Clinical test management system based on artificial intelligence

Abstract

The invention discloses a clinical test management system based on artificial intelligence, and the system comprises an acquisition unit which is used for collecting clinical test data of a control group and a test group, and carrying out the preprocessing of the data; the simulation surface construction unit is used for respectively constructing a reference bottom surface and a test side surface according to the clinical test data of the control group and the test group; the simulation surface connecting unit is used for connecting the test side surface above the reference bottom surface and adjusting the angle of the test side surface; the difference calculation unit is used for calculating the difference degree between the test side surface and the reference bottom surface; the comprehensive adjustment unit is used for adjusting the clinical test scheme according to the difference degree; based on a control group and a test group in a clinical test, a reference bottom surface and a test side surface are respectively constructed, a three-dimensional graph representing single clinical test data is formed, the difference between the test group and the control group can be accurately obtained through calculation in the three-dimensional graph, a clinical test scheme is adjusted based on the difference degree, closed-loop feedback is realized, and the clinical test accuracy is improved. And the clinical test effect and efficiency are improved.

Description

Technical Field

[0001] The present invention relates to the field of intelligent medical technology, and in particular to an artificial intelligence-based clinical trial management system. Background Art

[0002] In the field of clinical trial management, traditional methods rely on manual data collection, progress monitoring, risk assessment, and resource allocation. These methods suffer from low data processing efficiency, difficulty in cross-system integration, and delayed risk warning, resulting in extended trial cycles, high costs, and difficulty in quality control. With the popularization of electronic data capture, IoT devices, and wearable technology, the scale of multimodal data has grown exponentially, but existing systems lack efficient intelligent analysis capabilities, making it difficult to achieve deep data correlation and dynamic decision support. Although some systems have attempted to introduce rule engines or simple statistical models for process optimization, they are unable to cope with complex clinical scenarios. In particular, in the difference analysis between the control group and the experimental group, traditional methods can only complete basic mean difference and risk ratio calculations through manual statistics. It is difficult to use machine learning algorithms to explore potential differences between groups in multiple dimensions such as efficacy, safety, and biomarkers in real time, and it is even more impossible to automatically judge the actual significance of the differences and trigger regimen adjustments in combination with clinical guidelines. Summary of the Invention

[0003] In view of this, the present invention proposes an artificial intelligence-based clinical trial management system, which can adjust the clinical trial plan based on the differences between the control group and the experimental group, thereby improving the accuracy of the clinical trial results.

[0004] The technical solution of the present invention is achieved as follows:

[0005] An artificial intelligence-based clinical trial management system, comprising:

[0006] The collection unit is used to collect clinical trial data of the control group and the experimental group and perform preprocessing;

[0007] A simulation surface construction unit, used to construct a reference bottom surface and a test side surface according to the clinical trial data of the control group and the test group respectively;

[0008] A simulation surface connection unit is used to connect the test side surface to the top of the reference bottom surface and adjust the angle of the test side surface;

[0009] A difference calculation unit, used to calculate the difference between the test side surface and the reference bottom surface;

[0010] Comprehensive adjustment unit, used to adjust clinical trial plans based on the degree of variability;

[0011] The acquisition unit, the simulation surface construction unit, the simulation surface connection unit, the difference calculation unit and the comprehensive adjustment unit are sequentially data-connected, and the simulation surface construction unit is data-connected with the simulation surface connection unit.

[0012] Preferably, the execution steps of the acquisition unit include:

[0013] Step S11: extract key features from the clinical trial protocol, input the key features into the trained neural network, and process the neural network to obtain the indicators to be monitored;

[0014] Step S12: Acquire clinical trial data of the control group and the experimental group according to the indicators to be monitored, and perform data cleaning, denoising, and standardization on the clinical trial data.

[0015] Preferably, the training steps of the neural network are: obtaining a large number of historical clinical trial plans and historical monitoring indicators collected during the trial, performing feature extraction on the historical clinical trial plans, and obtaining historical key features, taking the historical key features and the corresponding historical monitoring indicators as a set of data, obtaining a training set and a test set by dividing, training the neural network with the training set, and testing it with the test set after the training is completed.

