Remote monitoring method for clinical test
Through intelligent algorithms and big data analysis, the monitoring strategies are automatically adjusted in clinical trials, which solves the shortcomings of traditional methods in data analysis and strategy adjustment, and achieves efficient and accurate monitoring, reducing resource waste and data security risks, ensuring that The experiment was carried out smoothly.
Patent Information
- Application Number
- CN202411036382.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-31
- Publication Date
- 2024-12-03
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
Traditional clinical trial remote monitoring methods have shortcomings in data analysis and strategy adjustment, and it is difficult to deeply explore the complex relationships and potential risks behind the data, and lack real-time and dynamic nature, resulting in waste of monitoring resources and omission of potential problems.
Using intelligent algorithms to deeply mine and analyze the experimental data collected in real time, automatically adjust the monitoring strategies and plans based on the analysis results, identify potential problems and risks through big data analysis and artificial intelligence algorithms, and build a remote monitoring platform to achieve real-time feedback and security.
It significantly improves the monitoring efficiency and quality, reduces the interference of external factors on the trial, enhances data security, and provides guarantees for the smooth progress of clinical trials.
Smart Images

Figure CN119069054A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of clinical trials, and in particular to a remote monitoring method for clinical trials. Background Art
[0002] With the rapid development of medical technology, clinical trials, as an important link before new drugs and new therapies are launched on the market, are directly related to the pace of medical innovation and the improvement of patient welfare in terms of quality and efficiency. In recent years, with the continuous maturity and popularization of advanced technologies such as big data and artificial intelligence, these technologies have gradually penetrated into all aspects of clinical trials, providing new possibilities for the modern management of clinical trials. In particular, the rise of remote monitoring technology has not only broken the limitations of geographical space, but also provided support for the real-time monitoring and efficient management of clinical trials.
[0003] Although remote monitoring technology has begun to be applied in clinical trials, traditional methods still have many shortcomings in handling intelligent analysis and dynamic adjustments. First, traditional data analysis methods are often limited to simple statistical descriptions, and it is difficult to deeply explore the complex relationships and potential risks behind the data. This leads to the easy omission of important information during the monitoring process and the inability to timely discover and deal with potential problems. Secondly, the formulation and adjustment of traditional monitoring strategies usually rely on the experience and intuition of monitors and lack scientific and systematic decision-making support. This not only leads to a waste of monitoring resources, but also introduces deviations due to human factors, affecting the monitoring effect. Finally, traditional methods lack real-time and dynamics, and cannot adjust monitoring strategies in time according to the progress of the trial and data analysis results, making it difficult to cope with emergencies and complex changes.
[0004] In response to the above problems, it is necessary to optimize the existing remote monitoring technology, conduct in-depth mining and analysis of the test data collected in real time through intelligent algorithms, and automatically adjust the monitoring strategies and plans based on the analysis results. Therefore, it is of great significance to develop a remote monitoring method for clinical trials that can comprehensively realize the above characteristics. Summary of the invention
[0005] The purpose of the present invention is to make up for the shortcomings of the existing technology and provide a remote monitoring method for clinical trials. It can deeply mine and analyze the test data collected in real time through intelligent algorithms, and automatically adjust the monitoring strategies and plans based on the analysis results, so as to achieve efficient and accurate monitoring of clinical trials. It can not only significantly improve the monitoring efficiency and quality, reduce the interference of external factors on the trial, but also enhance the security of data, and provide strong guarantee for the smooth progress of clinical trials.
