Clinical trial based risk identification and early warning method and system

Abstract

The present application relates to the technical field of risk early warning, and particularly relates to a risk identification and early warning method and system based on clinical trials, comprising: a collection and storage unit, a risk identification unit, an evaluation and correction unit, and a data grabbing unit. The present application collects the rate of serious adverse events and the basic information and examination data of subjects in time through an input module, provides a data basis for risk assessment, calculates basic representation values and liver function representation values through these data, provides quantitative indicators for risk identification, makes a grade determination on drug risks according to these representation values and preset early warning evaluation standards, effectively distinguishes different risk levels, dynamically corrects early warning evaluation according to sample size and abnormal threshold, improves the accuracy and adaptability of the early warning system, compares the data of similar drugs, provides a comprehensive, dynamic and adjustable risk management solution for the safety supervision of clinical trials, and speeds up the drug development process.

Description

Technical Field

[0001] The present invention relates to the technical field of risk warning, and in particular to a risk identification and warning method and system based on clinical trials. Background Art

[0002] Clinical trials are usually divided into several phases, each with its own specific objectives and risk management needs. The purpose of the first phase is to evaluate the preliminary safety of the drug, determine the maximum tolerated dose of the drug, and understand the pharmacokinetic and pharmacodynamic properties of the drug; the purpose of the second phase is to evaluate the efficacy of the drug, determine the drug's therapeutic effect on specific conditions, and monitor short-term side effects; the purpose of the third phase is to further verify the efficacy and safety of the drug in a wider population and monitor long-term side effects and drug interactions.

[0003] During Phase II and III clinical trials, namely the efficacy and safety evaluation phases, risk early warning for the trial population is crucial. This phase typically involves a larger group of subjects, with the goal of evaluating the efficacy of the investigational drug, monitoring its side effects, and determining the optimal dosage range. Because Phase II and III may involve a wider patient population, a systematic approach is needed to identify and warn of potential risks to protect subject safety and optimize the development process of the investigational drug. Timely identification and response to potential risks during clinical trials is crucial for protecting subject safety, maintaining trial integrity, and promoting new drug development. The clinical application of risk early warning systems is crucial in clinical trials, as they enable real-time monitoring and assessment of subject safety throughout the trial, enabling timely identification of potential risks and adverse events. By accurately calculating subject baseline and liver function markers, the system can stratify individualized risk levels, providing a scientific basis for clinical decision-making. Furthermore, by dynamically adjusting early warning assessments and flagging abnormal samples, the system enhances subject protection and optimizes clinical trial design and management. At different stages of drug development, risk warning systems help balance innovation and safety, ensuring that new drugs can safely and effectively serve patients, while accelerating the process of drug launch, which is of great significance to improving public health and promoting medical progress.

[0004] Although risk warning systems in existing technologies play an important role in the second and third phases of clinical trials, they still have some shortcomings, including providing accurate personalized risk assessments for diverse subject groups, adapting to dynamic changes in risk patterns during trial progress, strengthening monitoring of long-term side effects, effectively interpreting and applying increasing amounts of data, achieving efficient integration of multiple data sources, rationally allocating resources to prioritize the identification of high-risk subjects, combining the professional judgment of clinicians to avoid over-reliance on automated systems, improving the interpretability of the system to enhance user trust, fully considering the diversity of subjects, and adapting to regulatory requirements in different regions. These shortcomings require continuous improvement and optimization of the existing system to enhance its effectiveness and reliability in clinical trials. Summary of the Invention

[0005] To this end, the present invention provides a risk identification and early warning method and system based on clinical trials, which is used to overcome the problem that the risk identification and early warning system in the prior art has poor real-time monitoring data capabilities when the subject group expands and the amount of data increases significantly in the second and third phase trials, resulting in the inability to accurately identify and warn risks.

[0006] To achieve the above objectives, the present invention provides a risk identification and early warning system based on clinical trials, comprising:

[0007] An acquisition and storage unit, comprising an input module and a calculation module, wherein the input module is used to receive and store externally input serious adverse event rates of clinical trial drugs, basic information of the subject population, and examination data of the subject population;

[0008] The basic information of the subjects included age, disease duration, and BMI;

[0009] The examination data of the subjects included hormone levels, alanine aminotransferase levels, total bilirubin levels, and albumin levels;

[0010] The calculation module is used to calculate the basic characterization value of the subject population according to age, length of illness and BMI, and calculate the liver function characterization value of the subject population according to the alanine aminotransferase level value, total bilirubin level value and albumin level value;

[0011] a risk identification unit connected to the acquisition and storage unit, for determining the rate of serious adverse events according to standard early warning evaluation and determining a risk level, wherein the risk level includes a low risk level, a medium risk level, and a high risk level;

[0012] An evaluation correction unit, connected to the risk identification unit, is used to correct the standard warning evaluation according to the sample size of the subject population and the evaluation amount of the standard sample, and determine whether each sample is an abnormal sample based on the basic characterization value of the subject population, the liver function characterization value of the subject population, and the hormone level value of any sample in the subject population and the corresponding preset abnormal threshold value, wherein the abnormal threshold value includes the basic characterization value threshold, the liver function characterization value threshold, and the hormone level value threshold, and check the proportion of abnormal samples to determine whether to correct the evaluation amount of the standard sample;

[0013] The data capture unit is connected to the evaluation and correction unit, and is used to capture the test data of similar drugs in the large database, compare the differences, and determine the direction of the differences based on the differences to determine whether to issue an extraordinary warning.

[0014] Furthermore, the risk identification unit is provided with a standard early warning evaluation, which includes a first evaluation event rate and a second evaluation event rate;

[0015] The risk identification unit can determine the adverse event rate entered into the collection and storage unit according to the standard early warning evaluation.