[0016] Preferably, the execution steps of the simulation surface construction unit include:

[0017] Step S21: randomly select a clinical trial data as a construction index, and map the construction index of the control group to the area data of the control group;

[0018] Step S22: construct a reference base of a regular polygon according to the number of experimental groups, adjust the size of the reference base based on the area data of the control group, and obtain the side length of the reference base;

[0019] Step S23: construct a rectangular test side surface for each test group, with the bottom side of the test side surface being consistent with the side length of the reference bottom surface, mapping the construction index of the test group to the test group area data, and adjusting the height of the test side surface according to the test group area data;

[0020] Step S24: Repeat steps S21-S23 until all clinical trial data are selected, and take the reference bottom surface and the test side surface corresponding to each construction indicator as a group.

[0021] Preferably, if there are multiple control groups, the clinical trial data of the control groups are averaged.

[0022] Preferably, the execution steps of the simulated surface connection unit include:

[0023] Step S31: establishing a virtual multi-dimensional space, and mapping the reference bottom surface and the test side surface into the virtual multi-dimensional space;

[0024] Step S32: align the bottom edge of each test side surface with the edge of the reference bottom surface, and keep each test side surface perpendicular to the reference bottom surface;

[0025] Step S33: Compare the same clinical trial data of the test group and the control group, obtain the difference, and adjust the angle of the test side based on the difference.

[0026] Preferably, the specific steps of step S33 are:

[0027] Step S331: Evaluate the positive and negative variability of the selected clinical trial data, subtract the clinical trial data of the experimental group from the clinical trial data of the control group, obtain the difference, and map the difference to a rotation angle;

[0028] Step S332: Determine whether the positive and negative signs of the difference are consistent with the positive and negative variability of the clinical trial data. If they are consistent, rotate the test side toward the center of the reference bottom surface according to the rotation angle. If they are inconsistent, rotate the test side toward the direction away from the reference bottom surface according to the rotation angle.

[0029] Preferably, the execution steps of the differential calculation unit include:

[0030] Step S41, respectively obtaining the center of the reference bottom surface, the center of the test side surface, and the center of the top edge of the test side surface;

[0031] Step S42, calculating a first Euclidean distance from the center of the reference bottom surface to the center of the test side surface and a second Euclidean distance from the center of the reference bottom surface to the center of the top edge of the test side surface;

[0032] Step S43: Obtain the difference between the test side surface and the reference bottom surface based on the first Euclidean distance and the second Euclidean distance.

[0033] Preferably, the specific steps of step S43 are: averaging the first Euclidean distance and the second Euclidean distance, and outputting the average value as the difference.

[0034] Preferably, the execution steps of the comprehensive adjustment unit are: comparing the difference with the expected difference in the initial clinical trial plan, and adjusting the test conditions of the test group patients based on the comparison results.

[0035] Compared with the prior art, the present invention has the following beneficial effects:

[0036] The present invention provides an artificial intelligence-based clinical trial management system, which can collect clinical trial data of a control group and a test group through an acquisition unit, and then construct a reference bottom surface and a test side surface for the control group and the test group respectively. The reference bottom surface is used as a basis, and the test side surface is connected to the top of the reference bottom surface. The angle of the test side surface can be adjusted uniformly, and finally a three-dimensional graph including a single reference bottom surface and multiple test side surfaces can be formed for each clinical trial data. Each clinical trial data will be constructed to obtain a corresponding three-dimensional graph, and then the difference between each test side surface and the reference bottom surface is calculated in the three-dimensional graph. Finally, the clinical trial plan can be adjusted according to the difference. By constructing simulation surfaces for the test group and the control group and converting them into spatial dimensions, it is convenient for management personnel to quickly determine the differences between each test group and the control group under the same clinical trial data, so as to quickly formulate adjustment plans and improve the efficiency and effectiveness of clinical trials. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only preferred embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0038] Figure 1 This is a schematic diagram of a clinical trial management system based on artificial intelligence of the present invention;

[0039] Figure 2 A diagram showing the execution steps of a collection unit of an artificial intelligence-based clinical trial management system of the present invention;

[0040] Figure 3 A diagram showing the execution steps of a simulation interface construction unit of an artificial intelligence-based clinical trial management system according to the present invention;

[0041] Figure 4 A diagram showing the execution steps of a simulation interface connection unit of an artificial intelligence-based clinical trial management system of the present invention;

[0042] Figure 5 Detailed step diagram of step S33 of an artificial intelligence-based clinical trial management system of the present invention;

[0043] Figure 6 A diagram of the execution steps of a differentiated computing unit of an artificial intelligence-based clinical trial management system of the present invention;

[0044] In the figure, 1. Acquisition unit; 2. Simulation surface construction unit; 3. Simulation surface connection unit; 4. Differentiation calculation unit; 5. Comprehensive adjustment unit. DETAILED DESCRIPTION

[0045] In order to better understand the technical content of the present invention, a specific embodiment is provided below, and the present invention is further described in conjunction with the accompanying drawings.