[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: a remote monitoring method for clinical trials, the system comprising the following specific steps:
[0007] S100, remote monitoring platform construction: establish a remote monitoring platform including data collection, intelligent analysis, dynamic adjustment and remote interaction functions;
[0008] S200, data collection and preprocessing: using electronic data collection system (EDC) tools to collect various data in clinical trials in real time and perform preprocessing;
[0009] S300, intelligent analysis and dynamic adjustment: using big data analysis and artificial intelligence algorithms to conduct in-depth mining of pre-processed data, identify potential problems and risks, and automatically trigger the adjustment of monitoring strategies based on the results of intelligent analysis;
[0010] S400, real-time feedback and visual reports: Generate visual monitoring reports to intuitively display data analysis results, monitoring strategy adjustments and trial progress, and provide real-time feedback mechanisms;
[0011] S500, Security Assurance: Build a secure and reliable communication connection and data encryption mechanism, and monitor and record the operating behavior of the inspectors to prevent data leakage and illegal access.
[0012] Further, the S100 includes:
[0013] S101, design the platform architecture, clarify the interfaces for data collection, intelligent analysis, dynamic adjustment and remote interaction functions, and select and deploy the infrastructure of servers, databases and network communication equipment;
[0014] S102, develop remote interactive interface, including data display, report generation and user authority management functions.
[0015] Furthermore, the S200 includes:
[0016] S201, configuring the interface between the electronic data collection system (EDC) and the hospital information system (HIS) data source to realize automatic data collection;
[0017] S202, clean the collected data, remove duplicate, erroneous and irrelevant data items, and perform data desensitization to protect the privacy and sensitive information of the subjects;
[0018] S203, storing the processed data into a database for subsequent intelligent analysis and monitoring.
[0019] Furthermore, the S300 includes:
[0020] S301, further processing the collected data, including data conversion and normalization, and extracting key features from the processed data for subsequent analysis;
[0021] S302: Use the trained machine learning or deep learning model to predict the data, identify abnormal patterns in the data, including outliers and trend deviations, and based on the abnormal detection results, evaluate their impact on the quality and progress of the clinical trial and determine the risk level;
[0022] S303, automatically adjust the monitoring strategy according to the risk assessment results.
[0023] Furthermore, the S301 uses principal component analysis to extract key features from the processed data. Specifically, the covariance matrix C of the data set X is calculated, where X T is the transpose of X, and the covariance matrix C is decomposed to obtain the eigenvalues λ1,λ2,…,λ m and its corresponding eigenvectors v1,v2,…,v m , select the eigenvectors corresponding to the first k eigenvalues according to the size of the eigenvalue, where k is less than or equal to m, and these eigenvectors constitute a projection matrix V; project the data set X into the subspace defined by the projection matrix V to obtain the reduced-dimensional data set Y, where Y = XV, and each column in Y corresponds to a principal component, representing the key features in the data.
[0024] Furthermore, the S302 uses the ARIMA algorithm to predict the data and identify abnormal patterns in the data. Specifically, the ARIMA (p, d, q) model is used to model the preprocessed data, where p is the order of the autoregressive model, d is the number of differences, and q is the order of the moving average model. The mathematical expression of the ARIMA model is: Among them, y t is the observed value at the current time point, y t-i is the observed value at the past i time points, φ i is the autoregressive coefficient, ∈ t is the white noise error term, θ j is the moving average coefficient, c is a constant term. If the time series data is non-stationary, d times of difference processing is performed to make the series stable. The trained ARIMA model is used to predict future time series data, and an abnormal threshold is set. If the predicted value and the actual observed value exceed the set threshold, the system recognizes it as an abnormal mode and triggers corresponding dynamic adjustment measures.
[0025] Furthermore, the specific steps of evaluating the impact of abnormal detection results on the quality and progress of clinical trials in S302 are as follows:
[0026] (1) Classify and process abnormal results according to the nature and severity of the abnormality;
[0027] (2) Analyze the scope of each abnormal result and predict the potential consequences based on the nature and scope of the abnormality;
[0028] (3) Quantify the risk factors of severity, probability of occurrence, and duration of abnormal impacts;
[0029] (4) Comprehensively consider various risk factors and use the risk assessment model to comprehensively evaluate the anomaly detection results to determine their risk level.
