A method, system and storage medium for dynamic scheduling of test follow-up based on rule engine

Abstract

The present invention relates to the field of clinical information, specifically a method and system for dynamic scheduling of trial follow-up based on a rule engine, which obtains basic data of clinical trial participants, generates information containers of participants, and configures a rule engine for the information container; obtains trial-related data of participants, predicts changes in clinical symptoms, uses a rule engine to make decisions and judgments on clinical symptom changes and participants' questionnaire responses, and generates a follow-up schedule; based on the implementation results of the follow-up schedule, updates the information container and reconfigures the rule engine, thereby adjusting the follow-up schedule for the next predicted clinical symptom changes and questionnaire responses, and dynamically updates the follow-up schedule. The present invention adaptively changes the follow-up schedule based on changes in the physical condition, somatosensory sensations, and drug efficacy of participants during drug trials, ensuring that participants obtain follow-up docking that matches the actual trial situation during each follow-up process, thereby improving the effectiveness and quality of the follow-up schedule.

Description

Technical Field

[0001] The present invention relates to the field of clinical information, and in particular to a rule engine-based dynamic trial follow-up scheduling method, system and storage medium. Background Art

[0002] Clinical trials are systematic studies of drugs conducted on human volunteers and are essential for the certification and marketing of new drugs. As direct participants in clinical trials, the feedback provided by volunteers during clinical trials directly impacts the quality of these trials. Therefore, establishing a follow-up mechanism for participants during clinical trials is crucial for identifying efficacy and side effects. However, given the diverse mechanisms of action and intended treatment of different drugs, there is no single standardized follow-up model or standard for clinical trials.

[0003] In order to reduce the complexity of follow-up, a regular scheduling mode is adopted in the follow-up operation for the clinical trial of the same drug. For example, the invention patent CN104820903A discloses a mobile management method and management system for medical follow-up schedules. By recommending follow-up dates, recording and arranging personal schedules, counting the number of all follow-up patients on a specific date, sending SMS reminders of follow-up time, and sending patient education notifications, patients can be given standardized SMS reminders according to the post-operative schedule, thereby realizing standardized and regular follow-up schedules for patients. However, the follow-up arrangement provided by the above invention patent is regular, and cannot adapt to the actual physical condition, physical condition and drug efficacy of the patient (participant) to dynamically arrange follow-up tasks and processes during the entire follow-up period, thereby reducing the quality and reliability of the follow-up. Therefore, how to make rule decisions based on the actual physical condition, physical condition and drug efficacy of different participants and dynamically set follow-up schedules is of great significance for obtaining test data in real time and accurately and improving the quality and reliability of the test. Summary of the Invention

[0004] In order to adapt the follow-up schedule to the changes in participants' physical condition, physical sensations, and drug effects during drug trials, ensure that participants receive follow-up matching that matches the actual trial situation during each follow-up, improve the effectiveness of follow-up scheduling, and provide reliable reference data for the trial, the present invention provides a rule-based dynamic scheduling method for trial follow-up, which includes the following steps:

[0005] S100: Obtain basic data of clinical trial participants and generate an information container of the participants; configure a rule engine for the information container based on information elements of the information container;

[0006] S200: Acquire the participant's trial-related data and predict clinical symptom changes; use the rule engine to make decisions based on the clinical symptom changes and the participant's questionnaire responses to generate a follow-up schedule;

[0007] S300, based on the implementation results of the follow-up schedule, the information container is updated and the rule engine is reconfigured; the follow-up schedule is adjusted based on the next predicted clinical symptom changes and questionnaire responses using the reconfigured rule engine, and the follow-up schedule is updated and set.

[0008] Preferably, in S100, basic data of clinical trial participants are obtained to generate information containers of the participants; and a rule engine is configured for the information containers according to information elements of the information containers, specifically:

[0009] Obtaining the clinical trial participant's medical condition data, physical status data, and trial protocol data, performing natural semantic recognition and conversion on the medical condition data, physical status data, and trial protocol data to generate an information container for the participant; wherein the information container includes a medical condition information element, a physical status information element, and a trial protocol information element;

[0010] Based on the information elements of the information container, a mapping relationship between the disease information, the physical condition information, and the test plan information is constructed; based on the mapping relationship, a matching decision rule set is selected to configure a rule engine for the information container; wherein, the rule engine executes follow-up scheduling based on the decision rule set.

