A method and system for dynamic visualization of health events of clinical trial patients

Abstract

The present invention relates to the technical field of clinical trial data analysis, and in particular to a method for dynamic visualization of health events of clinical trial patients, which collects and organizes health indicator data to construct a data set by acquiring clinical trial information and matching it with monitoring data requirements. The historical health event data is further associated to generate a measured three-dimensional visual model related to a common timeline, and the models in continuous time are fused to generate a dynamic three-dimensional visual model. The model has interactive functions, allowing the analysis of changing trends of health events through timeline sliders and filter condition controls. The final generated visualization report contains a time series graph of individual health events, providing an intuitive and dynamic health event assessment tool for clinical trials. This application has the effect of improving the efficiency and quality of clinical research.

Description

Technical Field

[0001] The present invention relates to the technical field of clinical trial data analysis, and in particular to a method and system for dynamically visualizing health events of clinical trial patients. Background Art

[0002] In modern medical research, clinical trials are a crucial tool for evaluating drug efficacy, medical device safety, and the effectiveness of treatments. As clinical trials grow in complexity, the number of patients and the volume of health data involved are also surging. Effectively managing and analyzing this data has become a critical issue.

[0003] Traditional data recording and display methods (such as static tables and simple charts) are no longer able to meet researchers' needs for in-depth data mining and dynamic monitoring. While existing technologies can achieve basic clinical data visualization, they have many limitations. For example, they only provide single-dimensional data display and have difficulty integrating multimodal health indicators. This restricts exploratory data analysis and makes it difficult to reveal the dynamic evolution of patient health events, thereby reducing the quality of clinical trial analysis. Summary of the Invention

[0004] In order to improve the analysis quality of clinical trials, the present application provides a method and system for dynamically visualizing health events of clinical trial patients.

[0005] In the first aspect, the above-mentioned invention object of the present application is achieved through the following technical solutions:

[0006] A method for dynamically visualizing health events of clinical trial patients, comprising the steps of:

[0007] Obtain clinical trial information and input it into a pre-set clinical trial requirements database to match corresponding monitoring data requirements, where the monitoring data requirements are used to indicate the health indicator data that needs to be collected in the clinical trial;

[0008] Acquire health indicator data of all patients in the clinical trial based on the monitoring data requirements, and preprocess and organize the health indicator data to construct a patient health indicator dataset, wherein the health indicator data includes laboratory test result data, vital sign data, and imaging test data;

[0009] Acquire historical health event data of the patient, and associate the health indicator data with the health event data based on a preset common timeline to construct a health event-related data set;

[0010] Performing a comprehensive analysis on the corresponding health indicator data in the health event-related data set to generate a measured three-dimensional visual model associated with the common time axis;

[0011] Continuously fusing the measured three-dimensional visual models over continuous time based on a common timeline to generate a dynamic three-dimensional visual model, wherein interactive functions are provided in the dynamic visualization display, allowing the dynamic three-dimensional visual model to be analyzed through a timeline slider and filter condition controls to observe the changing trends and patterns of patient health events over time;

[0012] A visualization report of the clinical trial is generated based on the dynamic three-dimensional visual model, and the visualization report also includes a time series diagram of individual health events of the patient.

[0013] By adopting the above technical solution, comprehensive, dynamic, and intuitive visual management of clinical trial patient health events can be achieved. This process starts with obtaining clinical trial information and matching it with monitoring data needs. Through systematic data collection, organization, and analysis, it ultimately generates dynamic three-dimensional visualization models and related reports with high interactivity and clinical value. Compared with existing technologies, the advantages of this method lie in its integration and dynamism. Existing technologies often only provide static data display or limited interactive functions. However, this method uses three-dimensional visualization and dynamic models to enable researchers to more intuitively observe the changing process of patient health events. In addition, this method emphasizes the integration of multi-dimensional data, including laboratory test results, vital signs, and imaging test data, which makes the generated visualization reports more comprehensive and in-depth. Existing clinical trial data visualization tools can usually only display a single type of data. However, this method provides a more complete and three-dimensional view of patient health status by comprehensively analyzing multiple health indicators, which helps to improve the efficiency and quality of clinical research and better support clinical decision-making and research discoveries.

[0014] In a preferred example, the present application may be further configured to: after comprehensively analyzing the corresponding health indicator data in the health event-related data set to generate a measured three-dimensional visual model associated with the common timeline, include the following steps:

[0015] Obtaining all corresponding index baseline value intervals in the clinical trial, and comparing each of the monitored health index data with the corresponding index baseline value intervals. If the health index data is not within the index baseline value interval, the corresponding health index data is marked as abnormal;

[0016] The health indicator data with abnormal marks are marked on the measured three-dimensional visual model in a highlighted form.

[0017] By adopting the above-mentioned technical solution, abnormal health indicator data is marked in a highlighted form, allowing researchers to quickly locate abnormal situations when viewing complex three-dimensional visualization models, thereby improving the efficiency and accuracy of data interpretation. This intuitive abnormality prompt function helps researchers to promptly identify potential health problems, provide strong support for clinical decision-making, and further improve the monitoring level and research quality of clinical trials.

[0018] In a preferred example, the present application may be further configured to: after obtaining the patient's historical health event data and associating the health indicator data with the health event data based on a preset common timeline to construct a health event-related dataset, the following steps are included:

[0019] Obtain clinical medical records corresponding to each health event;

[0020] Recording health indicator data before and after a health event based on a preset time length, and associating the health event, clinical medical records, and health indicator data one-to-one based on the common timeline to construct a nursing effect dataset;

[0021] The preset nursing effect analysis model analyzes the nursing effect data set based on a machine self-learning algorithm to generate corresponding effect quantitative data;

[0022] Predicting the effect of the current health event based on the effect quantification data to generate prediction indicator data;

[0023] The prediction index data are comprehensively analyzed to generate a three-dimensional visual model for effect prediction.

[0024] By adopting the above technical solutions, the three-dimensional visual model of effect prediction can intuitively display the effect prediction of nursing measures, helping researchers to better understand and communicate the potential benefits of nursing plans. This visualization tool not only improves the readability and usability of data, but also provides strong support for the reporting and presentation of clinical trials, helping to promote the dissemination and application of research results. By combining predictive indicator data with three-dimensional models of organs, researchers can more comprehensively evaluate the impact of nursing measures on the overall health status of patients, thereby optimizing treatment plans and improving patients' treatment effects and quality of life.