[0030] Furthermore, the specific measures of the automatic adjustment of the monitoring strategy in S303 are:
[0031] (1) Increase the frequency of monitoring for areas where abnormalities are found or where risks are high, and appropriately reduce the frequency of monitoring for areas where risks are stable and low;
[0032] (2) Expand the scope of monitoring for new risk points or changes in business needs, and reduce the scope of monitoring for areas that are already stable and have controllable risks;
[0033] (3) Update monitoring rules in a timely manner based on business needs and data analysis results.
[0034] Furthermore, the S400 includes:
[0035] S401, converting the intelligent analysis results and the monitoring strategy adjustment into charts and reports, and automatically generating a monitoring report;
[0036] S402, through a remote interactive interface, provides real-time feedback to sponsors, researchers, and monitors, supporting online discussions and collaboration.
[0037] Furthermore, the S500 includes:
[0038] S501, use HTTPS security protocol, and set up firewalls and intrusion detection systems to prevent external attacks;
[0039] S502, encrypting the storage and transmission of sensitive data, including subject information and trial data, and using a key management system;
[0040] S503, implement strict access control policies to ensure that only authorized personnel can access the system and data and record access logs.
[0041] Compared with the existing technology, this remote monitoring method for clinical trials has the following beneficial effects:
[0042] 1. The present invention introduces a dynamic adjustment mechanism to automatically trigger the adjustment of the monitoring strategy according to the results of intelligent analysis. This dynamically optimized monitoring strategy helps to control trial risks more efficiently and accurately and ensure the smooth progress of clinical trials. At the same time, the system can also automatically expand or reduce the scope of monitoring according to new risk points or changes in business needs to ensure the comprehensiveness and pertinence of the monitoring work.
[0043] 2. The present invention realizes in-depth mining and intelligent analysis of clinical trial data by introducing big data analysis and artificial intelligence algorithms. This method can automatically identify potential problems and risks in the data, thereby avoiding omissions and delays in traditional manual monitoring. The results of intelligent analysis can trigger the adjustment of monitoring strategies in real time, ensuring that monitoring work can be accurately focused on high-risk or abnormal areas, significantly improving the efficiency and accuracy of monitoring work.
[0044] Other advantages, objectives and features of the present invention will be set forth in part in the following description and, in part, will be apparent to those skilled in the art based on an examination of the following or may be taught from the practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS
[0045] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.
[0046] Figure 1 The figure is a flowchart of a remote monitoring method for clinical trials. DETAILED DESCRIPTION
[0047] The technical solutions in the embodiments of the present invention are described clearly and completely below. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0048] Embodiment 1
[0049] This embodiment specifically describes the detailed implementation process of a remote monitoring method for clinical trials, which not only improves monitoring efficiency and accuracy, but also ensures data security and compliance through multi-level security measures.
[0050] During the remote monitoring platform construction phase (S100), when designing the platform architecture, you should clarify the interface specifications of core functions such as data collection, intelligent analysis, dynamic adjustment and remote interaction to ensure smooth data exchange and command transmission between modules, select and deploy high-performance servers, databases and network communication equipment, build a stable and reliable infrastructure environment, and provide solid support for remote monitoring. At the same time, develop an intuitive and easy-to-use remote interaction interface, which includes functions such as data display, report generation and user authority management. Through the graphical interface, sponsors, researchers and monitors can view test data, analysis results and monitoring strategy adjustments in real time, and support online discussion and collaboration.
[0051] In the data collection and preprocessing stage (S200), the interface between the electronic data collection system (EDC) and the hospital information system (HIS) is configured to realize the automatic and real-time collection of clinical trial data. This not only reduces manual input errors, but also greatly improves data collection efficiency. The collected data is strictly cleaned to remove duplicate, erroneous and irrelevant data items to ensure the accuracy and completeness of the data. At the same time, data desensitization is performed to protect the privacy and sensitive information of the subjects, comply with relevant laws and regulations, and store the processed data in the database for subsequent intelligent analysis and monitoring. The design of the database must take into account both data storage efficiency and query performance to ensure rapid response to various data analysis needs.