[0016] If the adverse event rate is lower than the first evaluation event rate, the risk identification unit will determine that the current clinical trial drug has low risk and will not issue a risk warning;

[0017] If the adverse event rate is greater than or equal to the first evaluation event rate and less than or equal to the second evaluation event rate, the risk identification unit will control the data capture unit to capture the test data of similar drugs in the large database and make a judgment to determine whether a risk warning is issued;

[0018] If the adverse event rate is greater than the second evaluation event rate, the risk identification unit determines to issue a risk warning.

[0019] Furthermore, a standard difference degree is set in the data capture unit. If the adverse event rate is greater than or equal to the first evaluation event rate and less than or equal to the second evaluation event rate, the data capture unit will capture the test data of similar drugs in the large database and input the test data results into the Mann-Whitney U model. The Mann-Whitney U model compares the serious adverse event rates of the clinical trial drugs and similar drugs, the basic information of the test population, and the examination data of the test population to obtain the same type difference degree. The data capture unit compares the standard difference degree with the same type difference degree.

[0020] If the difference between the same category is less than or equal to the standard difference, no risk warning will be issued;

[0021] If the difference between the same category is greater than the standard difference, a risk warning will be issued.

[0022] Furthermore, a standard sample evaluation range is provided in the evaluation correction unit. When the acquisition storage unit collects data, the evaluation correction unit can obtain the sample size of the test population to correct the standard warning evaluation in the risk identification unit, and compare the sample size of the test population with the standard sample evaluation range.

[0023] If the sample size of the test population is within the standard sample evaluation range, the evaluation correction unit determines not to perform correction;

[0024] If the sample size of the test population is lower than the minimum value of the standard sample evaluation range, the evaluation correction unit reduces the second evaluation event rate;

[0025] If the sample size of the test population is higher than the maximum value of the standard sample evaluation range, the evaluation correction unit increases the first evaluation event rate.

[0026] Furthermore, the evaluation and correction unit is also provided with a basic characterization value threshold, a liver function characterization value threshold and a hormone level value threshold. The evaluation and correction unit compares the basic characterization value of the subject population of any sample with the basic characterization value threshold, and compares the liver function characterization value of the subject population with the liver function characterization value threshold, and compares the hormone level value with the hormone level value threshold, and determines whether the sample is abnormal based on the comparison result, and marks the abnormal sample as abnormal when it is determined to be abnormal, thereby forming an abnormal sample.

[0027] Furthermore, the evaluation correction unit determines whether the sample is abnormal based on the comparison result.

[0028] If the basic characterization value of the test population, the liver function characterization value of the test population, and the hormone level value are all greater than the corresponding abnormal threshold value, the evaluation correction unit determines that the sample is abnormal.

[0029] Furthermore, the evaluation correction unit is further provided with a correction sample ratio. When the evaluation correction unit completes the abnormality determination of all samples, it calculates the abnormality ratio according to the number of abnormal samples and compares the abnormality ratio with the correction sample ratio.

[0030] If the abnormal proportion is greater than the corrected sample proportion, the evaluation correction unit will adjust the standard sample evaluation range according to the corrected sample proportion.

[0031] Furthermore, the calculation module is used to calculate the basic characterization values ​​of the test population according to age, disease duration and BMI.

[0032]

[0033] Among them, Q is the basic characterization value of the subject population, A is age, A0 is the average age, a is the standard deviation of age, B is the length of illness, B0 is the average length of illness, b is the standard deviation of length of illness, C is BMI, C0 is the average BMI, and c is the standard deviation of BMI.

[0034] Furthermore, the calculation module is used to calculate the liver function characteristic value of the test population based on the alanine aminotransferase level value, the total bilirubin level value and the albumin level value,

[0035] W=w1×ALT+w2×TBIL+w3×ALB,

[0036] Wherein, W is the liver function characteristic value of the test population, w1 is the alanine aminotransferase weight coefficient, ALT is the alanine aminotransferase level value, w2 is the total bilirubin weight coefficient, TBI L is the total bilirubin level value, w3 is the albumin weight coefficient, ALB is the albumin level value;

[0037] A risk identification and early warning method based on clinical trials, comprising:

[0038] Step S1, collecting the serious adverse event rate of the clinical trial drug, basic information of the test population, and examination data of the test population, the basic information of the test population includes age, duration of illness, and BMI, and the examination data of the test population includes hormone levels, alanine aminotransferase levels, total bilirubin levels, and albumin levels;

[0039] Step S2: The risk identification unit determines the serious adverse event rate according to the standard early warning evaluation, determines the risk level, which includes low risk level, medium risk level and high risk level, and outputs a warning of the risk level;

[0040] Step S3, the standard warning evaluation is corrected by the evaluation correction unit according to the sample size of the subject population and the evaluation amount of the standard sample, and the basic characterization value of the subject population, the liver function characterization value of the subject population and the hormone level value of any sample in the subject population sample and the corresponding preset abnormal threshold are judged, and the abnormal threshold includes the basic characterization value threshold, the liver function characterization value threshold and the hormone level value threshold, to determine whether each sample is an abnormal sample, and check the proportion of abnormal samples to determine whether to correct the evaluation amount of the standard sample.

[0041] Compared with existing technologies, the present invention has the beneficial effect of timely identifying potential safety issues through real-time monitoring and analysis of data from the test population, and adjusting risk assessment parameters based on the actual sample size and the proportion of abnormal samples, thereby improving the accuracy of early warnings. In addition, by comparing the system with the trial data of similar drugs in a large database, it can identify risk patterns that may have been overlooked, thereby enhancing the ability to monitor drug safety. This comparison not only helps to identify problems in a timely manner, but also can issue extraordinary early warnings when necessary to ensure the safety of subjects and the integrity of the trial.

[0042] Furthermore, by setting clear thresholds, the system can quickly respond to and address potential risks, thereby protecting the safety of subjects. This approach also helps improve the efficiency and quality of clinical trials by identifying potential issues at an early stage, avoiding waste of resources and ensuring the reliability of trial results. Furthermore, by comparing data from similar drugs, the system can provide more comprehensive safety assessments, helping researchers and regulators make more informed decisions.

