report table for the indication of the target clinical trial by patient source (outpatient, inpatient) and department distribution (based on the setting of the department in the hospital, corresponding to the registered specialty of the target clinical trial project) as the clinical trial center.
[0051] The prediction method provided by the present invention also includes generating a prediction report on the enrollment capacity of the clinical trial center for the target clinical project based on the maximum enrollment capacity report form and the minimum enrollment capacity report form, and the prediction report is sent to the data sponsor after encryption.
[0052] It should be noted that in the above prediction method of the present invention, the query data of the sponsor is received and processed by the prediction device deployed locally in the clinical trial center. At the same time, the search is performed locally in the clinical trial center. The search process is performed by the ciphertext algorithm. The search result is the encrypted processing result, and the prediction report fed back to the sponsor includes the prediction result of the enrollment capacity. The prediction result is not associated with any patient real information data in the clinical trial center. That is, the above prediction method does not retain or provide any patient's personal information in the process from data query to feedback of the prediction report, that is, there is no problem of leakage of any patient's personal information. Since the result of the retrieval module execution is also fed back as an encrypted result, the data of the clinical trial center can be used in compliance with regulations, and it is ensured that the patient's real data is always in the clinical trial center, and the data isolation between the sponsor and the clinical trial center is achieved geographically.
[0053] like Figure 5 and Figure 6 As shown, the present invention also provides a prediction device deployed locally in a clinical trial center, which includes: a receiving module 2, a data retrieval module 3, a first data analysis module 4, a second data analysis module 5, a report generation module 6 and a transmitter 7, wherein the receiving module 2 is used to receive a query parameter package from a query party, parse the query parameter package to obtain a first type of query parameter and a second type of query parameter; the data retrieval module 3 is connected to the receiving module 2 in communication, and retrieves the first type of query parameter and the second type of query parameter in the trial result library 8 of the clinical trial center to obtain the corresponding first virtual patient data set and the second virtual patient data set; the first data analysis module 4 and the second data analysis module 5 communicate with the data retrieval module 2, respectively, and .... Block 3 is connected in communication, the first data analysis module 4 receives the first virtual patient data set to generate a corresponding first prediction result representing the maximum enrollment capacity of the clinical trial center; the second data analysis module 5 receives the first virtual patient data set and the second virtual patient data set, and generates a second prediction result representing the minimum enrollment capacity of the clinical trial center based on the difference between the two; the report generation module 6 is connected in communication with the first data analysis module 4 and the second data analysis module 5, and generates a prediction report including the maximum enrollment capacity and the minimum enrollment capacity of the clinical trial center based on the first prediction result and the second prediction structure; the transmitter 7 is connected in communication with the report generation module 6, and the prediction report is transmitted to the query party 1 by the transmitter 7 after encryption.
[0054] In this embodiment, the prediction device also includes an input unit 11 displayed on the operation interface of the query party 1, and the query party inputs the first type of query information including the indications of the target clinical trial project and the filing information of the planned clinical trial project through the input unit 11.
[0055] The first type of query information input by the input unit 11 is transmitted to the query parameter generation module 9 of the prediction device in an encrypted or unencrypted manner. The query module 91 in the query parameter generation module 9 retrieves a set of parameters to be confirmed in an external database according to the first query element in the first query parameter, such as indication, and the parameters to be confirmed are potential competing projects of the current clinical trial center. Potential competing projects refer to other clinical trial projects under development that are highly similar to or overlap with the target clinical trial project.
[0056] The selection module 92 in the query parameter generation module 9 is displayed on the operation interface of the query party 1, and the query party can select at least one parameter in a group of parameters to be confirmed as a competitive item in the operation interface. It can be understood that when no competitive item is retrieved, the prediction device automatically starts to search and analyze the first type of query parameters, and finally forms a prediction report representing the maximum enrollment capacity of the current clinical trial center.
[0057] It is also understandable that when it is found that there is a competing project in the current clinical trial center, the indication, main researcher and department information of the competing project are selected as the second type of query parameters.
[0058] It is also understandable that when it is found that there are two or more competing projects in the current clinical trial center, the indications, principal investigators and department information of each of the two or more competing projects are selected as sub-query parameters, and the two or more sub-query parameters corresponding to the two or more competing projects are used as the second type of query parameters.