[0046] See also Figures 1 to 6 The present invention provides an artificial intelligence-based clinical trial management system, comprising:

[0047] Collection unit 1, used to collect clinical trial data of the control group and the experimental group and perform preprocessing;

[0048] A simulation surface construction unit 2 is used to construct a reference bottom surface and a test side surface according to the clinical trial data of the control group and the test group respectively;

[0049] The simulation surface connection unit 3 is used to connect the test side surface above the reference bottom surface and adjust the angle of the test side surface;

[0050] a difference calculation unit 4, for calculating the difference between the test side surface and the reference bottom surface;

[0051] Comprehensive adjustment unit 5, used to adjust the clinical trial plan according to the degree of difference;

[0052] The acquisition unit 1 , the simulation surface construction unit 2 , the simulation surface connection unit 3 , the difference calculation unit 4 and the comprehensive adjustment unit 5 are sequentially data-connected, and the simulation surface construction unit 2 is data-connected to the simulation surface connection unit 3 .

[0053] The present invention provides an artificial intelligence-based clinical trial management system for use in clinical trials, wherein clinical trials are divided into different phases. Different patients or users are administered medication or use of devices at different phases, and the effectiveness of the medication or device is evaluated based on the patient's or user's response. To determine whether the medication or device is indeed effective, a control group and a test group are set up during the clinical trial. The patients in the control and test groups are in substantially the same environment and have similar physical conditions, ages, etc. After patients or users are recruited through community-directed push or voluntary recruitment, they can be divided into a control group and a test group. The patients or users in the control group do not receive any medication or device treatment, while the patients or users in the test group are given different medication doses or different device usage amounts. After a period of time, the corresponding physiological indicators of the test and control groups can be evaluated to assess the effectiveness of the medication or device. However, the current evaluation process mostly relies on experienced medical staff, and it is impossible to quickly and accurately obtain the correlation and differences between the control and test groups to adjust the clinical trial plan, resulting in a failure to improve the efficiency of clinical trials.

[0054] During the clinical trial process of the present invention, after the clinical trial data of the control group and the experimental group are collected by the acquisition unit 1, the clinical trial data are preprocessed to ensure the accuracy of the data and the uniformity of the dimensions. Then, the simulation surface construction unit 2 can respectively construct the reference bottom surface and the test side surface, the reference bottom surface corresponds to the control group, and the test side surface corresponds to the experimental group. The simulation surface connection unit 3 connects the test side surface to the top of the reference bottom surface, and at the same time adjusts the angle of the test side surface so that the test side surface offsets the reference bottom surface, that is, it can simulate the differences caused by different experimental conditions between the experimental group and the control group, and obtain a three-dimensional graph including a reference bottom surface and multiple test side surfaces. There are many types of clinical trial data, and a separate three-dimensional graph will be formed for each type of clinical trial data. The difference calculation unit 4 can calculate the difference between each test side surface in each three-dimensional graph and the reference bottom surface. The difference degree is used to describe the difference between the experimental group and the control group. Finally, according to different difference degrees, the clinical trial plan can be adjusted by the comprehensive adjustment unit 5 to achieve dynamic feedback adjustment and improve the efficiency and effect of the clinical trial.

[0055] Preferably, the execution steps of the acquisition unit 1 include:

[0056] Step S11: extract key features from the clinical trial protocol, input the key features into the trained neural network, and process the neural network to obtain the indicators to be monitored;

[0057] Step S12: Acquire clinical trial data of the control group and the experimental group according to the indicators to be monitored, and perform data cleaning, denoising, and standardization on the clinical trial data.

[0058] Before conducting a clinical trial, a complete clinical trial plan will be formed. The clinical trial plan includes the purpose of the trial, the trial method, the trial conditions, the number of selected control groups, the number of experimental groups, the number of patients in each group, the basic information of each patient, etc. In addition, it also includes the goals to be achieved in the clinical trial and the type of data that needs to be collected to verify the effect. Natural language processing methods are used to extract key features from the clinical trial plan, and then the key features are input into the trained neural network. The neural network processes the indicators to be tested, and then the clinical trial data of the control group and the experimental group can be queried according to the indicators to be tested, and the clinical trial data are preprocessed. The preprocessing includes data cleaning, denoising and standardization. Data cleaning can be used to process missing values, remove outliers, etc., while denoising can remove noise contained in the data, and standardization can unify data coding and units.