[0025] In a preferred example, the present application may be further configured as follows: after the step of predicting the effect of the current health event based on the effect quantification data to generate prediction index data, the following steps are included:

[0026] Classify patient health events in the same clinical trial and generate corresponding event labels;

[0027] Obtaining quantitative effect data of different patients within the same event label, and comparing the quantitative effect data of different patients with each other to generate effect difference data;

[0028] Obtaining vital sign data and basic data of different patients, and comparing the vital sign data and basic data of different patients to generate vital sign difference data and basic characteristic difference data;

[0029] Associating the physical sign difference data, basic feature difference data and effect difference data to construct a patient effect difference data set;

[0030] The preset effect difference analysis model analyzes the patient effect difference data set based on a preset machine self-learning algorithm to generate an effect deviation factor;

[0031] The prediction index data is corrected based on the effect deviation factor so that the prediction result is more consistent with the actual situation of the patient himself.

[0032] By adopting the above technical solutions, comparing the basic data and vital signs data of different patients can help discover individual differences that may affect the nursing effect, and provide clues for analyzing the reasons for the difference in effects. The effect deviation factor can improve the accuracy and personalization of the prediction, making the prediction results more in line with the patient's actual situation, and helping to provide more accurate care and treatment plans.

[0033] In a preferred example, the present application may be further configured to: after obtaining the patient's historical health event data and associating the health indicator data with the health event data based on a preset common timeline to construct a health event-related dataset, the following steps are included:

[0034] The preset health event warning model analyzes the health event related data set based on a machine self-learning algorithm to generate a health event warning trend, wherein the health event warning trend includes corresponding indicator features;

[0035] When the changing trend of the health indicator data is similar to the indicator characteristics corresponding to the health event warning trend, a warning signal is issued.

[0036] By adopting the above technical solutions, through real-time monitoring of health indicator data and comparing it with warning trends, early warning signals can be issued in a timely manner to remind researchers to pay attention to high-risk patients. This timely early warning mechanism helps to quickly respond to potential health crises and improve the safety of clinical trials and patient treatment effects.

[0037] In a preferred example, the present application may be further configured to include the following steps after the step of issuing a warning signal when the changing trend of the health indicator data is similar to the indicator characteristics corresponding to the health event warning trend:

[0038] Predict the patient's health indicator data based on a dynamic time series analysis model and generate predicted health indicator trends;

[0039] A comprehensive analysis is performed on the data features of each health indicator in the predicted health indicator trend to generate a dynamic prediction three-dimensional visual model, and the prediction three-dimensional visual model is used to predict the occurrence of health events in the time period corresponding to the predicted health indicator trend.

[0040] By adopting the above technical solutions, the dynamic predictive three-dimensional visual model can visualize complex predictive data, allowing researchers to quickly understand the changing trends of patients' health status. This visualization tool not only improves the readability and usability of data, but also provides strong support for the reporting and presentation of clinical trials, helping to promote the dissemination and application of research results. In this way, researchers can better predict the occurrence of health events, formulate intervention measures in advance, and improve patients' treatment effects and quality of life.

[0041] In a preferred example, the present application may be further configured to include the following steps after comprehensively analyzing the data features of each health indicator in the predicted health indicator trend to generate a dynamic predictive three-dimensional visual model:

[0042] If the predicted three-dimensional visual model indicates a health event, the predicted three-dimensional visual model at that time point is marked as a risk event;

[0043] Simulating medical care behaviors on the predicted three-dimensional visual model with risk markers to generate a three-dimensional visual model of simulated effects;

[0044] Performing effect prediction on the three-dimensional visual model of the simulated effect based on the corresponding quantitative effect data to generate prediction index data corresponding to the simulated medical care behavior;

[0045] Generate a medical simulation report based on simulated medical behavior, three-dimensional visual model of simulation effect and corresponding prediction indicator data.

[0046] By adopting the above technical solutions, a medical simulation report containing the simulation process, effect evaluation and suggestions is generated for researchers and clinicians to discuss and make decisions. The medical simulation report summarizes the key information and expected effects of the simulation intervention, which helps all parties quickly understand the potential benefits of the simulation intervention, promote multidisciplinary team collaboration, jointly optimize patient treatment plans, and improve the quality and efficiency of medical services.

[0047] Secondly, the above-mentioned invention objectives of this application are achieved through the following technical solutions:

[0048] A device for dynamically visualizing health events of clinical trial patients, the device comprising: a monitoring data requirement matching unit for acquiring clinical trial information and inputting it into a pre-set clinical trial requirement database to match corresponding monitoring data requirements, the monitoring data requirements being used to indicate health indicator data that need to be collected in the clinical trial;

[0049] a health indicator data acquisition unit, configured to acquire health indicator data of all patients in the clinical trial based on the monitoring data requirements, and preprocess and organize the health indicator data to construct a patient health indicator dataset, wherein the health indicator data includes laboratory test result data, vital sign data, and imaging test data;

[0050] A historical health event data acquisition unit is used to acquire the patient's historical health event data and associate the health indicator data with the health event data based on a preset common timeline to construct a health event related data set;

[0051] a measured three-dimensional visual model generating unit, configured to comprehensively analyze the corresponding health indicator data in the health event-related data set to generate a measured three-dimensional visual model associated with the common time axis;

[0052] A dynamic three-dimensional visual model generating unit, configured to continuously fuse the measured three-dimensional visual models in a continuous time based on a common time axis to generate a dynamic three-dimensional visual model;

[0053] A clinical trial visualization report generating unit is used to generate a visualization report of the clinical trial based on the dynamic three-dimensional visual model, wherein the visualization report also includes a time series diagram of individual health events of the patient.

[0054] Thirdly, the above-mentioned purpose of this application is achieved through the following technical solutions:

[0055] An electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the above-mentioned method for dynamically visualizing health events of clinical trial patients are implemented.

[0056] Fourthly, the above-mentioned purpose of the present application is achieved through the following technical solutions:

[0057] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the above-mentioned method for dynamically visualizing health events of clinical trial patients.