[0011] Preferably, in S200, the trial-related data of the participant is obtained to predict clinical symptom changes; the rule engine is used to perform decision screening on the clinical symptom changes and the participant's questionnaire responses to generate a follow-up schedule, specifically:

[0012] Obtaining the trial strategy data and trial monitoring data already executed by the participant from the knowledge base, correlating and benchmarking the trial strategy data and the trial monitoring data to obtain a trial dataset; using the trial dataset and the large model, predicting clinical symptom changes occurring in the participant, and generating a clinical symptom feature vector;

[0013] The participant's questionnaire responses are subjected to natural text semantic recognition and conversion to obtain a questionnaire response text vector; the rule engine is used to make decisions and judgments on the clinical symptom feature vector and the questionnaire response text vector to obtain the feasibility of each task in the follow-up task set for the participant, thereby generating a follow-up schedule.

[0014] Preferably, in S300, according to the implementation result of the follow-up schedule, the information container is updated and the rule engine is reconfigured, specifically:

[0015] Performing semantic recognition and information extraction on the implementation results of the follow-up schedule to obtain feedback data that is content-related to the basic data; using the feedback data, updating the invalid information elements of the information container;

[0016] According to the mapping relationship between the information elements updated in the information container, a matching decision rule set is reselected, thereby reconfiguring a rule engine for the information container.

[0017] Preferably, in S300, the reconfigured rule engine is used to adjust the follow-up schedule based on the next predicted clinical symptom changes and questionnaire responses, and the follow-up schedule is updated and set, specifically:

[0018] Obtaining a clinical symptom feature vector corresponding to the next predicted clinical symptom change and a questionnaire response text vector corresponding to the next questionnaire response; utilizing a reconfigured rule engine to make decisions and judgments on the clinical symptom feature vector and the questionnaire response text vector, and adjusting the follow-up tasks included in the follow-up schedule;

[0019] Adjust the content according to the follow-up tasks in the follow-up schedule, set up online notifications and visualize the schedule calendar.

[0020] In another aspect, the present invention provides a rule engine-based dynamic scheduling system for trial follow-up, the system comprising the following modules:

[0021] An information container generation module, configured to obtain basic data of clinical trial participants and generate information containers for the participants;

[0022] A rule engine configuration module, configured to configure a rule engine for the information container according to the information element of the information container;

[0023] A prediction module, used to obtain trial-related data of the participant and predict changes in clinical symptoms;

[0024] a schedule generation module, configured to use the rule engine to make decisions based on the clinical symptom changes and the participant's questionnaire responses to generate a follow-up schedule;

[0025] An updating module, configured to update the information container and reconfigure the rule engine according to the implementation result of the follow-up schedule;

[0026] The schedule adjustment and setting module is used to adjust the follow-up schedule based on the clinical symptom changes and questionnaire responses for the next prediction using the reconfigured rule engine, and to update and set the follow-up schedule.

[0027] Preferably, the information container generation module is used to obtain basic data of clinical trial participants and generate the information container of the participants, specifically:

[0028] Obtaining the clinical trial participant's medical condition data, physical status data, and trial protocol data, performing natural semantic recognition and conversion on the medical condition data, physical status data, and trial protocol data to generate an information container for the participant; wherein the information container includes a medical condition information element, a physical status information element, and a trial protocol information element;

[0029] The rule engine configuration module is used to configure a rule engine for the information container according to the information element of the information container, specifically:

[0030] Based on the information elements of the information container, a mapping relationship between the disease information, the physical condition information, and the test plan information is constructed; based on the mapping relationship, a matching decision rule set is selected to configure a rule engine for the information container; wherein, the rule engine executes follow-up scheduling based on the decision rule set.

[0031] Preferably, the prediction module is used to obtain the participant's test-related data and predict clinical symptom changes, specifically:

[0032] Obtaining the trial strategy data and trial monitoring data already executed by the participant from the knowledge base, correlating and benchmarking the trial strategy data and the trial monitoring data to obtain a trial dataset; using the trial dataset and the large model, predicting clinical symptom changes occurring in the participant, and generating a clinical symptom feature vector;

[0033] The schedule generation module is used to use the rule engine to make decisions based on the clinical symptom changes and the participants' questionnaire responses to generate a follow-up schedule, specifically:

[0034] The participant's questionnaire responses are subjected to natural text semantic recognition and conversion to obtain a questionnaire response text vector; the rule engine is used to make decisions and judgments on the clinical symptom feature vector and the questionnaire response text vector to obtain the feasibility of each task in the follow-up task set for the participant, thereby generating a follow-up schedule.