[0043] Furthermore, by using the Mann-Whitney U model to scientifically compare data from clinical trial drugs with similar drugs, the system can more accurately assess the drug's risk level. The advantage of this approach is that it does not rely on the data's distribution pattern and is therefore applicable to various types of datasets. Furthermore, by setting a standard difference as a threshold, the system can objectively determine whether an early warning is necessary, thereby avoiding the potential bias caused by human subjective judgment. This approach helps to promptly identify potential risks that may have been overlooked, while reducing unnecessary early warnings, ensuring the effective use of resources, and improving the overall safety and efficiency of clinical trials.

[0044] Furthermore, through the dynamic adjustment mechanism of the evaluation correction unit, the early warning system can adapt to the statistical differences brought about by different sample sizes, thereby improving the accuracy and reliability of risk assessment. This adaptive adjustment helps avoid false negative results caused by too small sample sizes, while also reducing false positive results caused by too large sample sizes. In addition, by dynamically adjusting the evaluation event rate, the system can more accurately identify and warn of potential risks, which is crucial for protecting the safety of subjects and ensuring the scientific integrity of clinical trials.

[0045] Furthermore, by comparing key physiological indicators against pre-set thresholds, the system can promptly identify samples that may indicate a subject is at risk. Abnormal flagging not only facilitates rapid response and resolution of potential medical issues, protecting the health of subjects, but also improves the quality of clinical trial data.

[0046] Furthermore, by identifying outliers, the system ensures data accuracy and reliability, thereby improving the validity of clinical trial results. This approach helps promptly identify and eliminate outliers that could affect trial conclusions, reducing data bias, ensuring subject safety, and providing researchers with clearer and more accurate datasets for analysis.

[0047] Furthermore, by setting a correction sample ratio and adjusting the standard sample evaluation range accordingly, the evaluation correction unit enhances the system's adaptability and flexibility to abnormal situations. This approach helps maintain the quality and integrity of clinical trial data and ensures that risk assessments are not affected by accidental fluctuations in abnormal values. Furthermore, timely adjustments to the evaluation range can reduce the risk of false positive and false negative warnings, improving the accuracy and credibility of the early warning system.

[0048] Furthermore, by calculating baseline characterization values ​​for the trial population, the early warning system can more comprehensively assess the health of the subjects and, based on this, predict the risks they may face in clinical trials. This approach helps identify subjects who may respond poorly to the trial drug or have other health risks. Furthermore, the calculation of baseline characterization values ​​can provide researchers with a macro perspective on the overall health of the trial population, enabling more informed decisions regarding trial design and risk management.

[0049] Furthermore, by integrating multiple indicators, the liver function characterization value provides a comprehensive view of liver function, helping to detect possible liver problems early, allowing timely prevention or intervention measures to protect the health of subjects. In addition, this approach also improves the objectivity and accuracy of risk assessment in clinical trials, helping to generate more reliable and valid trial results.

[0050] Furthermore, through continuous data collection and dynamic risk assessment, potential safety issues in clinical trials can be promptly identified. By setting risk levels and outputting early warnings, rapid responses and measures can be taken to protect subject safety and trial integrity. Furthermore, the introduction of an evaluation and correction unit enhances the system's adaptability and accuracy, ensuring that risk assessments can be adjusted accordingly based on actual sample size and abnormalities as the trial progresses. This not only improves the relevance and accuracy of risk assessments, but also helps optimize clinical trial design and execution, increasing the success rate of drug development. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 This is a schematic diagram of the structure of the risk identification and early warning system based on clinical trials in this embodiment;

[0052] Figure 2 This is a logic diagram for determining whether to issue a risk warning in this embodiment;

[0053] Figure 3 This is a logic diagram for determining whether to issue a risk warning in this embodiment;

[0054] Figure 4 This is a decision logic diagram for determining whether to perform correction in this embodiment. DETAILED DESCRIPTION

[0055] In order to make the objects and advantages of the present invention more clearly understood, the present invention is further described below in conjunction with embodiments; it should be understood that the specific embodiments described herein are merely used to explain the present invention and are not intended to limit the present invention.

[0056] The preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood by those skilled in the art that these embodiments are only used to explain the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0057] It should be noted that, in the description of the present invention, terms such as "up", "down", "left", "right", "inside", and "outside" indicating directions or positional relationships are based on the directions or positional relationships shown in the accompanying drawings. This is only for the convenience of description and does not indicate or imply that the device or element must have a specific orientation, be constructed and operated in a specific orientation. Therefore, it cannot be understood as a limitation on the present invention.

[0058] Furthermore, it should be noted that, in the description of the present invention, unless otherwise expressly specified or limited, the terms "mounted," "connected," and "connected" should be understood in a broad sense. For example, they may refer to fixed connections, detachable connections, or integral connections; mechanical connections or electrical connections; direct connections or indirect connections through an intermediate medium; and internal communication between two components. Those skilled in the art will understand the specific meanings of the above terms in the present invention based on specific circumstances.