[0052] In the intelligent analysis and dynamic adjustment stage (S300), the collected data is further processed, including data conversion and normalization, to improve data quality and analysis results. The principal component analysis (PCA) method is used to extract key features from the processed data, reduce data dimensions and retain main information. The trained machine learning or deep learning model is used to predict the data, identify abnormal patterns in the data (such as outliers and trend deviations), and establish a data prediction model based on time series analysis algorithms such as ARIMA. The abnormality threshold is set to identify potential abnormal points, and the abnormalities are classified and processed according to their nature and severity. The impact on the quality and progress of clinical trials is evaluated, and the risk level is determined. According to the risk assessment results, the monitoring strategy is automatically adjusted. For areas where abnormalities are found or risks are higher, the monitoring frequency is increased. For stable and low-risk areas, the monitoring frequency is appropriately reduced. At the same time, according to business needs and data analysis results, the monitoring rules are updated in a timely manner to ensure the pertinence and effectiveness of the monitoring work.
[0053] In the real-time feedback and visual reporting stage (S400), the intelligent analysis results and monitoring strategy adjustments are converted into charts and reports, and an intuitive and easy-to-understand monitoring report is automatically generated. The report content covers important information such as data analysis results, monitoring strategy adjustments and trial progress. It provides real-time feedback to sponsors, researchers and monitors through a remote interactive interface, supports online discussion and collaboration, promotes communication and information sharing among all parties, and ensures the smooth progress of clinical trials.
[0054] In the security assurance stage (S500), a secure and reliable communication connection and data encryption mechanism are built, the HTTPS security protocol is adopted and a firewall and intrusion detection system are set up to prevent external attacks, sensitive data is encrypted for storage and transmission, a key management system is used to ensure data security, a strict access control policy is implemented to ensure that only authorized personnel can access the system and data, and access logs are recorded for tracking and auditing.
[0055] In summary, this embodiment not only improves the monitoring efficiency and accuracy by building a clinical trial remote monitoring platform that integrates intelligent analysis and dynamic adjustment strategies, but also ensures the security and compliance of data through multi-level security measures, providing strong support for the smooth progress of clinical trials.
[0056] Embodiment 2
[0057] Based on the first embodiment, this embodiment describes in detail a method for implementing the intelligent analysis and dynamic adjustment phase of a remote monitoring method for a clinical trial.
[0058] In the intelligent analysis and dynamic adjustment stage (S300), the collected data is further processed, including data conversion and normalization, to improve data quality and analysis effect. The principal component analysis (PCA) method is used to extract key features from the processed data, reduce the data dimension and retain the main information. Specifically, the covariance matrix C of the data set X is calculated, where X T is the transpose of X, and the covariance matrix C is decomposed to obtain the eigenvalues λ1,λ2,…,λ m and its corresponding eigenvectors v1,v2,…,v m , select the eigenvectors corresponding to the first k eigenvalues according to the size of the eigenvalue, where k is less than or equal to m, and these eigenvectors constitute a projection matrix V; project the data set X into the subspace defined by the projection matrix V to obtain the reduced-dimensional data set Y, where Y = XV, and each column in Y corresponds to a principal component, representing the key features in the data.
[0059] Use the trained machine learning or deep learning model to predict the data, identify abnormal patterns in the data, establish a data prediction model based on time series analysis algorithms such as ARIMA, and set an abnormal threshold to identify potential anomalies. Specifically, use the ARIMA (p, d, q) model to model the preprocessed data, where p is the order of the autoregressive model, d is the number of differences, and q is the order of the moving average model. The mathematical expression of the ARIMA model is:, where is the observation value at the current time point, is the observation value at the past time point, is the autoregressive coefficient, is the white noise error term, is the moving average coefficient, and is the constant term. For non-stationary time series data, perform d times of difference processing to make the series stable. Use the trained ARIMA model to predict future time series data and set an abnormal threshold. If the predicted value and the actual observed value exceed the set threshold, the system identifies it as an abnormal pattern and triggers corresponding dynamic adjustment measures.