[0035] Preferably, the updating module is used to update the information container and reconfigure the rule engine according to the implementation result of the follow-up schedule, specifically:

[0036] Performing semantic recognition and information extraction on the implementation results of the follow-up schedule to obtain feedback data that is content-related to the basic data; using the feedback data, updating the invalid information elements of the information container;

[0037] Reselecting a matching decision rule set based on the mapping relationship between the information elements updated in the information container, thereby reconfiguring a rule engine for the information container;

[0038] The schedule adjustment and setting module is used to adjust the follow-up schedule based on the next predicted clinical symptom changes and questionnaire responses using the reconfigured rule engine, and update the follow-up schedule, specifically:

[0039] Obtaining a clinical symptom feature vector corresponding to the next predicted clinical symptom change and a questionnaire response text vector corresponding to the next questionnaire response; utilizing a reconfigured rule engine to make decisions and judgments on the clinical symptom feature vector and the questionnaire response text vector, and adjusting the follow-up tasks included in the follow-up schedule;

[0040] Adjust the content according to the follow-up tasks in the follow-up schedule, set up online notifications and visualize the schedule calendar.

[0041] In addition, the present invention also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it implements the above-mentioned rule engine-based dynamic scheduling method for trial follow-up.

[0042] Compared with the prior art, the present invention has the following beneficial effects:

[0043] The follow-up schedule is adaptively changed according to the changes in the participants' physical condition, somatic sensations and drug efficacy during the drug trial. After each drug treatment course is completed during the drug clinical trial, the participant's information container and rule engine are updated to provide a new data information basis and decision engine for resetting the content of the next follow-up task, ensuring that participants receive follow-up docking that matches the actual trial situation during each follow-up process, thereby improving the effectiveness and quality of the follow-up schedule. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without inventive efforts. Among them:

[0045] Figure 1 This is a flow chart of a method for dynamic trial follow-up scheduling based on a rule engine provided by the present invention.

[0046] Figure 2 This is a schematic diagram of information container construction.

[0047] Figure 3 This is a schematic diagram of rule engine construction.

[0048] Figure 4 It is the interface of the trial follow-up management platform.

[0049] Figure 5 It is the knowledge base operation interface.

[0050] Figure 6 This is the questionnaire interface.

[0051] Figure 7 This is the follow-up scheduling notification interface.

[0052] Figure 8 It is a visual interface for scheduling calendar.

[0053] Figure 9 This is a structural diagram of a test follow-up dynamic scheduling system based on a rule engine provided by the present invention. DETAILED DESCRIPTION

[0054] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below in conjunction with the accompanying drawings. It will be understood that the specific embodiments described herein are only used to explain the present invention, rather than to limit the present invention. It should also be noted that, for ease of description, only parts related to the present invention, rather than all structures, are shown in the accompanying drawings. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0055] The terms "comprise," "comprising," and "having," and any variations thereof, as used herein, are intended to cover non-exclusive inclusions. For example, a process, method, system, product, or apparatus comprising a series of steps or elements is not limited to the listed steps or elements, but may optionally include steps or elements not listed, or may optionally include other steps or elements inherent to the process, method, product, or apparatus.

[0056] References herein to "embodiments" mean that a particular feature, structure, or characteristic described in connection with the embodiments may be included in at least one embodiment of the present invention. The appearance of this phrase in various places in the specification does not necessarily refer to the same embodiment, nor does it constitute a separate or alternative embodiment that is mutually exclusive of other embodiments. It is understood, both explicitly and implicitly, by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0057] See also Figure 1 As shown, the present invention provides a method for dynamic scheduling of trial follow-up based on a rule engine, the method comprising the following steps:

[0058] S100, obtaining basic data of clinical trial participants and generating information containers of the participants; configuring a rule engine for the information container based on information elements of the information container.

[0059] Furthermore, in S100, basic data of clinical trial participants are obtained to generate information containers of the participants; and a rule engine is configured for the information container based on the information elements of the information container, specifically:

[0060] Obtaining the clinical trial participant's medical condition data, physical status data, and trial protocol data, performing natural semantic recognition and conversion on the medical condition data, physical status data, and trial protocol data, and generating an information container for the participant; wherein the information container includes a medical condition information element, a physical status information element, and a trial protocol information element;

[0061] Based on the information elements of the information container, a mapping relationship between the disease information, physical status information, and test plan information is constructed; based on the mapping relationship, a matching decision rule set is selected to configure a rule engine for the information container; wherein the rule engine executes follow-up scheduling based on the decision rule set.

[0062] Drug clinical trials aim to verify the efficacy of a drug in different populations. To minimize the influence of subjective factors on trial results, a double-blind trial is typically used. This involves dividing all participants into an experimental group and a control group, each receiving the drug and a placebo, respectively. The provision of medication to participants in the experimental group requires a personalized trial plan, meaning that different participants receive different dosages of the drug. Given the unique characteristics of clinical trial participants, the follow-up needs of different participants in the experimental group vary during the trial. Specifically, when developing a follow-up schedule for a participant (including required follow-up examinations and their timing), a dedicated container for each participant is pre-built to store relevant data from the participant throughout the drug trial. This not only isolates data for different participants during the designation and adjustment of follow-up schedules, but also provides a comprehensive and reliable data foundation for follow-up schedule development.