[0059] See also Figure 1 As shown, it is a schematic diagram of the structure of the risk identification and early warning system based on clinical trials in this embodiment;

[0060] This embodiment provides a risk identification and early warning system based on clinical trials, including:

[0061] An acquisition and storage unit, comprising an input module and a calculation module, wherein the input module is used to receive and store externally input serious adverse event rates of clinical trial drugs, basic information of the subject population, and examination data of the subject population;

[0062] The basic information of the subjects included age, disease duration, and BMI;

[0063] The examination data of the subjects included hormone levels, alanine aminotransferase levels, total bilirubin levels, and albumin levels;

[0064] The calculation module is used to calculate the basic characterization value of the subject population according to age, length of illness and BMI, and calculate the liver function characterization value of the subject population according to the alanine aminotransferase level value, total bilirubin level value and albumin level value;

[0065] a risk identification unit connected to the acquisition and storage unit, for determining the rate of serious adverse events according to standard early warning evaluation and determining a risk level, wherein the risk level includes a low risk level, a medium risk level, and a high risk level;

[0066] An evaluation correction unit, connected to the risk identification unit, is used to correct the standard warning evaluation according to the sample size of the subject population and the evaluation amount of the standard sample, and determine whether each sample is an abnormal sample based on the basic characterization value of the subject population, the liver function characterization value of the subject population, and the hormone level value of any sample in the subject population and the corresponding preset abnormal threshold value, wherein the abnormal threshold value includes the basic characterization value threshold, the liver function characterization value threshold, and the hormone level value threshold, and check the proportion of abnormal samples to determine whether to correct the evaluation amount of the standard sample;

[0067] The data capture unit is connected to the evaluation and correction unit, and is used to capture the test data of similar drugs in the large database, compare the differences, and determine the direction of the differences based on the differences to determine whether to issue an extraordinary warning.

[0068] First, the data entry module of the acquisition and storage unit collects and stores drug serious adverse event rates, basic subject information (such as age, duration of illness, and BMI), and detailed examination data (such as hormone levels and liver function indicators) from clinical trials. The calculation module further processes this data to generate subject baseline and liver function indicators. The risk identification unit then evaluates the serious adverse event rate based on a pre-set standard early warning evaluation system to determine the drug's risk level, which is categorized as low, medium, or high. The evaluation and correction unit compares the sample size of the trial population with the standard sample size and makes necessary adjustments to the early warning evaluation. It also identifies abnormal samples by comparing the sample's baseline indicator values, liver function indicators, and hormone levels with pre-set abnormal thresholds. Furthermore, this unit examines the proportion of abnormal samples to determine whether adjustments to the standard sample evaluation size are necessary. The data capture unit captures trial data for similar drugs from a large database and uses statistical methods (such as the Mann-Whitney U model) to compare and analyze any differences. Based on the differences, the system determines whether an extraordinary early warning is necessary.

[0069] By monitoring and analyzing data from the trial population in real time, the system can promptly identify potential safety issues and adjust risk assessment parameters based on the actual sample size and the proportion of abnormal samples, thereby improving the accuracy of early warnings. Furthermore, by comparing data with trial data from similar drugs in a large database, the system can identify risk patterns that may have been overlooked, enhancing drug safety monitoring capabilities. This comparison not only helps to promptly identify issues but also enables the issuance of extraordinary early warnings when necessary to safeguard the safety of subjects and the integrity of the trial.

[0070] Please continue reading Figure 2 As shown, it is a decision logic diagram for determining whether to issue a risk warning in this embodiment;

[0071] Specifically, the risk identification unit is provided with a standard early warning evaluation, which includes a first evaluation event rate and a second evaluation event rate;

[0072] The risk identification unit can determine the adverse event rate entered into the collection and storage unit according to the standard early warning evaluation.

[0073] If the adverse event rate is lower than the first evaluation event rate, the risk identification unit will determine that the current clinical trial drug has low risk and will not issue a risk warning;

[0074] If the adverse event rate is greater than or equal to the first evaluation event rate and less than or equal to the second evaluation event rate, the risk identification unit will control the data capture unit to capture the test data of similar drugs in the large database and make a judgment to determine whether a risk warning is issued;

[0075] If the adverse event rate is greater than the second evaluation event rate, the risk identification unit determines to issue a risk warning.

[0076] The first evaluation event rate refers to a lower threshold for serious adverse event rates. When the actual adverse event rate is lower than this threshold, the drug is generally considered to be low-risk and does not require further risk warning;

[0077] The second evaluation event rate refers to a higher threshold for serious adverse events. When the actual adverse event rate exceeds this threshold, the drug is considered high-risk and requires an immediate risk warning.

[0078] It all depends on the purpose and design of the clinical trial. Different drugs and indications may have different risk tolerances; regulatory requirements. Regulators may provide guidance or requirements on adverse event rates; historical data and safety data of similar drugs. Thresholds are set with reference to historical data and adverse event rates of similar drugs; statistical principles ensure that there is sufficient statistical power to detect actual risks; ethical considerations ensure the safety of subjects and avoid unnecessary risks, which are usually determined by the designers and statisticians of the clinical trial based on specific circumstances.

[0079] Suppose, in a clinical trial of a new drug, based on historical data and safety assessments of similar drugs, the research team decides to set the primary evaluation event rate at 2% and the secondary evaluation event rate at 5%. During the trial, if the adverse event rate is below 2%, the drug is considered safe and no risk warning is required. If the adverse event rate is between 2% and 5%, the system triggers a data capture unit to compare data with similar drugs in a large database to determine whether a warning is needed. If the adverse event rate is above 5%, the system immediately issues a high-risk warning, prompting researchers to take appropriate measures.

[0080] Through this setup, research teams can ensure that they are neither overly sensitive nor ignore real risks in drug safety assessments, thereby finding a balance between protecting the safety of subjects and advancing drug development.

[0081] The risk level of a drug in a clinical trial is assessed by setting two key parameters: the first evaluation event rate and the second evaluation event rate. When the adverse event rate collected by the data entry module is lower than the first evaluation event rate, the system determines that the drug risk is low and does not trigger a risk warning. If the adverse event rate is between the first and second evaluation event rates, the system activates the data capture unit to obtain trial data of similar drugs from a large database for comparative analysis to further determine whether an early warning is necessary. When the adverse event rate is higher than the second evaluation event rate, the system directly determines it as high risk and issues a risk warning.