[0058] In summary, this application includes at least one of the following beneficial technical effects:

[0059] 1. It can achieve comprehensive, dynamic, and intuitive visual management of clinical trial patient health events. This process starts with obtaining clinical trial information and matching it with monitoring data needs. Through systematic data collection, organization, and analysis, it ultimately generates dynamic three-dimensional visualization models and related reports with high interactivity and clinical value. Compared with existing technologies, the advantages of this method lie in its integration and dynamism. Existing technologies often only provide static data display or limited interactive functions. This method, through three-dimensional visualization and dynamic models, enables researchers to more intuitively observe the changing process of patient health events. In addition, this method emphasizes the integration of multi-dimensional data, including laboratory test results, vital signs, and imaging test data, which makes the generated visualization reports more comprehensive and in-depth. Existing clinical trial data visualization tools can usually only display a single type of data. This method provides a more complete and three-dimensional view of patient health status by comprehensively analyzing multiple health indicators, which helps to improve the efficiency and quality of clinical research and better support clinical decision-making and research discoveries.

[0060] 2. Highlighting abnormal health indicator data allows researchers to quickly locate anomalies when viewing complex 3D visualization models, improving the efficiency and accuracy of data interpretation. This intuitive abnormality prompt function helps researchers promptly identify potential health issues, providing strong support for clinical decision-making and further improving the monitoring level and research quality of clinical trials.

[0061] 3. Dynamic predictive 3D visualization models can visualize complex predictive data, enabling researchers to quickly understand changing trends in patients' health status. This visualization tool not only improves data readability and usability, but also provides strong support for clinical trial reporting and presentation, helping to promote the dissemination and application of research results. In this way, researchers can better predict the occurrence of health events, formulate intervention measures in advance, and improve patients' treatment outcomes and quality of life.

[0062] 4. Generate a medical simulation report that includes the simulation process, effect evaluation, and recommendations for researchers and clinicians to discuss and make decisions. The medical simulation report summarizes the key information and expected effects of the simulation intervention, helping all parties to quickly understand the potential benefits of the simulation intervention, promote multidisciplinary team collaboration, jointly optimize patient treatment plans, and improve the quality and efficiency of medical services. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] Figure 1This is a flow chart of a method for dynamically visualizing health events of clinical trial patients in one embodiment of the present application;

[0064] Figure 2 This is a principle block diagram of a device for dynamically visualizing health events of clinical trial patients in one embodiment of the present application;

[0065] Figure 3 It is a schematic diagram of an electronic device in an embodiment of the present application.

[0066] Figure Number:

[0067] 1. Monitoring data demand matching unit; 2. Health indicator data acquisition unit; 3. Historical health event data acquisition unit; 4. Measured 3D visual model generation unit; 5. Dynamic 3D visual model generation unit; 6. Clinical trial visualization report generation unit. DETAILED DESCRIPTION

[0068] The present application is further described in detail below with reference to the accompanying drawings.

[0069] In one embodiment, if Figure 1 As shown, the present application discloses a method for dynamically visualizing health events of clinical trial patients, which specifically includes the following steps:

[0070] S10: Acquire clinical trial information and input it into a pre-set clinical trial requirement database to match corresponding monitoring data requirements, where the monitoring data requirements are used to indicate the health indicator data that need to be collected in the clinical trial;

[0071] Specifically, consider a clinical trial designed to evaluate the effectiveness of a new antihypertensive drug in patients with hypertension and investigate its impact on the liver. Researchers enter trial information into a requirements database, including trial name, purpose, and duration. The system then matches this information with corresponding monitoring data requirements. For example, the health indicators to be collected include the patient's systolic and diastolic blood pressure, heart rate, blood and urine routine, liver and kidney function indicators, and so on.

[0072] Through standardized collection, we can avoid missing key indicators or over-monitoring irrelevant indicators, so that safety assessments meet regulatory requirements while reducing the monitoring burden and improving monitoring efficiency.

[0073] S20: Acquire health indicator data of all patients in the clinical trial based on the monitoring data requirements, and preprocess and organize the health indicator data to construct a patient health indicator dataset;

[0074] In the embodiments of the present application, health indicator data includes laboratory test results, vital signs, and imaging data. Similarly, using the aforementioned antihypertensive drug clinical trial as an example, researchers collected health indicator data from all patients participating in the trial. These data included vital signs such as systolic blood pressure, diastolic blood pressure, and heart rate recorded at each follow-up visit, laboratory test results such as blood and urine routine before and after the trial, and imaging data such as cardiac ultrasound. These data were preprocessed, such as missing value filling, outlier detection and correction, and data standardization, and then organized into a structured patient health indicator dataset.

[0075] It should be noted that in the embodiment of the present application, the unstructured doctor's notes are quantified into level / degree classifications that can be used for data processing (such as "moderate edema" is coded as CTCAE grade 2 edema), and the health indicator data are comprehensively preprocessed and sorted. This can improve the quality and availability of the data, ensure the accuracy and reliability of the subsequently constructed health event-related data sets, and provide high-quality data support for dynamic visualization analysis.

[0076] S30: Acquire historical health event data of the patient, and associate the health indicator data with the health event data based on a preset common timeline to construct a health event-related data set;

[0077] Specifically, continuing with the example of a clinical trial of antihypertensive drugs, researchers obtained the medical history data of patients during the trial, and with the trial timeline as the common timeline, associated the patients' health indicator data (such as blood pressure measurements at different time points) with historical health event data. For example, the blood pressure values ​​of patients at the beginning of the trial, at 1 month, 2 months, 3 months, and other time points of the trial are recorded, as well as whether cardiovascular-related health events (such as angina attacks, dizziness, etc.) occurred at these time points. In this application, whenever a patient has a health event, it needs to be recorded, rather than revisiting at specific time intervals, so as to avoid missing registration of health events and better achieve full-cycle monitoring of clinical trial patients.

[0078] By integrating a patient's health indicator data with historical health event data and linking them based on a common timeline, a time series dataset can be constructed that comprehensively reflects the patient's health status. This helps researchers gain a deeper understanding of the relationship between a patient's health events and health indicators, providing a rich data context for subsequent dynamic visualization analysis.

[0079] S40: performing a comprehensive analysis on the corresponding health indicator data in the health event related data set to generate a measured three-dimensional visual model associated with the common time axis;

[0080] Specifically, professional data analysis software is used to perform a comprehensive analysis of the health indicator data in the health event-related data set. In the embodiment of the present application, Mimics software and Simpleware are used to generate a three-dimensional visual model (other software such as MITK software, InVesalius software, etc. can be used in other embodiments), and the MarchingCubes algorithm is used to generate a surface mesh model to retain internal texture features.

[0081] Suppose, in a clinical trial, researchers are concerned about a patient's lung health. They use CT scans to obtain a series of chest image data. Using specialized medical image processing software (such as Mimics or 3DSlicer), these CT images are reconstructed into a 3D visual model of the lungs. During this process, researchers can analyze the CT images at different time points to observe changes in lung structure, such as the extent and morphological changes of lung lesions. This 3D visual model clearly displays the lung's anatomy and can be linked to a common timeline to observe changes in lung health over time.