[0060] Classify and handle abnormalities according to their nature and severity, evaluate their impact on the quality and progress of clinical trials, determine the risk level, and automatically adjust the monitoring strategy based on the risk assessment results. The specific steps are as follows:
[0061] (1) Classify and process abnormal results according to the nature and severity of the abnormality;
[0062] (2) Analyze the scope of each abnormal result and predict the potential consequences based on the nature and scope of the abnormality;
[0063] (3) Quantify the risk factors of severity, probability of occurrence, and duration of abnormal impacts;
[0064] (4) Comprehensively consider various risk factors and use the risk assessment model to comprehensively evaluate the anomaly detection results to determine their risk level.
[0065] According to the risk assessment results, the monitoring strategy is automatically adjusted. The specific measures are as follows:
[0066] (1) Increase the frequency of monitoring for areas where abnormalities are found or where risks are high, and appropriately reduce the frequency of monitoring for areas where risks are stable and low;
[0067] (2) Expand the scope of monitoring for new risk points or changes in business needs, and reduce the scope of monitoring for areas that are already stable and have controllable risks;
[0068] (3) Update monitoring rules in a timely manner based on business needs and data analysis results.
[0069] In summary, this embodiment introduces a dynamic adjustment mechanism to automatically trigger the adjustment of the monitoring strategy according to the results of intelligent analysis. This dynamically optimized monitoring strategy helps to control trial risks more efficiently and accurately and ensure the smooth progress of clinical trials. At the same time, the system can also automatically expand or reduce the scope of monitoring according to new risk points or changes in business needs to ensure the comprehensiveness and pertinence of the monitoring work.
[0070] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.
Claims
1. A remote monitoring method for clinical trials, It is characterized in that The method comprises the following specific steps: S100, remote monitoring platform construction: establish a remote monitoring platform including data collection, intelligent analysis, dynamic adjustment and remote interaction functions; S200, data collection and preprocessing: using electronic data collection system (EDC) tools to collect various data in clinical trials in real time and perform preprocessing; S300, intelligent analysis and dynamic adjustment: using big data analysis and artificial intelligence algorithms to conduct in-depth mining of pre-processed data, identify potential problems and risks, and automatically trigger the adjustment of monitoring strategies based on the results of intelligent analysis; S400, real-time feedback and visual reports: Generate visual monitoring reports to intuitively display data analysis results, monitoring strategy adjustments and trial progress, and provide real-time feedback mechanisms; S500, Security Assurance: Build a secure and reliable communication connection and data encryption mechanism, and monitor and record the operating behavior of the inspectors to prevent data leakage and illegal access.
2. A remote monitoring method for clinical trials according to claim 1, It is characterized in that The S100 includes: S101, design the platform architecture, clarify the interfaces for data collection, intelligent analysis, dynamic adjustment and remote interaction functions, and select and deploy the infrastructure of servers, databases and network communication equipment; S102, develop remote interactive interface, including data display, report generation and user authority management functions.
3. A remote monitoring method for clinical trials according to claim 1, It is characterized in that The S200 includes: S201, configuring the interface between the electronic data collection system (EDC) and the hospital information system (HIS) data source to realize automatic data collection; S202, clean the collected data, remove duplicate, erroneous and irrelevant data items, and perform data desensitization to protect the privacy and sensitive information of the subjects; S203, storing the processed data into a database for subsequent intelligent analysis and monitoring.