[0082] By setting clear thresholds, the system can quickly respond to and address potential risks, thereby protecting the safety of subjects. This approach also helps improve the efficiency and quality of clinical trials by identifying potential issues at an early stage, avoiding waste of resources and ensuring the reliability of trial results. Furthermore, by comparing data from similar drugs, the system can provide more comprehensive safety assessments, helping researchers and regulators make more informed decisions.

[0083] Please continue reading Figure 3 As shown, it is a decision logic diagram for determining whether to further determine whether to issue a risk warning in this embodiment;

[0084] Specifically, the data capture unit is provided with a standard difference degree. If the adverse event rate is greater than or equal to the first evaluation event rate and less than or equal to the second evaluation event rate, the data capture unit will capture the test data of similar drugs in the large database and input the test data results into the Mann-Whitney U model. The Mann-Whitney U model compares the serious adverse event rates of the clinical trial drugs and similar drugs, the basic information of the test population, and the examination data of the test population to obtain the same type difference degree. The data capture unit compares the standard difference degree with the same type difference degree.

[0085] If the difference between the same category is less than or equal to the standard difference, no risk warning will be issued;

[0086] If the difference between the same category is greater than the standard difference, a risk warning will be issued.

[0087] When the adverse event rate in a clinical trial falls within the defined range of the first and second evaluation event rates, the data capture unit is activated. This unit is responsible for retrieving and capturing trial data for drugs similar to the current trial drug from a large database. This data is then input into the Mann-Whitney U model, a nonparametric statistical method used to compare the median difference between two independent samples. The model compares the adverse event rate, basic information about the study population, and examination data between the current drug and similar drugs to calculate the degree of similarity. The data capture unit compares this degree of similarity with a preset standard degree of similarity. If the degree of similarity does not exceed the standard degree of similarity, the system determines that the risk of the current drug is within an acceptable range and does not trigger a risk warning. Conversely, if the degree of similarity exceeds the standard degree of similarity, the system deems the risk significant and warrants a risk warning.

[0088] By using the Mann-Whitney U model to scientifically compare data from clinical trial drugs with similar drugs, the system can more accurately assess a drug's risk level. This approach is advantageous because it is independent of the data's distribution and can therefore be applied to a wide variety of datasets. Furthermore, by setting a threshold for standard deviation, the system can objectively determine whether an early warning is necessary, thus avoiding potential biases introduced by subjective judgment. This approach helps promptly identify potential risks that might otherwise be overlooked, while reducing unnecessary early warnings, ensuring efficient resource utilization, and improving the overall safety and efficiency of clinical trials.

[0089] Please continue reading Figure 4 As shown, it is a decision logic diagram for determining whether to perform correction in this embodiment;

[0090] Specifically, the evaluation correction unit is provided with a standard sample evaluation range. When the acquisition storage unit collects data, the evaluation correction unit can obtain the sample size of the test population to correct the standard warning evaluation in the risk identification unit, and compare the sample size of the test population with the standard sample evaluation range.

[0091] If the sample size of the test population is within the standard sample evaluation range, the evaluation correction unit determines not to perform correction;

[0092] If the sample size of the test population is lower than the minimum value of the standard sample evaluation range, the evaluation correction unit reduces the second evaluation event rate, wherein E2'=E2×[1-kd×(Nmin-N) / Nmin], E2' is the reduced second evaluation event rate, E2 is the preset second evaluation event rate, kd is the evaluation event rate reduction adjustment coefficient, Nmin is the minimum value of the standard sample evaluation range, and N is the sample size of the test population;

[0093] If the sample size of the subject population is higher than the maximum value of the standard sample evaluation range, the evaluation correction unit increases the first evaluation event rate, wherein E1'=E1×[1+ku×(N-Nmax) / (N-Nmin)], E1' is the increased first evaluation event rate, E1 is the preset first evaluation event rate, ku is the evaluation event rate increase adjustment coefficient, and Nmax is the maximum value of the standard sample evaluation range.

[0094] The adjustment factor for decreased event rate (kd) and the adjustment factor for increased event rate (ku) are used to adjust risk assessment parameters based on the comparison of the sample size of the test population with the standard sample evaluation range. The specific values ​​of these factors depend on a variety of factors, including but not limited to:

[0095] The objectives and requirements of the clinical trial, i.e., determining the risk tolerance based on the purpose and importance of the trial;

[0096] Statistical principles, i.e. determining the appropriate adjustment range based on statistical principles and power analysis;

[0097] Historical data, that is, referring to historical data of similar experiments to estimate the adjustment factor;

[0098] Expert opinion, which is determined based on the advice of clinical trial design experts and statisticians;

[0099] Ethical and regulatory requirements, i.e. ensuring that the adjustment factors comply with ethical and regulatory standards;

[0100] Risk management strategy, i.e., determining the adjustment factor based on the risk management strategy to balance sensitivity and specificity;

[0101] In some cases, the evaluation event rate decrease adjustment coefficient kd and the evaluation event rate increase adjustment coefficient ku will start from smaller coefficients, for example kd = 0.05 and ku = 0.05, which means that for every 1% decrease or increase in sample size, the second evaluation event rate or the first evaluation event rate will decrease or increase by 0.5%, respectively.

[0102] In this embodiment, assuming that the preset second evaluation event rate E2=3%, the minimum value of the standard sample evaluation range Nmin=100, and the sample size of the test population N=80, then:

[0103] E2'=3%×[1-0.1×(100-80) / 100]=2.64%;

[0104] Assuming that the preset first evaluation event rate E1 = 1%, the maximum value of the standard sample evaluation range Nmax = 200, and the sample size of the test population N = 220, then:

[0105] E1'=1%×[1+0.2×(220-200) / (220-100)]=1.1%;

[0106] In this way, the system can dynamically adjust risk assessment parameters based on the sample size to ensure the accuracy and effectiveness of risk warnings.