[0059] Reference Figure 5 , and also includes packaging the first type of query parameters and the second type of query parameters, encrypting them into a query parameter package, and sending the query parameter package to the receiving module 2. The receiving module 2 decrypts the query parameters to obtain the first type of query parameters and the second type of query parameters, and then performs retrieval in the above-mentioned data retrieval module to generate the corresponding first virtual patient data set and second virtual patient data set.
[0060] The following combination Figure 6The retrieval process of the first virtual patient data set is described. After the data receiving module 2 receives the first type of query parameters, the data extraction module 31 extracts and obtains the first temporary retrieval result from the trial results library 8 of the clinical trial center according to the data range of the first type of query parameters. The first temporary retrieval result is cached in the cache module 32. The retrieval engine 33 retrieves the first patient data set in the cache module 32 according to the indication information in the first type of query parameters; the encryption unit 34 encrypts the identity information, department information and attending physician information of each patient in the first patient data set to obtain the identity virtual information, department information virtual information and attending physician virtual information of each patient, and uses the obtained identity virtual information, department information virtual information and attending physician virtual information of each patient as the first virtual patient data set. After the retrieval of the first type of query parameters is completed, the first temporary retrieval result in the cache module 32 is cleared.
[0061] The above-mentioned retrieval module effectively avoids the leakage of the patient's real data by encrypting the result feedback when outputting data; the retrieval data is pre-processed by the data extraction module 31 and the cache module 32, so that the data calculation amount of the retrieval process is small, avoiding the high calculation amount and high equipment cost caused by direct retrieval in the trial results library of the clinical trial center.
[0062] It is understandable that the retrieval process of the second type of query parameters in the data retrieval module 3 is similar to the retrieval process of the first type of query parameters in the data retrieval module 3, and the retrieval process of the first type of query parameters may be referred to and will not be described in detail.
[0063] It should be noted that when the second type of query parameters includes multiple sub-query parameters, each sub-query parameter is executed according to the above data retrieval process, and multiple virtual patient subsets are obtained correspondingly, and the multiple virtual patient subsets together constitute the second virtual patient set. In addition, the second prediction report representing the minimum enrollment capacity of the clinical trial center is calculated based on the first virtual patient set and the second virtual patient set, and the difference between each virtual patient subset and the first virtual patient set is calculated to obtain multiple sub-difference sets, and the multiple sub-difference sets are merged and deduplicated as the second virtual patient set; or, after merging multiple virtual patient subsets and performing deduplication, the difference between the union of multiple virtual patient subsets and the first virtual patient set is calculated as the second virtual patient set.
[0064] Continue to refer to Figure 5, the selection module 92 displayed on the operation interface of the query party 1 also includes comparing each parameter in a set of parameters to be confirmed with the preset parameters of the target clinical trial project, and taking the parameters that meet the preset parameters as the second type of query parameters; wherein, when 2 or more parameters in a set of parameters to be confirmed meet the preset parameters, 2 or more corresponding sub-query parameters are generated according to the 2 or more parameters, and the 2 or more sub-query parameters are sorted according to the preset priority as the second type of query parameters, and displayed on the operation interface for selection.
[0065] Among them, the preset parameters refer to the entry conditions for clinical trial projects; the preset priority ranking refers to the ranking according to the scoring results of indications, principal investigators and their department information.
[0066] In addition, in one embodiment, if the querying party does not select a parameter to be confirmed on the operation interface, the query parameter generating module selects all parameters to be confirmed by default to generate corresponding sub-query parameter sets, and uses the sub-query parameter sets as the second type of query parameters.
[0067] In summary, the present invention provides a prediction method and prediction device for the enrollment capacity of a clinical trial center. Compared with the enrollment capacity assessment in the traditional SSV survey, the present invention is based on the analysis of real patient visit data within the site, and the conclusions obtained are more accurate and reliable than the traditional focus group or questionnaire survey method. In addition, due to the use of the privacy computing model, the data and patient information within the site do not leave the site, and ciphertext encryption and calculation are used in the entire process of data transmission, storage, and processing, which has advantages in the rigor of personal information protection and the security of data compliance.