[0063] The effectiveness of the drug on the participant is closely related to the participant's own condition, physical condition, and trial plan. In order to implement personalized follow-up scheduling for each participant and ensure that each participant receives follow-up arrangements when the drug truly takes effect during the trial, it is necessary to build an independent information container for each participant; the above information container can be understood as a dedicated storage area set for the participant in the trial follow-up management platform. Figure 2At the initial stage of a clinical trial, participants' medical condition data, physical condition data, and trial protocol data are collected. Medical condition data may include, but are not limited to, the participant's disease type and duration of illness; physical condition data may include, but are not limited to, the participant's gender, age, weight, blood pressure, blood sugar, and history of drug allergies; and trial protocol data may include, but are not limited to, the participant's medication dosage and the interval between medications. Natural language semantic recognition and conversion are performed on these medical condition data, physical condition data, and trial protocol data to extract medical condition information elements, physical condition information elements, and trial protocol information elements. These medical condition information elements, physical condition information elements, and trial protocol information elements all take the form of information pairs, i.e., information elements consisting of a one-to-one correspondence between an information name and information content. For example, a medical condition information element may include, for example, disease type: diabetes, duration: 3 years; a physical condition information element may include, for example, gender: male, age: 45 years, weight: 80 kg; and a trial protocol information element may include, for example, medication dosage: metformin hydrochloride sustained-release tablets, active ingredient 1 mg, and medication interval: 4 hours. The participants' medical condition information elements, physical condition information elements, and test plan information elements are stored together in an information container, effectively isolating the information of different participants and avoiding information mixing and interference; when subsequent test conditions change, only the information elements of the information container need to be updated separately, without involving a large number of information changes, reducing the workload of changing the follow-up schedule of participants.

[0064] The purpose of follow-up visits in drug clinical trials is to assess the physical changes in different types of participants after drug administration, providing comprehensive and reliable research data to determine drug efficacy, side effects, and appropriate populations. Follow-up scheduling primarily involves two aspects: follow-up tasks and their implementation time (frequency). Follow-up tasks refer to the physical examinations performed on participants during the follow-up period. Depending on the participant's stage of the clinical trial and their physical condition, the tasks and their implementation time (frequency) vary during each follow-up period. For example, in the initial stages of a drug clinical trial, follow-up tasks such as blood draws, CT scans, and MRI scans may be performed on participants, and these tasks may be performed more frequently (i.e., more frequently). In the final stages of a drug clinical trial, only blood draws and tests may be required, and these tasks may be performed less frequently (i.e., less frequently). The above analysis demonstrates that the requirements for follow-up scheduling vary across different stages of a drug clinical trial. The specific follow-up schedule for each stage depends on the participant's medical condition, physical condition, and trial protocol information. In order to accurately formulate appropriate follow-up schedules for participants based on their medical condition information, physical status information, and trial plan information, a planning engine is used to analyze and make decisions on all the information in the information container corresponding to the participant, thereby formulating and executing a follow-up schedule that matches the current drug clinical trial needs. In practical applications, it is preferred to use rule engines such as the ODM rule engine or the Drools rule engine. The information container contains multiple information elements corresponding to the medical condition information, physical status information, and trial plan information. These information elements are independent of each other in their original state. If these information elements are directly used to set the decision procedure set of the rule engine (that is, the set of triggering logic condition rules for the rule engine to make decisions), it will not only increase the amount of calculation required for the rule engine to run, but also fail to reflect the joint impact of different types of information elements in the information container on the execution of the follow-up schedule. For this purpose, please refer to Figure 3First, all information elements in the information container are converted to semantic vectors. Based on the clustering characteristics between different semantic vectors, a mapping relationship is established between the disease information, physical status information, and trial plan information. The clustering characteristics between different semantic vectors can be determined by calculating the distance between different semantic vectors. The smaller the distance between different semantic vectors, the higher the clustering degree of the corresponding semantic vector. This is a conventional technical means in the field and will not be described in detail here. Then, based on the above clustering characteristics, the disease information elements, physical status information elements, and trial plan information elements belonging to the same cluster are mapped and compared to generate corresponding mapping relationships (i.e., mapping logical relationships). For example, disease type: diabetes, age: 55 years old, and drug name: metformin hydrochloride sustained-release tablets can be considered to belong to the same cluster, and a mapping relationship is established between these information elements. Based on the above mapping relationship, a portion of decision rules are selected from the original decision rule set to form a matching decision rule set. This matching decision rule set is then assigned to the information container. In this way, the rule engine can identify and judge follow-up tasks based on the matching decision rule set and generate a follow-up schedule that matches the information elements currently contained in the information container. For example, selecting a decision rule according to the mapping relationship may include selecting a decision rule from an original decision rule set according to the mapping correlation whose logical correlation with the mapping relationship is greater than a preset correlation threshold.