[0107] The evaluation and correction unit is responsible for dynamically adjusting the risk assessment parameters to adapt to changes in the sample size of the subject population. When the acquisition and storage unit collects new clinical trial data, the evaluation and correction unit will check the sample size of the subject population and compare it with the preset standard sample evaluation range (defined by the minimum value Nmin and the maximum value Nmax). If the sample size is within this range, the evaluation and correction unit decides not to adjust the early warning evaluation. If the sample size is lower than Nmin, the unit will reduce the second evaluation event rate E2 according to the adjustment coefficient kd to reduce the risk assessment bias caused by the small sample size. On the contrary, if the sample size is higher than Nmax, the unit will increase the first evaluation event rate E1 and increase the sensitivity of the risk assessment according to the adjustment coefficient ku. This adjustment strategy ensures that the risk assessment can be appropriately adjusted according to the size of the sample to maintain the accuracy and consistency of the assessment.

[0108] Through the dynamic adjustment mechanism of the evaluation correction unit, the early warning system can adapt to the statistical differences brought about by different sample sizes, thereby improving the accuracy and reliability of risk assessment. This adaptive adjustment helps avoid false negative results caused by too small sample sizes, while also reducing false positive results caused by too large sample sizes. In addition, by dynamically adjusting the evaluation event rate, the system can more accurately identify and warn of potential risks, which is crucial for protecting the safety of subjects and ensuring the scientific integrity of clinical trials.

[0109] Specifically, the evaluation and correction unit is also provided with a basic characterization value threshold, a liver function characterization value threshold and a hormone level value threshold. The evaluation and correction unit compares the basic characterization value of the subject population of any sample with the basic characterization value threshold, and compares the liver function characterization value of the subject population with the liver function characterization value threshold, and compares the hormone level value with the hormone level value threshold, and determines whether the sample is abnormal based on the comparison results, and marks the abnormal sample as abnormal when it is determined to be abnormal, forming an abnormal sample.

[0110] The abnormality of the samples from the test population is evaluated through a series of set thresholds. These thresholds include basic characterization value thresholds, liver function characterization value thresholds, and hormone level value thresholds. When the acquisition storage unit provides new data, the evaluation and correction unit will compare the basic characterization value, liver function characterization value, and hormone level value of the test population of each sample with the corresponding thresholds one by one. If these indicators of the sample exceed the preset thresholds, the evaluation and correction unit will determine the sample as abnormal and mark it as abnormal, thus forming a collection of abnormal samples. This process helps to identify samples that may indicate potential health risks or do not meet the test requirements.

[0111] By comparing key physiological indicators against pre-set thresholds, the system can promptly identify samples that may indicate a subject is at risk. Abnormal flagging not only helps quickly respond to and address potential medical issues, protecting the health of subjects, but also improves the quality of clinical trial data.

[0112] Specifically, the evaluation correction unit determines whether the sample is abnormal based on the comparison result.

[0113] If the basic characterization value of the test population, the liver function characterization value of the test population, and the hormone level value are all greater than the corresponding abnormal threshold value, the evaluation correction unit determines that the sample is abnormal.

[0114] The test population's basic characteristic values ​​(such as a composite index of age, duration of illness, and BMI), liver function characteristic values ​​(calculated based on alanine aminotransferase, total bilirubin, and albumin levels), and hormone levels are compared with their respective abnormal thresholds. When all of these key indicators for a sample exceed their corresponding abnormal thresholds, the evaluation and correction unit will integrate this information and determine that the sample is abnormal.

[0115] By identifying outliers, the system ensures data accuracy and reliability, thereby improving the validity of clinical trial results. This approach helps promptly detect and exclude outliers that may affect trial conclusions, reduces data bias, ensures subject safety, and provides researchers with clearer and more accurate data sets for analysis.

[0116] Specifically, the evaluation correction unit is further provided with a correction sample ratio. When the evaluation correction unit completes the abnormality determination of all samples, it calculates the abnormality ratio according to the number of abnormal samples and compares the abnormality ratio with the correction sample ratio.

[0117] If the abnormal proportion is greater than the corrected sample proportion, the evaluation correction unit will adjust the standard sample evaluation range according to the corrected sample proportion.

[0118] in, is the minimum value of the adjusted standard sample evaluation range, X is the corrected sample proportion, and Nmax' is the maximum value of the adjusted standard sample evaluation range.

[0119] After completing the abnormality determination for all samples, the evaluation and correction unit will enter an additional analysis phase. In this phase, the unit will calculate the ratio of the number of abnormal samples to the total number of samples, that is, the abnormality ratio. The evaluation and correction unit will then compare this abnormality ratio with the preset corrected sample ratio. If the actual abnormality ratio exceeds the corrected sample ratio, it means that the number of abnormal samples is greater than expected, which may indicate that the current standard sample evaluation range is not strict enough or there are other problems. In this case, the evaluation and correction unit will adjust the standard sample evaluation range based on the corrected sample ratio to ensure that future risk assessments are more accurate and reliable.

[0120] By setting a correction sample ratio and adjusting the standard sample evaluation range accordingly, the evaluation correction unit enhances the system's adaptability and flexibility to abnormal situations. This practice helps maintain the quality and integrity of clinical trial data and ensures that risk assessments are not affected by accidental fluctuations in abnormal values. Furthermore, timely adjustments to the evaluation range can reduce the risk of false positive and false negative warnings, improving the accuracy and credibility of the early warning system.

[0121] Specifically, the calculation module is used to calculate the basic characterization values ​​of the test population based on age, disease duration and BMI.

[0122]

[0123] Among them, Q is the basic characterization value of the subject population, A is age, A0 is the average age, a is the standard deviation of age, B is the length of illness, B0 is the average length of illness, b is the standard deviation of length of illness, C is BMI, C0 is the average BMI, and c is the standard deviation of BMI.