[0068] 1) Strict measures to protect patient personal information to avoid the risk of patient privacy leakage. In the data retrieval of the prediction and evaluation of enrollment capacity, the present invention is based only on the patient's ciphertext code information, and no personal information of the patient appears, avoiding the risk of patient privacy leakage; in the enrollment capacity evaluation link, all data processing is completed on the privacy computing device within the site, and the evaluation results are only transmitted in the form of statistical reports. The original data and intermediate data are all in ciphertext form and do not leave the site to ensure data security and compliance.
[0069] 2) Accuracy and reliability of enrollment capacity assessment. The present invention conducts enrollment capacity assessment based on real patient visit data within the site. Compared with the expert discussion or questionnaire survey method used in the traditional SSV survey process, it has certain advantages in data accuracy. In addition, in terms of the analysis of competitive projects, the traditional evaluation method based on the judgment of the sponsor's evaluators on the content of the on-site discussion and personal experience has been changed. Through mathematical processing of real data, a more reliable and reasonable evaluation conclusion is obtained, reducing the risk of evaluation errors.
[0070] 3) Convenience of system deployment of prediction device. The present invention adopts lightweight interface development, does not need to change the existing business processes and business systems of the hospital, and has the characteristics of simple deployment and strong promotion. Through standardized components and hardware and software integrated equipment, it can be quickly deployed in the hospital where the clinical trial project is carried out to complete the enrollment capacity assessment of the clinical trial project.
[0071] The present invention has been described by the above-mentioned relevant embodiments, but the above-mentioned embodiments are only examples for implementing the present invention. In addition, the technical features involved in the different embodiments of the present invention described above can be combined with each other as long as they do not conflict with each other. It must be pointed out that the disclosed embodiments do not limit the scope of the present invention. On the contrary, changes and modifications made without departing from the spirit and scope of the present invention are all within the scope of patent protection of the present invention.
Claims
1. A method for predicting the enrollment capacity of a clinical trial center, characterized in that: The prediction method comprises: Receiving a query parameter packet from a querying party, parsing the query parameter packet to obtain a first type of query parameter and a second type of query parameter; Retrieve the first type of query parameter and the second type of query parameter from the trial results database of the target clinical trial center respectively to obtain the corresponding first virtual patient data set and second virtual patient data set; generating a first prediction result according to the first virtual patient data set, wherein the first prediction result corresponds to a maximum value of the enrollment capacity of the target clinical trial center; generating a second prediction result according to a difference set between the first virtual patient data set and the second virtual patient data set, wherein the second prediction result corresponds to a minimum value of the enrollment capacity of the target clinical trial center; generating a prediction report of the target clinical trial center's enrollment capacity according to the first prediction result and the second preset result; and The forecast report is encrypted and sent to the querying party.
2. The prediction method according to claim 1, characterized in that: The prediction report of the target clinical trial center's enrollment capacity displays the prediction results in text and chart format, and the prediction results at least include the enrollment capacity rankings of departments matching the target clinical project.
3. The prediction method according to claim 2, characterized in that: The ranking of the enrollment capacity of departments matching the target clinical project includes: the maximum enrollment value and the degree of competitiveness, where the degree of competitiveness is the difference between the maximum number of patients corresponding to the first prediction result and the minimum number of patients corresponding to the second prediction result / the percentage interval of the maximum number of patients corresponding to the first prediction result.
4. The prediction method according to claim 1, characterized in that: Also includes: According to a first type of query parameter input by the querying party, the first type of query parameter includes a first query element; Generate a second type of query parameter according to the first query element; The first type of query parameters and the second type of query parameters are packaged and encrypted as the query parameter package.
5. The prediction method according to claim 4, characterized in that: Also includes: According to the first type of query elements, a set of parameters to be confirmed is detected in a public database; At least one of the group of parameters to be confirmed is selected as the second-category query parameter.
6. The prediction method according to claim 5, characterized in that: When two or more parameters are selected from the set of parameters to be confirmed, two or more corresponding sub-query parameters are generated according to the two or more parameters, and the two or more sub-query parameters serve as the second type of query parameters.
7. The prediction method according to claim 6, characterized in that: Retrieving the two or more sub-query parameters in the trial results database of the target clinical trial center, and correspondingly generating two or more sets of virtual patient data subsets; The second prediction result is generated according to the difference between the first virtual patient data set and the union of all virtual patient subsets of the two or more virtual patient data subsets.