[0082] By generating a three-dimensional visual model of organs associated with a common timeline, researchers can more intuitively observe the morphological changes and health status of organs. This three-dimensional visualization method helps improve the accuracy of assessing patient health events in clinical trials, and provides stronger support for disease diagnosis, evaluation of treatment effects, and medical research.

[0083] S50: Continuously fusing the measured three-dimensional visual models in continuous time based on a common timeline to generate a dynamic three-dimensional visual model, wherein an interactive function is provided in the dynamic visualization display, allowing the dynamic three-dimensional visual model to be analyzed through a timeline slider and a filter condition control to observe the changing trends and patterns of the patient's health events over time;

[0084] Specifically, the measured three-dimensional visual models generated above are continuously fused in chronological order to form a dynamic three-dimensional visualization. In the visualization interface, researchers can observe the blood pressure change trend of patients in different time periods by dragging the timeline slider. At the same time, the interface also provides filtering condition controls, allowing researchers to filter according to the patient's age, gender, underlying diseases and other characteristics to observe the change pattern of health events in specific subgroups of patients. For example, researchers can choose to only view the blood pressure changes of male patients over the age of 60, or filter out the blood pressure fluctuation patterns of patients with a history of diabetes;

[0085] The generation of dynamic 3D visual models and the provision of interactive features allow researchers to more flexibly explore and analyze patient health event data. By adjusting the timeline slider and filtering conditions, researchers can delve deeper into the underlying patterns in the data, such as observing the significant efficacy of a treatment regimen in a specific patient population within a specific time period, or discovering the high incidence of certain adverse reactions under specific conditions. This interactive, dynamic visual analysis helps improve the efficiency and accuracy of clinical trial data interpretation, providing deeper insights for clinical research.

[0086] S60: generating a visualization report of the clinical trial based on the dynamic three-dimensional visual model, wherein the visualization report further includes a time series diagram of individual health events of the patient;

[0087] Specifically, based on the dynamic three-dimensional visual model, a detailed clinical trial visualization report is generated. In addition to the three-dimensional visualization of the blood pressure change trend of the entire patient, the visualization report also generates a time series diagram of individual health events for each patient. These time series diagrams can be displayed in the form of simple two-dimensional line graphs to show the changes in systolic and diastolic blood pressure of each patient during the trial, marking the time points of key health events (such as the time of adverse reaction occurrence, the time of treatment plan adjustment, etc.). For example, the time series diagram of patient A shows that his systolic blood pressure dropped significantly in the second month of the trial, and it is also noted that the researcher adjusted the drug dosage at this time. In addition, the report also provides summary statistical charts, such as a comparison chart of the average blood pressure change trend of different treatment groups, a statistical chart of adverse reaction incidence, etc.

[0088] The generated visual reports present complex clinical trial data in an intuitive and easy-to-understand manner, helping researchers quickly understand the overall trial status and changes in individual patients' health events. They also provide strong support for clinical trial summaries and reporting. The individual health event time series graphs in the reports help researchers focus on each patient's unique responses, facilitating personalized medical decision-making. Meanwhile, summary statistical charts provide a clear basis for evaluating the effectiveness and safety of treatment options.

[0089] In summary, the method described in steps S10 to S60 enables comprehensive, dynamic, and intuitive visualization of clinical trial patient health events. This process begins with obtaining clinical trial information and matching it with monitoring data requirements. Through systematic data collection, organization, and analysis, it ultimately generates a dynamic three-dimensional visualization model and related reports with high interactivity and clinical value. Compared with existing technologies, the advantages of this method lie in its integration and dynamism. Existing technologies often only provide static data display or limited interactive functions. However, this method, through three-dimensional visualization and dynamic models, enables researchers to more intuitively observe the changing process of patient health events. In addition, this method emphasizes the integration of multi-dimensional data, including laboratory test results, vital signs, and imaging test data, which makes the generated visualization reports more comprehensive and in-depth. Existing clinical trial data visualization tools are generally only able to display a single type of data. However, this method provides a more complete and three-dimensional view of patient health status by comprehensively analyzing multiple health indicators, which helps improve the efficiency and quality of clinical research and better support clinical decision-making and research discoveries.

[0090] After the step of S30: acquiring the patient's historical health event data and associating the health indicator data with the health event data based on a preset common timeline to construct a health event related data set, the following steps are included:

[0091] S31: Obtain clinical medical records corresponding to each health event;

[0092] Specifically, in a clinical trial, Patient A experienced an adverse reaction (such as dizziness) during the first month of the trial. Researchers obtained the clinical medical records corresponding to this health event, including detailed information such as the time of consultation, symptom description, diagnosis, and treatment measures (such as drug dosage adjustment).

[0093] S32: Recording health indicator data before and after the health event based on a preset time length, and associating the health event, clinical medical records, and health indicator data one-to-one based on the common time axis to construct a nursing effect data set;

[0094] Specifically, using Patient A as an example, the researchers set a predefined timeframe (e.g., one month before and one month after the health event). They recorded the patient's health indicator data, including vital signs like blood pressure and heart rate, as well as laboratory test results, for the month before and one month after the health event. Then, based on a common timeline, they established a one-to-one correspondence between the health event (dizziness episode), clinical care records (the timing and content of medication dosage adjustments), and health indicator data to construct a nursing effectiveness dataset.

[0095] By recording health indicator data before and after health events and linking them to clinical medical records, we can comprehensively reflect the impact of nursing interventions on patients' health status. The construction of this dataset helps researchers deeply analyze the effectiveness of nursing measures and provide data support for optimizing nursing plans.

[0096] S33: The preset nursing effect analysis model analyzes the nursing effect data set based on a machine self-learning algorithm to generate corresponding effect quantitative data;

[0097] Specifically, the nursing effectiveness analysis model uses machine learning algorithms (such as random forests or neural networks) to analyze nursing effectiveness datasets. For example, the model analysis found that after adjusting the medication dosage, Patient A's blood pressure gradually stabilized over the following month, and the frequency of dizziness symptoms decreased significantly. Based on this, the model generates quantitative data on the effectiveness, such as a 20% increase in the blood pressure stability index and a 30% decrease in the frequency of dizziness symptoms.