[0124] The calculation module calculates a comprehensive indicator—the basic characterization value (Q)—based on the individual characteristics of the subjects. This calculation process involves the subjects' three basic physiological indicators: age (A), duration of illness (B), and BMI (C), as well as their mean and standard deviation. Specifically, the calculation module uses each subject's indicator value, its corresponding mean and standard deviation, to calculate the Q value using a specific mathematical formula. This Q value reflects the basic health status of the subjects and is a key factor in assessing their potential risks in clinical trials.

[0125] By calculating baseline characterization values ​​for the trial population, the early warning system provides a more comprehensive assessment of the subjects' health and, based on this, predicts the risks they may face in clinical trials. This approach helps identify subjects who may respond poorly to the trial drug or have other health risks. Furthermore, the calculation of baseline characterization values ​​provides researchers with a macro perspective on the overall health of the trial population, enabling more informed decisions regarding trial design and risk management.

[0126] Specifically, the calculation module is used to calculate the liver function characteristic value of the test population based on the alanine aminotransferase level value, the total bilirubin level value and the albumin level value.

[0127] W=w1×ALT+w2×TBIL+w3×ALB,

[0128] Wherein, W is the liver function characteristic value of the test population, w1 is the alanine aminotransferase weight coefficient, ALT is the alanine aminotransferase level value, w2 is the total bilirubin weight coefficient, TBI L is the total bilirubin level value, w3 is the albumin weight coefficient, ALB is the albumin level value;

[0129] Through a specific formula, combined with the subject's alanine aminotransferase (ALT), total bilirubin (TBIL) and albumin (ALB) levels, a comprehensive indicator - liver function characterization value (W) is calculated. During the calculation process, each indicator will be multiplied by its corresponding weight coefficient (w1, w2, w3). These weight coefficients reflect the importance and influence of different indicators in assessing liver function. In this way, the calculation module can comprehensively consider multiple liver function indicators and generate a single value representing the subject's overall liver function status.

[0130] By integrating multiple indicators, liver function characterization values ​​provide a comprehensive view of liver function, helping to identify potential liver problems early, allowing timely preventive or intervention measures to protect the health of subjects. Furthermore, this approach improves the objectivity and accuracy of risk assessments in clinical trials, helping to generate more reliable and valid trial results.

[0131] A risk identification and early warning method based on clinical trials, comprising:

[0132] Step S1, collecting the serious adverse event rate of the clinical trial drug, basic information of the test population, and examination data of the test population, the basic information of the test population includes age, duration of illness, and BMI, and the examination data of the test population includes hormone levels, alanine aminotransferase levels, total bilirubin levels, and albumin levels;

[0133] Step S2: The risk identification unit determines the serious adverse event rate according to the standard early warning evaluation, determines the risk level, which includes low risk level, medium risk level and high risk level, and outputs a warning of the risk level;

[0134] Step S3, the standard warning evaluation is corrected by the evaluation correction unit according to the sample size of the subject population and the evaluation amount of the standard sample, and the basic characterization value of the subject population, the liver function characterization value of the subject population and the hormone level value of any sample in the subject population sample and the corresponding preset abnormal threshold are judged, and the abnormal threshold includes the basic characterization value threshold, the liver function characterization value threshold and the hormone level value threshold, to determine whether each sample is an abnormal sample, and check the proportion of abnormal samples to determine whether to correct the evaluation amount of the standard sample.

[0135] First, in step S1, data collection is performed to collect clinical trial drug-related data, including serious adverse event rates, as well as basic information and examination data of the subjects. Basic information involves age, duration of illness, and BMI, while examination data includes key physiological parameters such as hormone levels and liver function indicators. In step S2, the serious adverse event rate is evaluated by the risk identification unit, and the risk level of the drug is determined based on a preset standard early warning evaluation system. This step corresponds the adverse event rate to the risk level, dividing it into three categories: low, medium, and high, and outputs corresponding early warning information based on the judgment results. Step S3 involves an evaluation correction unit, which adjusts the early warning evaluation based on the sample size of the subject population to ensure the accuracy of the evaluation. This unit also compares the various physiological indicators of the subject population samples with the preset abnormal threshold to identify abnormal samples, and determines whether the standard sample evaluation volume needs to be further revised based on the proportion of abnormal samples.

[0136] Through continuous data collection and dynamic risk assessment, potential safety issues in clinical trials can be promptly identified. By setting risk levels and outputting early warnings, rapid responses and measures can be taken to protect subject safety and trial integrity. Furthermore, the introduction of an evaluation and correction unit enhances the system's adaptability and accuracy, ensuring that risk assessments can be adjusted accordingly based on actual sample size and abnormalities as the trial progresses. This not only improves the relevance and accuracy of risk assessments, but also helps optimize clinical trial design and execution, increasing the success rate of drug development.

[0137] Thus far, the technical solutions of the present invention have been described in conjunction with the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art may make equivalent changes or substitutions to the relevant technical features, and the technical solutions after such changes or substitutions will fall within the scope of protection of the present invention.

[0138] The foregoing description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Those skilled in the art will readily appreciate that the present invention is susceptible to various modifications and variations. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention are intended to be within the scope of protection of the present invention.