4. A remote monitoring method for clinical trials according to claim 1, It is characterized in that The S300 includes: S301, further processing the collected data, including data conversion and normalization, and extracting key features from the processed data for subsequent analysis; S302: Use the trained machine learning or deep learning model to predict the data, identify abnormal patterns in the data, including outliers and trend deviations, and based on the abnormal detection results, evaluate their impact on the quality and progress of the clinical trial and determine the risk level; S303, automatically adjust the monitoring strategy according to the risk assessment results.
5. A remote monitoring method for clinical trials according to claim 4, It is characterized in that The S301 uses principal component analysis to extract key features from the processed data. Specifically, the covariance matrix C of the data set X is calculated, where X T is the transpose of X, and the covariance matrix C is decomposed to obtain the eigenvalue λ 1 ,λ 2 ,…,λ m and its corresponding eigenvector v 1 ,v 2 ,…,v m , select the eigenvectors corresponding to the first k eigenvalues according to the size of the eigenvalue, where k is less than or equal to m, and these eigenvectors constitute a projection matrix V; project the data set X into the subspace defined by the projection matrix V to obtain the reduced-dimensional data set Y, where Y = XV, and each column in Y corresponds to a principal component, representing the key features in the data.
6. A remote monitoring method for clinical trials according to claim 4, It is characterized in that The S302 uses the ARIMA algorithm to predict the data and identify abnormal patterns in the data. Specifically, the ARIMA (p, d, q) model is used to model the preprocessed data, where p is the order of the autoregressive model, d is the number of differences, and q is the order of the moving average model. The mathematical expression of the ARIMA model is: Among them, y t is the observed value at the current time point, y t-i is the observed value at the past i time points, φ i is the autoregressive coefficient, ∈ t is the white noise error term, θ j is the moving average coefficient, c is a constant term. If the time series data is non-stationary, d times of difference processing is performed to make the series stable. The trained ARIMA model is used to predict future time series data, and an abnormal threshold is set. If the predicted value and the actual observed value exceed the set threshold, the system recognizes it as an abnormal mode and triggers corresponding dynamic adjustment measures.
7. According to the remote monitoring method of clinical trial according to claim 4, It is characterized in that The specific steps of evaluating the impact of abnormal detection results on the quality and progress of clinical trials in S302 are as follows: (1) Classify and process abnormal results according to the nature and severity of the abnormality; (2) Analyze the scope of each abnormal result and predict the potential consequences based on the nature and scope of the abnormality; (3) Quantify the risk factors of severity, probability of occurrence, and duration of abnormal impacts; (4) Comprehensively consider various risk factors and use the risk assessment model to comprehensively evaluate the anomaly detection results to determine their risk level.
8. A remote monitoring method for clinical trials according to claim 4, It is characterized in that The specific measures of the automatic adjustment of the monitoring strategy in S303 are: (1) Increase the frequency of monitoring for areas where abnormalities are found or where risks are high, and appropriately reduce the frequency of monitoring for areas where risks are stable and low; (2) Expand the scope of monitoring for new risk points or changes in business needs, and reduce the scope of monitoring for areas that are already stable and have controllable risks; (3) Update monitoring rules in a timely manner based on business needs and data analysis results.
9. A remote monitoring method for clinical trials according to claim 1, It is characterized in that The S400 includes: S401, converting the intelligent analysis results and the monitoring strategy adjustment into charts and reports, and automatically generating a monitoring report; S402, through a remote interactive interface, provides real-time feedback to sponsors, researchers, and monitors, supporting online discussions and collaboration.
10. A remote monitoring method for clinical trials according to claim 1, It is characterized in that The S500 includes: S501, use HTTPS security protocol, and set up firewalls and intrusion detection systems to prevent external attacks; S502, encrypting the storage and transmission of sensitive data, including subject information and trial data, and using a key management system; S503, implement strict access control policies to ensure that only authorized personnel can access the system and data and record access logs.
Citation Information
Cited By
Medical equipment remote control system based on Internet of Things data
CN119916667A