[0098] Analyzing nursing effectiveness datasets through machine learning algorithms can quantitatively evaluate the effects of nursing measures and provide researchers with objective quantitative indicators. This helps to more accurately assess the effectiveness of nursing plans.

[0099] S34: Predicting the effect of the current health event based on the effect quantification data to generate prediction indicator data;

[0100] Specifically, based on patient A's quantitative treatment outcome data (e.g., a 20% increase in blood pressure stability index and a 30% decrease in dizziness frequency), the nursing effectiveness analysis model predicts the effectiveness of the current health event (dizziness episode). The model predicts that, with continued treatment, dizziness frequency will decrease by a further 20% and blood pressure stability index will increase by another 15% over the next month. These predictions serve as predictive indicators, providing researchers with a reference for future health trends.

[0101] By predicting the effects of treatment, researchers can understand in advance the potential future health changes that may result from nursing interventions, allowing them to better plan and adjust care plans. This predictive capability helps optimize the design and execution of clinical trials, improving patient outcomes and quality of life.

[0102] S35: performing a comprehensive analysis on the prediction index data to generate a three-dimensional visual model for effect prediction;

[0103] Specifically, continuing to take patient A as an example, the nursing effect analysis model predicts that if the current treatment plan is continued, the frequency of dizziness symptoms will be further reduced by 20% and the blood pressure stability index will be increased by another 15% in the next month. The researchers combined these predictive indicator data with the previously generated three-dimensional visual model of the lungs to generate a three-dimensional visual model for effect prediction. In this model, the changing trend of the lung structure and the changes in predicted health indicators (such as the blood pressure stability index and the frequency of dizziness symptoms) can be intuitively displayed. For example, the model can use color coding or transparency changes to represent changes in health status in different areas, allowing researchers to more intuitively observe the impact of nursing measures on future health status;

[0104] The three-dimensional visual model of effect prediction can intuitively display the predicted effects of nursing measures, helping researchers better understand and communicate the potential benefits of nursing plans. This visualization tool not only improves the readability and usability of data, but also provides strong support for the reporting and presentation of clinical trials, helping to promote the dissemination and application of research results. By combining predictive indicator data with the three-dimensional model of the organ, researchers can more comprehensively evaluate the impact of nursing measures on the patient's overall health status, thereby optimizing treatment plans and improving patients' treatment effects and quality of life.

[0105] After the step S34: predicting the effect of the current health event based on the effect quantification data to generate prediction index data, the following steps are included:

[0106] S341: Classify patient health events in the same clinical trial and generate corresponding event labels;

[0107] S342: Obtaining quantitative effect data of different patients within the same event tag, and comparing the quantitative effect data of different patients to generate effect difference data;

[0108] S343: Obtain vital sign data and basic data of different patients, and compare the vital sign data and basic data of different patients to generate vital sign difference data and basic characteristic difference data;

[0109] S344: Correlating the physical sign difference data, basic feature difference data, and effect difference data to construct a patient effect difference data set;

[0110] S345: Analyzing the patient effect difference dataset using a preset effect difference analysis model based on a preset machine self-learning algorithm to generate an effect deviation factor;

[0111] S346: Correcting the prediction index data based on the effect deviation factor so that the prediction result is more consistent with the actual situation of the patient.

[0112] For example, regarding steps S341-S346, in a clinical trial, possible health events that may occur in patients are categorized into categories, such as "dizziness," "nausea," and "abnormal blood pressure." Assuming both Patient A and Patient B experience a "dizziness" event, the event is labeled "dizziness." Under the "dizziness" event label, the quantitative effect data for Patient A and Patient B are compared. Assuming that Patient A's dizziness frequency decreases by 30% after care, while Patient B's decrease is 20%, the difference in effect is 10%.

[0113] Obtain basic data (such as age, gender, and weight) and vital sign data (such as blood pressure and heart rate) for Patient A and Patient B. Assuming Patient A's blood pressure is 120 / 80 mmHg and Patient B's is 130 / 85 mmHg, the difference in vital signs is a blood pressure difference of 10 / 5 mmHg. Associate the vital sign difference data (blood pressure difference of 10 / 5 mmHg), basic characteristic difference data (age difference of 5 years), and effect difference data (dizziness frequency reduction difference of 10%) between Patient A and Patient B to form a comprehensive data set. Comparing basic data and vital sign data across different patients can help identify individual differences that may affect nursing outcomes and provide clues for analyzing the causes of these differences in outcomes.

[0114] Further analysis of the patient effect difference dataset was performed using machine learning algorithms (such as random forests or logistic regression). The model identified blood pressure and age differences as key factors influencing the effectiveness of dizziness care and generated corresponding weight coefficients as effect deviation factors. Based on these effect deviation factors, the predicted reduction in dizziness frequency for Patient B was corrected. Assuming the original predicted value was a 20% reduction, after correction for the effect deviation factors for blood pressure and age, the predicted value was adjusted to a 25% reduction. Using the effect deviation factors to modify the predictive indicator data improves the accuracy and personalization of the predictions, making the predictions more tailored to the patient's actual situation and helping to provide more precise care and treatment plans.

[0115] After the step S40 of comprehensively analyzing the corresponding health indicator data in the health event-related data set to generate a measured three-dimensional visual model associated with the common time axis, the following steps are included:

[0116] S41: Obtaining all corresponding index baseline value intervals in the clinical trial, and comparing each monitored health index data with the corresponding index baseline value interval. If the health index data is not within the index baseline value interval, marking the corresponding health index data as abnormal;

[0117] For example, in a clinical trial of a blood pressure-lowering drug, researchers set baseline values ​​for patients' systolic and diastolic blood pressure (e.g., 90-140 mmHg for systolic and 60-90 mmHg for diastolic blood pressure) before the trial. While monitoring the health indicator data, they discovered that Patient B's systolic blood pressure was 150 mmHg in the second month of the trial, exceeding the established normal range. The system then flagged this systolic blood pressure as abnormal.

[0118] It should be noted that in the embodiment of the present application, step S41 uses a dynamic threshold calibration algorithm to set the upper limit of the floating threshold for the indicator reference value interval.

[0119] By comparing monitored health indicator data with pre-set baseline values, abnormalities in patients' health indicators can be detected promptly, providing timely warnings for subsequent clinical intervention. This abnormality flagging function helps researchers quickly identify patients at risk of health problems, enabling them to take timely action and improve the safety and effectiveness of clinical trials.