Claims

1. A risk identification and early warning system based on clinical trials, characterized in that: include: An acquisition and storage unit, comprising an input module and a calculation module, wherein the input module is used to receive and store externally input serious adverse event rates of clinical trial drugs, basic information of the subject population, and examination data of the subject population; The basic information of the subjects included age, disease duration, and BMI; The examination data of the subjects included hormone levels, alanine aminotransferase levels, total bilirubin levels, and albumin levels; The calculation module is used to calculate the basic characterization values ​​of the test population based on age, disease duration and BMI: Among them, Q is the basic characterization value of the test population, A is age, A0 is the mean age, a is the standard deviation of age, B is the disease age, B0 is the mean age, b is the standard deviation of disease age, C is BMI, C0 is the mean BMI, and c is the standard deviation of BMI; The calculation module is used to calculate the liver function characteristic value of the test population according to the alanine aminotransferase level value, the total bilirubin level value and the albumin level value, Among them, W is the liver function characteristic value of the test population, w1 is the alanine aminotransferase weight coefficient, ALT is the alanine aminotransferase level value, w2 is the total bilirubin weight coefficient, TBIL is the total bilirubin level value, w3 is the albumin weight coefficient, ALB is the albumin level value; a risk identification unit connected to the acquisition and storage unit, configured to determine a risk level by judging a serious adverse event rate based on a standard early warning evaluation, wherein the standard early warning evaluation includes a first evaluation event rate and a second evaluation event rate, and the risk levels include a low risk level, a medium risk level, and a high risk level; An evaluation correction unit, connected to the risk identification unit, is used to correct the standard warning evaluation according to the sample size of the subject population and the evaluation amount of the standard sample, and determine whether each sample is an abnormal sample based on the basic characterization value of the subject population, the liver function characterization value of the subject population, and the hormone level value of any sample in the subject population and the corresponding preset abnormal threshold value, wherein the abnormal threshold value includes the basic characterization value threshold, the liver function characterization value threshold, and the hormone level value threshold, and check the proportion of abnormal samples to determine whether to correct the evaluation amount of the standard sample; The evaluation and correction unit determines whether the sample is abnormal based on the comparison results. If the basic characteristic value of the test population, the liver function characteristic value of the test population, and the hormone level value are all greater than the corresponding abnormal threshold value, the evaluation and correction unit determines that the sample is abnormal; The evaluation correction unit is further provided with a correction sample ratio. When the evaluation correction unit completes the abnormality determination of all samples, it calculates the abnormality ratio according to the number of abnormal samples and compares the abnormality ratio with the correction sample ratio. If the abnormality ratio is greater than the correction sample ratio, the evaluation correction unit will adjust the standard sample evaluation range according to the correction sample ratio. The evaluation correction unit is provided with a standard sample evaluation range. When the acquisition storage unit collects data, the evaluation correction unit can obtain the sample size of the test population to correct the standard warning evaluation in the risk identification unit, and compare the sample size of the test population with the standard sample evaluation range. If the sample size of the test population is within the standard sample evaluation range, the evaluation correction unit determines not to perform correction; If the sample size of the test population is lower than the minimum value of the standard sample evaluation range, the evaluation correction unit reduces and adjusts the second evaluation event rate; If the sample size of the test population is higher than the maximum value of the standard sample evaluation range, the evaluation correction unit increases the first evaluation event rate; A data capture unit is connected to the evaluation and correction unit, and is used to capture the test data of similar drugs in a large database, compare the differences, and determine the direction of the differences based on the differences to determine whether to issue an extraordinary warning; The data capture unit is provided with a standard difference degree. If the adverse event rate is greater than or equal to the first evaluation event rate and less than or equal to the second evaluation event rate, the data capture unit will capture the test data of similar drugs in the large database and input the test data results into the Mann-Whitney U model. The Mann-Whitney U model compares the serious adverse event rates of the clinical trial drugs and similar drugs, the basic information of the test population, and the examination data of the test population to obtain the same type difference degree. The data capture unit compares the standard difference degree with the same type difference degree. If the difference between the same category is less than or equal to the standard difference, no risk warning will be issued; If the difference between the same category is greater than the standard difference, a risk warning will be issued.

2. The clinical trial-based risk identification and early warning system according to claim 1, characterized in that: The risk identification unit is provided with a standard early warning evaluation, which includes a first evaluation event rate and a second evaluation event rate; The risk identification unit can determine the adverse event rate entered into the collection and storage unit according to the standard early warning evaluation. If the adverse event rate is lower than the first evaluation event rate, the risk identification unit will determine that the current clinical trial drug has low risk and will not issue a risk warning; If the adverse event rate is greater than or equal to the first evaluation event rate and less than or equal to the second evaluation event rate, the risk identification unit will control the data capture unit to capture the test data of similar drugs in the large database and make a judgment to determine whether a risk warning is issued; If the adverse event rate is greater than the second evaluation event rate, the risk identification unit determines to issue a risk warning.

3. The clinical trial-based risk identification and early warning system according to claim 1, characterized in that: The evaluation and correction unit is also provided with a basic characterization value threshold, a liver function characterization value threshold and a hormone level value threshold. The evaluation and correction unit compares the basic characterization value of the subject population of any sample with the basic characterization value threshold, and compares the liver function characterization value of the subject population with the liver function characterization value threshold, and compares the hormone level value with the hormone level value threshold, and determines whether the sample is abnormal based on the comparison result, and marks the abnormal sample as abnormal when it is determined to be abnormal, thereby forming an abnormal sample.

4. A method for risk identification and early warning based on clinical trials, using the risk identification and early warning system based on clinical trials according to any one of claims 1 to 3, characterized in that: The steps include: Step S1, collecting the serious adverse event rate of the clinical trial drug, basic information of the test population, and examination data of the test population, the basic information of the test population includes age, duration of illness, and BMI, and the examination data of the test population includes hormone levels, alanine aminotransferase levels, total bilirubin levels, and albumin levels; Step S2: The risk identification unit determines the serious adverse event rate according to the standard early warning evaluation, determines the risk level, which includes low risk level, medium risk level and high risk level, and outputs a warning of the risk level; Step S3, the standard warning evaluation is corrected by the evaluation correction unit according to the sample size of the subject population and the evaluation amount of the standard sample, and the basic characterization value of the subject population, the liver function characterization value of the subject population and the hormone level value of any sample in the subject population sample and the corresponding preset abnormal threshold are judged, and the abnormal threshold includes the basic characterization value threshold, the liver function characterization value threshold and the hormone level value threshold, to determine whether each sample is an abnormal sample, and check the proportion of abnormal samples to determine whether to correct the evaluation amount of the standard sample.

Images