8. The prediction method according to claim 1, characterized in that: Also includes: Extracting a first temporary search result from the trial result library of the target clinical trial center according to the data request range corresponding to the first type of query parameter, and caching the first temporary search result in a cache unit; Retrieving a first patient data set from the cache unit according to the first type of query parameters; encrypting the identity information, department information, and attending physician information of each patient in the first patient data set to obtain virtual identity information, virtual department information, and virtual attending physician information of each patient, and using the obtained virtual identity information, virtual department information, and virtual attending physician information of each patient as the first virtual patient data set; and The first temporary search result in the cache unit is cleared.
9. The prediction method according to claim 8, characterized in that: Also includes: Extracting a second temporary search result from the trial results library of the target clinical trial center according to the data request range corresponding to the second type of query parameter, and caching the second temporary search result in the cache unit; Retrieving a second patient data set from the cache unit according to the second type of query parameters; Encrypting the identity information, department information, and attending physician information of each patient in the second patient data set to obtain virtual identity information, virtual department information, and virtual attending physician information of each patient, and using the obtained virtual identity information, virtual department information, and virtual attending physician information of each patient as the second virtual patient data set; and clearing the second temporary search result in the cache unit, When the second query parameter includes multiple sub-query parameters, each sub-query parameter is executed in sequence according to the above steps.
10. A device for predicting the enrollment capacity of a clinical trial center, characterized in that: The prediction device comprises: A receiving module, configured to receive a query parameter packet from a querying party, and parse the query parameter packet to obtain a first type of query parameter and a second type of query parameter; A data retrieval module, used to sequentially retrieve the first type of query parameters and the second type of query parameters in the trial results library of the target clinical trial center to obtain the corresponding first virtual patient data set and second virtual patient data set; A first data analysis module, configured to generate a first prediction result according to the first virtual patient data set, wherein the first prediction result corresponds to a maximum value of the enrollment capacity of the target clinical trial center; A second data analysis module, configured to generate a second prediction result according to a difference between the first virtual patient data set and the second virtual patient data set, wherein the second prediction result corresponds to a minimum value of the enrollment capacity of the target clinical trial center; a report generating module, configured to generate a prediction report of the target clinical trial center's enrollment capacity according to the first prediction result and the second preset result; and A transmitter is used to send the encrypted prediction report to the querying party.
11. The prediction device according to claim 10, characterized in that: Also includes: A query parameter packet generating module, configured to generate a query parameter packet according to a first type of query parameter input by a querying party, wherein the first type of query parameter includes a first query element; Generate a second type of query parameter according to the first query element; Furthermore, the first type of query parameters and the second type of query parameters are packaged and encrypted to generate the query parameter package.
12. The prediction device according to claim 11, characterized in that: The query parameter packet generation module further includes: a second query parameter generation module, which includes a query module and a selection module, A query module, used for detecting and obtaining a set of parameters to be confirmed in a public database according to the first type of query elements; The selection module is used to select at least one of the group of parameters to be confirmed as the second type of query parameter.
13. The prediction device according to claim 12, characterized in that: The selection module is further configured to generate corresponding two or more sub-query parameters according to the two or more parameters when two or more parameters are selected from the group of parameters to be confirmed, and the two or more sub-query parameters serve as the second type of query parameters.
14. The prediction device according to claim 13, characterized in that: The data retrieval module is also used to retrieve the two or more sub-query parameters in the trial results library of the target clinical trial center, and correspondingly generate two or more sets of virtual patient data subsets; The second data analysis module is further used to generate the second prediction result according to the difference between the first virtual patient data set and the collection of all virtual patient subsets of the two or more virtual patient data subsets.
15. The prediction device according to claim 12, characterized in that: The selection module is further used to compare each parameter in the set of parameters to be confirmed with the preset parameters of the target clinical trial project and use the parameters that meet the preset parameters as the second type of query parameters; Among them, when two or more parameters in the group of parameters to be confirmed meet the preset parameters, two or more corresponding sub-query parameters are generated according to the two or more parameters, and the two or more sub-query parameters are sorted according to the preset priority as the second type of query parameters and displayed on an operation interface.