[0120] S42: marking the measured three-dimensional visual model with the health indicator data having abnormal marks in a highlighted form;

[0121] Specifically, in the dynamic 3D visualization model, all monitored health indicator data points are represented by dots of different colors. Data points within the normal range are displayed in green, while data points marked as abnormal (such as Patient B's systolic blood pressure of 150 mmHg) are highlighted in red. When viewing the 3D visualization model, researchers can immediately identify abnormal data points by the color difference and focus on these abnormalities. It should be noted that in this embodiment of the application, all 3D visualization models use highlighting to mark abnormalities.

[0122] It should be noted that the marking of the structural layer and the marking of the functional layer in this application are differentiated. For the marking of the structural layer: the abnormal area flashes with a red pulse light effect; for the marking of the functional layer: the abnormal area is displayed as a high-temperature color spot diffuse halo (the halo radius is proportional to the degree of abnormality).

[0123] By highlighting abnormal health indicator data, researchers can quickly locate abnormalities when viewing complex three-dimensional visualization models, improving the efficiency and accuracy of data interpretation. This intuitive abnormality prompt function helps researchers to promptly identify potential health problems, provide strong support for clinical decision-making, and further improve the monitoring level and research quality of clinical trials.

[0124] After the step of S30: obtaining the patient's historical health event data and associating the health indicator data with the health event data based on a preset common timeline to construct a health event related data set, the following steps are also included:

[0125] S36: The preset health event warning model analyzes the health event related data set based on a machine self-learning algorithm to generate a health event warning trend, wherein the health event warning trend includes corresponding indicator features;

[0126] Specifically, in clinical trials, researchers used machine learning algorithms (such as logistic regression or support vector machines) to analyze health event-related datasets. For example, model analysis found that when a patient's systolic blood pressure exceeded 140 mmHg and their heart rate exceeded 90 bpm for three consecutive days, the risk of a serious cardiovascular event increased significantly. These characteristics were defined as indicators of health event warning trends. By analyzing health event-related datasets using machine learning algorithms, potential health risk characteristics can be identified, providing a scientific basis for early warning. This helps to promptly identify high-risk patients, take preventive measures in advance, and reduce the incidence of serious health events.

[0127] S37: When the change trend of the health indicator data is similar to the indicator characteristics corresponding to the health event warning trend, issuing a warning signal;

[0128] Specifically, during a clinical trial, patient C's health indicator data showed that their systolic blood pressure had exceeded 140 mmHg for three consecutive days, and their heart rate had consistently exceeded 90 bpm. The health event warning model detected that this trend matched the pre-set health event warning trend and immediately triggered a warning signal, notifying researchers to conduct further evaluation and intervention for patient C. By monitoring health indicator data in real time and comparing it with the warning trend, timely warning signals can be issued, alerting researchers to focus on high-risk patients. This timely warning mechanism helps quickly respond to potential health crises, improving the safety of clinical trials and patient treatment outcomes.

[0129] After the step S37 of issuing a warning signal when the changing trend of the health indicator data is similar to the indicator characteristics corresponding to the health event warning trend, the following steps are included:

[0130] S371: Predicting the patient's health indicator data based on a dynamic time series analysis model to generate a predicted health indicator trend;

[0131] Suppose that in a clinical trial, patient D's health indicator data shows an upward trend in systolic blood pressure over the past week, with the following values: 130 mmHg, 135 mmHg, 140 mmHg, and 145 mmHg. Using a dynamic time series analysis model (such as an ARIMA or LSTM neural network), predict the systolic blood pressure trend over the next three days to be 150 mmHg, 155 mmHg, and 160 mmHg.

[0132] S372: Comprehensively analyzing the data features of each health indicator in the predicted health indicator trend to generate a dynamic prediction three-dimensional visual model, wherein the prediction three-dimensional visual model is used to predict the occurrence of health events within a time period corresponding to the predicted health indicator trend;

[0133] Based on the predicted systolic blood pressure trends (150 mmHg, 155 mmHg, and 160 mmHg), combined with patient D's other health indicators (such as heart rate and blood sugar levels), a dynamic, three-dimensional predictive visual model was generated. In this model, time serves as the X-axis, systolic blood pressure serves as the Y-axis, and heart rate serves as the Z-axis. The model visualizes the predicted health indicator trends, helping researchers intuitively understand the patient's potential future health risks.

[0134] Dynamic predictive three-dimensional visual models can visualize complex predictive data, allowing researchers to quickly understand changing trends in patients' health status. This visualization tool not only improves the readability and usability of data, but also provides strong support for the reporting and presentation of clinical trials, helping to promote the dissemination and application of research results. In this way, researchers can better predict the occurrence of health events, formulate intervention measures in advance, and improve patients' treatment outcomes and quality of life.

[0135] After the step S372 of comprehensively analyzing the data features of each health indicator in the predicted health indicator trend to generate a dynamic prediction three-dimensional visual model, the following steps are included:

[0136] S3721: If the predicted three-dimensional visual model has a health event, then a risk mark is applied to the predicted three-dimensional visual model at that time point;

[0137] S3722: Simulating medical care behavior on the predicted three-dimensional visual model with risk markers to generate a three-dimensional visual model of simulated effects;

[0138] S3723: performing effect prediction on the three-dimensional visual model of the simulated effect based on the corresponding quantitative effect data to generate prediction index data corresponding to the simulated medical care behavior;

[0139] S3724: It should be understood that medical simulation reports are generated based on simulated medical behaviors, three-dimensional visual models of simulation effects and corresponding predictive indicator data.

[0140] For example, in steps S3721-S3724, if a patient is found to be at risk of a serious health event (such as a myocardial infarction) at a certain point in the future in the predictive 3D visual model, the model for that point in time is highlighted in red. The specific time of risk occurrence and the health indicators affected are also recorded. For the time point marked as high-risk for myocardial infarction, medical interventions such as preemptive thrombolytic therapy or interventional surgery are simulated. By updating model parameters and regenerating the predictive 3D visual model, the health indicator trends after these interventions are displayed. The simulated medical interventions and the resulting 3D visual model of the simulated effects can help researchers evaluate the effectiveness of different interventions, providing important evidence for clinical decision-making and improving the scientific and effective nature of medical decisions. Based on the model after simulated thrombolytic therapy, specific indicators such as a 40% reduction in myocardial infarction risk and a 25% improvement in cardiac function are predicted. New predictive indicator data is generated to help researchers quantify the effectiveness of the simulated medical interventions. By quantifying the effects of the simulated medical interventions, data is provided for comparing the pros and cons of different intervention plans, helping to select the optimal treatment plan and improve patient treatment outcomes and survival rates. Integrate simulated medical behaviors (such as thrombolytic therapy), three-dimensional visual models of simulation effects (displaying changes in health indicators after treatment) and predictive indicator data (such as 40% risk reduction) to generate a medical simulation report containing simulation processes, effect evaluation and recommendations for researchers and clinicians to discuss and make decisions. The medical simulation report summarizes the key information and expected effects of the simulated intervention, helps all parties quickly understand the potential benefits of the simulated intervention, promotes multidisciplinary team collaboration, jointly optimizes patient treatment plans, and improves the quality and efficiency of medical services.