[0065] S200, obtaining the trial-related data of the participants and predicting the changes in clinical symptoms; using the rule engine to make decisions and judgments on the changes in clinical symptoms and the participants' questionnaire responses, and generating a follow-up schedule.

[0066] Furthermore, in S200, the trial-related data of the participants is obtained to predict changes in clinical symptoms. A rule engine is used to make decisions and screen the changes in clinical symptoms and the participants' questionnaire responses to generate a follow-up schedule, specifically:

[0067] Obtain the trial strategy data and trial monitoring data that participants have already implemented from the knowledge base, and associate the benchmarked trial strategy data and trial monitoring data to obtain the trial dataset; use the trial dataset and the large model to predict clinical symptom changes in participants and generate clinical symptom feature vectors;

[0068] The natural text semantics of the participants' questionnaire responses are recognized and converted to obtain questionnaire response text vectors. The rule engine is used to make decisions and judgments on the clinical symptom feature vectors and questionnaire response text vectors to obtain the feasibility of each task in the follow-up task set for the participants, thereby generating a follow-up schedule.

[0069] See also Figure 4 Provides a trial follow-up management platform. By logging into the trial follow-up management platform (corresponding to Figure 4 (a)), enter the platform main interface (corresponding to Figure 4(b)-(e)). The trial follow-up management platform provides different functional modules such as intelligent outbound calls, intelligent reports, online customer service, voice SMS, questionnaires, knowledge base management, data visualization, education center, symptom management, and patient data center. Among them, the knowledge base corresponding to the knowledge base management stores the trial strategy data and trial monitoring data that the participants have implemented. Please refer to Figure 5 The two types of data can be obtained by accessing the knowledge base interface. Trial strategy data may include, but is not limited to, data on medication courses taken by participants during the clinical trial (e.g., medication type and quantity); and trial monitoring data may include, but is not limited to, test results from blood draws, CT scans, and MRI scans performed during follow-up visits. To ensure timely access to clinical trial results, participants are scheduled for follow-up visits after completing a course of treatment to capture data on changes in their physical condition under the effects of the corresponding medication. To this end, the two types of data are temporally correlated based on their respective acquisition times. For example, if the acquisition time of the trial strategy data is earlier than the acquisition time of the trial monitoring data, and the interval between the two acquisition times falls within a preset time interval, the corresponding trial strategy data and trial monitoring data are combined to form a trial dataset, thereby ensuring a causal relationship between the trial strategy data and the trial monitoring data in the trial dataset. The platform's artificial intelligence large model (such as the DeepSeek large model, etc.) is then used to process the experimental data set to predict the changes in clinical symptoms that will occur after the participants complete the current drug treatment course (such as changes in blood sugar, blood pressure, heart rate and other physical signs of the participants in clinical tests). The above clinical symptom changes are then processed through natural language recognition and converted into clinical symptom feature vectors for subsequent import into the rule engine for processing.

[0070] The physical condition of the participants during the drug clinical trial is also an important reference factor for formulating follow-up schedules. Figure 6By entering the questionnaire interface, the participant's questionnaire responses after each completed medication course are obtained. The questionnaire questions are pre-designed based on the specific medication course and can be associated with the participant's current medication course and its potential effects. The questionnaire responses are subjected to natural language text semantic recognition and conversion to generate questionnaire response text vectors, which can be directly imported into the rule engine for processing. The rule engine then uses the clinical symptom feature vectors and the questionnaire response text vectors to determine the feasibility of each task in the follow-up task set for the participant, thereby generating a follow-up schedule. The follow-up task set includes multiple testing tasks that need to be performed during the follow-up period. The rule engine uses the clinical symptom feature vectors and the questionnaire response text vectors to determine the reliability of each test task's execution result for the participant during the follow-up period. If the reliability of the execution result is greater than or equal to a preset reliability threshold, the corresponding test task is set in the follow-up schedule; otherwise, the corresponding test task is not set in the follow-up schedule. All test tasks in the follow-up task set are verified and checked one by one in the above manner to generate a follow-up schedule.

[0071] S300, based on the implementation results of the follow-up schedule, the information container is updated and the rule engine is reconfigured; the follow-up schedule is adjusted based on the clinical symptom changes and questionnaire responses for the next prediction using the reconfigured rule engine, and the follow-up schedule is updated and set.