[0141] The size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0142] In one embodiment, a device for dynamically visualizing the health events of clinical trial patients is provided. The device for dynamically visualizing the health events of clinical trial patients corresponds to the method for dynamically visualizing the health events of clinical trial patients in the above embodiment. Figure 2 As shown, the clinical trial patient health event dynamic visualization device includes:

[0143] Monitoring data requirement matching unit 1, used to obtain clinical trial information and input it into a pre-set clinical trial requirement database to match corresponding monitoring data requirements, wherein the monitoring data requirements are used to indicate the health indicator data that need to be collected in the clinical trial;

[0144] Health indicator data acquisition unit 2, used to acquire health indicator data of all patients in the clinical trial based on the monitoring data requirements, and preprocess and organize the health indicator data to construct a patient health indicator data set, wherein the health indicator data includes laboratory test result data, vital sign data and imaging test data;

[0145] The historical health event data acquisition unit 3 is used to acquire the patient's historical health event data and associate the health indicator data with the health event data based on a preset common time axis to construct a health event related data set;

[0146] A measured three-dimensional visual model generating unit 4 is used to perform a comprehensive analysis on the corresponding health indicator data in the health event related data set to generate a measured three-dimensional visual model associated with the common time axis;

[0147] A dynamic three-dimensional visual model generating unit 5 is configured to continuously fuse the measured three-dimensional visual models in a continuous time based on a common time axis to generate a dynamic three-dimensional visual model;

[0148] The clinical trial visualization report generating unit 6 is configured to generate a visualization report of the clinical trial based on the dynamic three-dimensional visual model, wherein the visualization report further includes a time series diagram of individual health events of the patient.

[0149] The specific definition of the dynamic visualization device for clinical trial patient health events can be found in the definition of the dynamic visualization method for clinical trial patient health events above, and will not be repeated here. The various modules in the above-mentioned dynamic visualization device for clinical trial patient health events can be implemented in whole or in part by software, hardware, and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the electronic device in the form of hardware, or can be stored in the memory of the electronic device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0150] In one embodiment, an electronic device is provided. The electronic device may be a server, and its internal structure diagram may be as follows: Figure 3As shown. The electronic device includes a processor, a memory, a network interface and a database connected via a system bus. The processor of the electronic device is used to provide computing and control capabilities. The memory of the electronic device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the electronic device is used to store a database. The network interface of the electronic device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, a method for dynamically visualizing health events of patients in a clinical trial is implemented.

[0151] In one embodiment, an electronic device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the following steps are performed:

[0152] Obtain clinical trial information and input it into a pre-set clinical trial requirements database to match corresponding monitoring data requirements, where the monitoring data requirements are used to indicate the health indicator data that needs to be collected in the clinical trial;

[0153] Acquire health indicator data of all patients in the clinical trial based on the monitoring data requirements, and preprocess and organize the health indicator data to construct a patient health indicator dataset, wherein the health indicator data includes laboratory test result data, vital sign data, and imaging test data;

[0154] Acquire historical health event data of the patient, and associate the health indicator data with the health event data based on a preset common timeline to construct a health event-related data set;

[0155] Performing a comprehensive analysis on the corresponding health indicator data in the health event-related data set to generate a measured three-dimensional visual model associated with the common time axis;

[0156] Continuously fusing the measured three-dimensional visual models over continuous time based on a common timeline to generate a dynamic three-dimensional visual model, wherein interactive functions are provided in the dynamic visualization display, allowing the dynamic three-dimensional visual model to be analyzed through a timeline slider and filter condition controls to observe the changing trends and patterns of patient health events over time;

[0157] A visualization report of the clinical trial is generated based on the dynamic three-dimensional visual model, and the visualization report also includes a time series diagram of individual health events of the patient.

[0158] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the following steps are implemented:

[0159] Obtain clinical trial information and input it into a pre-set clinical trial requirements database to match corresponding monitoring data requirements, where the monitoring data requirements are used to indicate the health indicator data that needs to be collected in the clinical trial;

[0160] Acquire health indicator data of all patients in the clinical trial based on the monitoring data requirements, and preprocess and organize the health indicator data to construct a patient health indicator dataset, wherein the health indicator data includes laboratory test result data, vital sign data, and imaging test data;

[0161] Acquire historical health event data of the patient, and associate the health indicator data with the health event data based on a preset common timeline to construct a health event-related data set;

[0162] Performing a comprehensive analysis on the corresponding health indicator data in the health event-related data set to generate a measured three-dimensional visual model associated with the common time axis;

[0163] Continuously fusing the measured three-dimensional visual models over continuous time based on a common timeline to generate a dynamic three-dimensional visual model, wherein interactive functions are provided in the dynamic visualization display, allowing the dynamic three-dimensional visual model to be analyzed through a timeline slider and filter condition controls to observe the changing trends and patterns of patient health events over time;

[0164] A visualization report of the clinical trial is generated based on the dynamic three-dimensional visual model, and the visualization report also includes a time series diagram of individual health events of the patient.

[0165] Those skilled in the art will appreciate that all or part of the processes in the above-described method embodiments can be implemented by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the above-described method embodiments. Any reference to memory, storage, database, or other media used in the various embodiments provided herein may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct RAMbus dynamic RAM (DRDRAM), and RAMbus dynamic RAM (RDRAM).

[0166] Those skilled in the art will clearly understand that for the sake of convenience and brevity of description, only the division of the above-mentioned functional units and modules is used as an example. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0167] The above-described embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present application, and should all be included in the scope of protection of the present application.