[0072] Furthermore, in S300, based on the implementation result of the follow-up schedule, the information container is updated and the rule engine is reconfigured, specifically:

[0073] Perform semantic recognition and information extraction on the implementation results of the follow-up scheduling to obtain feedback data that is content-related to the basic data; use the feedback data to update the expired information elements in the information container;

[0074] According to the mapping relationship between the information elements updated in the information container, the matching decision rule set is reselected, thereby reconfiguring the rule engine for the information container.

[0075] Follow-up scheduling can be used to perform several testing tasks corresponding to a participant's completion of a drug treatment course during a drug clinical trial. To fully understand the drug's efficacy after taking the drug, the results of all testing tasks (i.e., test result data) are obtained after the participant completes the follow-up schedule. Natural language text semantic recognition and specific information extraction are performed on these results to obtain feedback data that is content-linked to the participant's initial clinical trial data, including basic data such as the patient's medical condition, physical condition, and trial protocol. This feedback data may include, but is not limited to, data on the severity of the participant's condition after completing the drug treatment course, physiological parameters such as blood pressure and blood sugar, and the effective duration of the drug's effect in the human body. After a participant completes a drug treatment course in a drug clinical trial, some of the information elements originally in the corresponding information container are considered to no longer accurately reflect the participant's medical condition and physical condition. These information elements are considered invalid. To ensure that the information container can provide valid information elements for the next follow-up schedule, new information elements are generated based on the feedback data and replaced with these new information elements. In the same way as described above, a mapping relationship between updated information elements of the information container is generated, and a matching decision rule set is reselected to reconfigure the rule engine for the information container to ensure that the reconfigured rule engine matches the requirements of the next random scheduling.

[0076] Furthermore, in S300, the reconfigured rule engine is used to adjust the follow-up schedule based on the next predicted clinical symptom changes and questionnaire responses, and the follow-up schedule is updated and set, specifically:

[0077] Obtain the clinical symptom feature vector corresponding to the next predicted clinical symptom change and the questionnaire response text vector corresponding to the next questionnaire response; use the reconfigured rule engine to make decisions based on the clinical symptom feature vector and the questionnaire response text vector, and adjust the follow-up tasks included in the follow-up schedule;

[0078] Adjust the content according to the follow-up tasks in the follow-up schedule, set up online notifications and visualize the schedule calendar.

[0079] After updating the participant's corresponding information container and reconfiguring the rule engine, the same process as step S200 is adopted, using the reconfigured rule engine to make decisions based on the clinical symptom feature vector corresponding to the next predicted clinical symptom change and the questionnaire response text vector corresponding to the next questionnaire response, and formulate the follow-up tasks corresponding to the next follow-up visit. This adjusts the follow-up tasks included in the follow-up schedule, making the adjusted follow-up schedule adaptable to the changes in the participant's condition and body after the previous drug treatment trial, thereby improving the match between the follow-up schedule and the participant. In addition, based on the follow-up task adjustment content of the follow-up schedule, online notification settings and scheduling calendar visualization are performed. Please refer to Figure 7-8 , respectively, through the follow-up scheduling notification interface and the scheduling calendar visualization interface, participants are provided with follow-up scheduling notification operations in the form of SMS and follow-up scheduling calendar in the form of reports, ensuring that participants fully and timely understand the specific task content of the next follow-up.

[0080] See also Figure 9 As shown, the present invention provides a dynamic scheduling system for trial follow-up based on a rule engine, which includes the following modules:

[0081] An information container generation module is used to obtain basic data of clinical trial participants and generate information containers for the participants;

[0082] A rule engine configuration module, configured to configure a rule engine for an information container according to information elements of the information container;

[0083] The prediction module is used to obtain trial-related data of participants and predict changes in clinical symptoms;

[0084] The schedule generation module is used to make decisions based on clinical symptom changes and participants' questionnaire responses using a rule engine to generate a follow-up schedule;

[0085] An updating module, used for updating the information container and reconfiguring the rule engine according to the implementation results of the follow-up scheduling;

[0086] The schedule adjustment and setting module is used to adjust the follow-up schedule based on the clinical symptom changes and questionnaire responses for the next prediction by using the reconfigured rule engine, and to update the follow-up schedule.

[0087] Furthermore, the information container generation module is used to obtain basic data of clinical trial participants and generate information containers for the participants, specifically:

[0088] Obtaining the clinical trial participant's medical condition data, physical status data, and trial protocol data, performing natural semantic recognition and conversion on the medical condition data, physical status data, and trial protocol data, and generating an information container for the participant; wherein the information container includes a medical condition information element, a physical status information element, and a trial protocol information element;

[0089] The rule engine configuration module is used to configure the rule engine for the information container according to the information element of the information container, specifically:

[0090] Based on the information elements of the information container, a mapping relationship between the disease information, physical status information, and test plan information is constructed; based on the mapping relationship, a matching decision rule set is selected to configure a rule engine for the information container; wherein the rule engine executes follow-up scheduling based on the decision rule set.

[0091] Furthermore, the prediction module is used to obtain the participant's trial-related data and predict changes in clinical symptoms, specifically:

[0092] Obtain the trial strategy data and trial monitoring data that participants have already implemented from the knowledge base, and associate the benchmarked trial strategy data and trial monitoring data to obtain the trial dataset; use the trial dataset and the large model to predict clinical symptom changes in participants and generate clinical symptom feature vectors;

[0093] The schedule generation module is used to use the rule engine to make decisions based on clinical symptom changes and participants' questionnaire responses, and generate a follow-up schedule. Specifically:

[0094] The natural text semantics of the participants' questionnaire responses are recognized and converted to obtain questionnaire response text vectors. The rule engine is used to make decisions and judgments on the clinical symptom feature vectors and questionnaire response text vectors to obtain the feasibility of each task in the follow-up task set for the participants, thereby generating a follow-up schedule.

[0095] Furthermore, the update module is used to update the information container and reconfigure the rule engine according to the implementation results of the follow-up schedule, specifically:

[0096] Perform semantic recognition and information extraction on the implementation results of the follow-up scheduling to obtain feedback data that is content-related to the basic data; use the feedback data to update the expired information elements in the information container;

[0097] Reselect the matching decision rule set based on the mapping relationship between the updated information elements of the information container, thereby reconfiguring the rule engine for the information container;

[0098] The schedule adjustment and setting module is used to adjust the follow-up schedule based on the next predicted clinical symptom changes and questionnaire responses using the reconfigured rule engine, and to update the follow-up schedule. Specifically:

[0099] Obtain the clinical symptom feature vector corresponding to the next predicted clinical symptom change and the questionnaire response text vector corresponding to the next questionnaire response; use the reconfigured rule engine to make decisions based on the clinical symptom feature vector and the questionnaire response text vector, and adjust the follow-up tasks included in the follow-up schedule;

[0100] Adjust the content according to the follow-up tasks in the follow-up schedule, set up online notifications and visualize the schedule calendar.

[0101] The operation and effects of the rule engine-based test follow-up dynamic scheduling system of the present invention are corresponding to and consistent with the above-mentioned rule engine-based test follow-up dynamic scheduling method, and the rule engine-based test follow-up dynamic scheduling system will not be repeated here.

[0102] In one embodiment of the present invention, the present invention further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the rule engine-based dynamic scheduling method for trial follow-up is implemented as described above.

[0103] In one embodiment of the present invention, the present invention also provides a computer device, which includes at least a memory and a processor, and a computer program is stored in the memory. When the computer program is executed by the processor, it implements the dynamic scheduling method for trial follow-up based on the rule engine as described above.

[0104] Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented by adding the necessary general-purpose hardware platform, or of course, by combining hardware and software. Based on this understanding, the essence of the above technical solution, or the portion that contributes to the prior art, can be embodied in the form of a computer product. The present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0105] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it, and other embodiments may also be used. Although the present invention has been described in detail with reference to the aforementioned embodiments, ordinary technicians in this field should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A dynamic scheduling method for test follow-up based on rule engine, characterized in that: The method comprises the following steps: S100: Obtain basic data of clinical trial participants and generate an information container of the participants; configure a rule engine for the information container based on information elements of the information container; S200: Acquire the participant's trial-related data and predict clinical symptom changes; use the rule engine to make decisions based on the clinical symptom changes and the participant's questionnaire responses to generate a follow-up schedule, specifically: Obtaining the trial strategy data and trial monitoring data already executed by the participant from the knowledge base, correlating and benchmarking the trial strategy data and the trial monitoring data to obtain a trial dataset; using the trial dataset and the large model, predicting clinical symptom changes occurring in the participant, and generating a clinical symptom feature vector; Performing natural text semantic recognition and conversion on the participant's questionnaire responses to obtain a questionnaire response text vector; utilizing the rule engine to make decisions and judgments on the clinical symptom feature vector and the questionnaire response text vector to obtain the feasibility of each task in the follow-up task set for the participant, thereby generating a follow-up schedule; S300: Update the information container and reconfigure the rule engine based on the implementation result of the follow-up schedule, specifically: Performing semantic recognition and information extraction on the implementation results of the follow-up schedule to obtain feedback data that is content-related to the basic data; using the feedback data, updating the invalid information elements of the information container; Reselecting a matching decision rule set based on the mapping relationship between the information elements updated in the information container, thereby reconfiguring a rule engine for the information container; The reconfigured rule engine is used to adjust the follow-up schedule based on the clinical symptom changes and questionnaire responses for the next prediction, and the follow-up schedule is updated and set, specifically: Obtaining a clinical symptom feature vector corresponding to the next predicted clinical symptom change and a questionnaire response text vector corresponding to the next questionnaire response; utilizing a reconfigured rule engine to make decisions and judgments on the clinical symptom feature vector and the questionnaire response text vector, and adjusting the follow-up tasks included in the follow-up schedule; Adjust the content according to the follow-up tasks in the follow-up schedule, set up online notifications and visualize the schedule calendar.