Claims

1. A method for dynamic visualization of health events of clinical trial patients, characterized by: The method comprises the steps of: obtaining clinical trial information and inputting it into a pre-set clinical trial requirement database to match corresponding monitoring data requirements, wherein the monitoring data requirements are used to indicate health indicator data that need to be collected in the clinical trial; Acquire health indicator data of all patients in the clinical trial based on the monitoring data requirements, and preprocess and organize the health indicator data to construct a patient health indicator dataset, wherein the health indicator data includes laboratory test result data, vital sign data, and imaging test data; Acquire historical health event data of the patient, and associate the health indicator data with the health event data based on a preset common timeline to construct a health event-related data set; Performing a comprehensive analysis on the corresponding health indicator data in the health event-related data set to generate a measured three-dimensional visual model associated with the common time axis; Continuously fusing the measured three-dimensional visual models over continuous time based on a common timeline to generate a dynamic three-dimensional visual model, wherein interactive functions are provided in the dynamic visualization display, allowing the dynamic three-dimensional visual model to be analyzed through a timeline slider and filter condition controls to observe the changing trends and patterns of patient health events over time; generating a visualization report of the clinical trial based on the dynamic three-dimensional visual model, wherein the visualization report further includes a time series diagram of individual health events of the patient; After obtaining the patient's historical health event data and associating the health indicator data with the health event data based on a preset common timeline to construct a health event-related data set, the following steps are included: Obtain clinical medical records corresponding to each health event; Recording health indicator data before and after a health event based on a preset time length, and associating the health event, clinical medical records, and health indicator data one-to-one based on the common timeline to construct a nursing effect dataset; The preset nursing effect analysis model analyzes the nursing effect data set based on a machine self-learning algorithm to generate corresponding effect quantitative data; Predicting the effect of the current health event based on the effect quantification data to generate prediction indicator data; Performing a comprehensive analysis on the prediction indicator data to generate a three-dimensional visual model for effect prediction; Specifically, after the step of predicting the effect of the current health event based on the effect quantification data to generate prediction index data, the following steps are included: Classify patient health events in the same clinical trial and generate corresponding event labels; Obtaining quantitative effect data of different patients within the same event label, and comparing the quantitative effect data of different patients with each other to generate effect difference data; Obtaining vital sign data and basic data of different patients, and comparing the vital sign data and basic data of different patients to generate vital sign difference data and basic characteristic difference data; Associating the physical sign difference data, basic feature difference data and effect difference data to construct a patient effect difference data set; The preset effect difference analysis model analyzes the patient effect difference data set based on a preset machine self-learning algorithm to generate an effect deviation factor; The prediction index data is corrected based on the effect deviation factor so that the prediction result is more consistent with the actual situation of the patient himself.

2. A method for dynamic visualization of health events of clinical trial patients according to claim 1, characterized in that: After comprehensively analyzing the corresponding health indicator data in the health event-related data set to generate a measured three-dimensional visual model associated with the common time axis, the following steps are included: Obtaining all corresponding index baseline value intervals in the clinical trial, and comparing each of the monitored health index data with the corresponding index baseline value intervals. If the health index data is not within the index baseline value interval, the corresponding health index data is marked as abnormal; The health indicator data with abnormal marks are marked on the measured three-dimensional visual model in a highlighted form.

3. A method for dynamic visualization of health events of clinical trial patients according to claim 1, characterized in that: After obtaining the patient's historical health event data and associating the health indicator data with the health event data based on a preset common timeline to construct a health event related data set, the following steps are included: The preset health event warning model analyzes the health event related data set based on a machine self-learning algorithm to generate a health event warning trend, wherein the health event warning trend includes corresponding indicator features; When the changing trend of the health indicator data is similar to the indicator characteristics corresponding to the health event warning trend, a warning signal is issued.

4. A method for dynamic visualization of health events of clinical trial patients according to claim 3, characterized in that: After the step of issuing a warning signal when the changing trend of the health indicator data is similar to the indicator characteristics corresponding to the health event warning trend, the following steps are included: Predict the patient's health indicator data based on a dynamic time series analysis model and generate predicted health indicator trends; A comprehensive analysis is performed on the data features of each health indicator in the predicted health indicator trend to generate a dynamic prediction three-dimensional visual model, and the prediction three-dimensional visual model is used to predict the occurrence of health events in the time period corresponding to the predicted health indicator trend.

5. A method for dynamic visualization of health events of clinical trial patients according to claim 4, characterized in that: After comprehensively analyzing the data features of each health indicator in the predicted health indicator trend to generate a dynamic prediction three-dimensional visual model, the following steps are included: If the predicted three-dimensional visual model indicates a health event, the predicted three-dimensional visual model at that time point is marked as a risk event; Simulating medical care behaviors on the predicted three-dimensional visual model with risk markers to generate a three-dimensional visual model of simulated effects; Performing effect prediction on the three-dimensional visual model of the simulated effect based on the corresponding quantitative effect data to generate prediction index data corresponding to the simulated medical care behavior; Generate a medical simulation report based on simulated medical behavior, three-dimensional visual model of simulation effect and corresponding prediction indicator data.

6. A device for dynamically visualizing health events of clinical trial patients, applied to a method for dynamically visualizing health events of clinical trial patients according to any one of claims 1 to 5, characterized in that: The device comprises: A monitoring data requirement matching unit (1) is used to obtain clinical trial information and input it into a pre-set clinical trial requirement database to match corresponding monitoring data requirements, wherein the monitoring data requirements are used to indicate health indicator data that need to be collected in the clinical trial; A health indicator data acquisition unit (2) is used to acquire health indicator data of all patients in the clinical trial based on the monitoring data requirements, and pre-process and organize the health indicator data to construct a patient health indicator data set, wherein the health indicator data includes laboratory test result data, vital sign data and imaging test data; A historical health event data acquisition unit (3) is used to acquire the patient's historical health event data and associate the health indicator data with the health event data based on a preset common time axis to construct a health event related data set; A measured three-dimensional visual model generating unit (4) is used to comprehensively analyze the corresponding health indicator data in the health event related data set to generate a measured three-dimensional visual model associated with the common time axis; A dynamic three-dimensional visual model generation unit (5) is used to continuously fuse the measured three-dimensional visual models in continuous time based on a common time axis to generate a dynamic three-dimensional visual model; A clinical trial visualization report generating unit (6) is used to generate a visualization report of the clinical trial based on the dynamic three-dimensional visual model, wherein the visualization report also includes a time series diagram of individual health events of the patient.

7. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the steps of the method for dynamically visualizing health events of clinical trial patients as described in any one of claims 1 to 5 are implemented.

8. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the method for dynamically visualizing health events of clinical trial patients as described in any one of claims 1 to 5 are implemented.

Images