2. The method according to claim 1, characterized in that In S100, basic data of clinical trial participants are obtained to generate information containers of the participants; and a rule engine is configured for the information containers based on information elements of the information containers, specifically: Obtaining the clinical trial participant's medical condition data, physical status data, and trial protocol data, performing natural semantic recognition and conversion on the medical condition data, physical status data, and trial protocol data to generate an information container for the participant; wherein the information container includes a medical condition information element, a physical status information element, and a trial protocol information element; Based on the information elements of the information container, a mapping relationship between the disease information, the physical condition information, and the test plan information is constructed; based on the mapping relationship, a matching decision rule set is selected to configure a rule engine for the information container; wherein, the rule engine executes follow-up scheduling based on the decision rule set.

3. A dynamic scheduling system for trial follow-up based on rule engine, characterized by: The system includes the following modules: An information container generation module, configured to obtain basic data of clinical trial participants and generate information containers for the participants; A rule engine configuration module, configured to configure a rule engine for the information container according to the information element of the information container; The prediction module is used to obtain the trial-related data of the participants and predict changes in clinical symptoms, specifically: Obtaining the trial strategy data and trial monitoring data already executed by the participant from the knowledge base, correlating and benchmarking the trial strategy data and the trial monitoring data to obtain a trial dataset; using the trial dataset and the large model, predicting clinical symptom changes occurring in the participant, and generating a clinical symptom feature vector; The schedule generation module is used to use the rule engine to make decisions based on the clinical symptom changes and the participants' questionnaire responses to generate a follow-up schedule, specifically: Performing natural text semantic recognition and conversion on the participant's questionnaire responses to obtain a questionnaire response text vector; utilizing the rule engine to make decisions and judgments on the clinical symptom feature vector and the questionnaire response text vector to obtain the feasibility of each task in the follow-up task set for the participant, thereby generating a follow-up schedule; An updating module is used to update the information container and reconfigure the rule engine according to the implementation results of the follow-up schedule, specifically: Performing semantic recognition and information extraction on the implementation results of the follow-up schedule to obtain feedback data that is content-related to the basic data; using the feedback data, updating the invalid information elements of the information container; Reselecting a matching decision rule set based on the mapping relationship between the information elements updated in the information container, thereby reconfiguring a rule engine for the information container; The schedule adjustment and setting module is used to adjust the follow-up schedule based on the clinical symptom changes and questionnaire responses for the next prediction using the reconfigured rule engine, and to update and set the follow-up schedule, specifically: Obtaining a clinical symptom feature vector corresponding to the next predicted clinical symptom change and a questionnaire response text vector corresponding to the next questionnaire response; utilizing a reconfigured rule engine to make decisions and judgments on the clinical symptom feature vector and the questionnaire response text vector, and adjusting the follow-up tasks included in the follow-up schedule; Adjust the content according to the follow-up tasks in the follow-up schedule, set up online notifications and visualize the schedule calendar.

4. The system according to claim 3, characterized in that The information container generation module is used to obtain basic data of clinical trial participants and generate information containers of the participants, specifically: Obtaining the clinical trial participant's medical condition data, physical status data, and trial protocol data, performing natural semantic recognition and conversion on the medical condition data, physical status data, and trial protocol data to generate an information container for the participant; wherein the information container includes a medical condition information element, a physical status information element, and a trial protocol information element; The rule engine configuration module is used to configure a rule engine for the information container according to the information element of the information container, specifically: Based on the information elements of the information container, a mapping relationship between the disease information, the physical condition information, and the test plan information is constructed; based on the mapping relationship, a matching decision rule set is selected to configure a rule engine for the information container; wherein, the rule engine executes follow-up scheduling based on the decision rule set.

5. A computer-readable storage medium having a computer program stored thereon, characterized in that: When executed by a processor, the computer program implements the rule engine-based dynamic scheduling method for trial follow-up according to any one of claims 1 to 2.

Images