[0059] Preferably, the training steps of the neural network are: obtaining a large number of historical clinical trial plans and historical monitoring indicators collected during the trial, performing feature extraction on the historical clinical trial plans, and obtaining historical key features, taking the historical key features and the corresponding historical monitoring indicators as a set of data, obtaining a training set and a test set by dividing, training the neural network with the training set, and testing it with the test set after the training is completed.

[0060] The role of the neural network is to quickly identify the indicators to be monitored in clinical trials, and then query the corresponding clinical trial data based on the indicators to be monitored. After collecting a large amount of historical data, the present invention divides it into a training set and a test set in a ratio of 7:3, and then trains with the training set and tests with the test set. When the test accuracy meets the requirements, the training is stopped. At this time, the key features of the clinical trial plan can be input into the neural network to extract the indicators to be monitored.

[0061] Preferably, the execution steps of the simulation surface construction unit 2 include:

[0062] Step S21: determine whether the number of control groups is plural. If the number of control groups is plural, average the clinical trial data of the control groups, randomly select one clinical trial data as a construction index, and map the construction index of the control group to the area data of the control group.

[0063] Step S22: construct a reference base of a regular polygon according to the number of experimental groups, adjust the size of the reference base based on the area data of the control group, and obtain the side length of the reference base;

[0064] Step S23: construct a rectangular test side surface for each test group, with the bottom side of the test side surface being consistent with the side length of the reference bottom surface, mapping the construction index of the test group to the test group area data, and adjusting the height of the test side surface according to the test group area data;

[0065] Step S24: Repeat steps S21-S23 until all clinical trial data are selected, and take the reference bottom surface and the test side surface corresponding to each construction indicator as a group.

[0066] The simulation surface construction unit 2 is used to construct a reference bottom surface and a test side surface. The number of reference bottom surfaces is one, while the number of control groups is not necessarily one group and may be multiple. Therefore, before constructing the simulation surface, the number of control groups is first determined. If there are multiple control groups, the clinical trial data of all control groups are averaged and the averaged clinical trial data is used as the clinical trial data of a single control group. Then, one is randomly selected from the clinical trial data as a construction index. The specific value of the construction index of the control group is mapped into the area data of the control group. Then, the reference bottom surface can be constructed. The reference bottom surface is a regular polygon with the number of sides equal to the number of test groups. Given the number of sides and the area data of the control group, the reference bottom surface can be accurately constructed. The length of each side of the reference bottom surface can also be calculated accordingly. Then, the test side surface needs to be constructed. The test side surface is a rectangle with the base length consistent with the side length of the reference bottom surface. After mapping the construction index of the test group into the area data of the test group, the height of the test side surface can be obtained. Repeating this process with different control groups and different clinical trial data can obtain a three-dimensional graph containing a single reference bottom surface and multiple test side surfaces, each of which corresponds to different clinical trial data.

[0067] Preferably, the execution steps of the simulated surface connection unit 3 include:

[0068] Step S31: establishing a virtual multi-dimensional space, and mapping the reference bottom surface and the test side surface into the virtual multi-dimensional space;

[0069] Step S32: align the bottom edge of each test side surface with the edge of the reference bottom surface, and keep each test side surface perpendicular to the reference bottom surface;

[0070] Step S33: Compare the same clinical trial data of the test group and the control group, obtain the difference, and adjust the angle of the test side based on the difference.

[0071] After constructing the reference bottom surface and the test side surface, the test side surface needs to be connected to the reference point. After establishing the virtual multi-dimensional space, the reference point and the test side surface are mapped into the virtual multi-dimensional space, and then the bottom edge of each test side surface is directly connected to the edge of the reference bottom surface. Initially, each test side surface is perpendicular to the reference bottom surface. Then, based on the clinical trial data corresponding to each three-dimensional graph, the difference in clinical experimental data between the experimental group and the control group is calculated. The angle of the test side surface can be adjusted according to the difference, and differentiated graphs between different experimental groups and control groups can be formed.

[0072] Preferably, the specific steps of step S33 are:

[0073] Step S331: Evaluate the positive and negative variability of the selected clinical trial data, subtract the clinical trial data of the experimental group from the clinical trial data of the control group, obtain the difference, and map the difference to a rotation angle;

[0074] Step S332: Determine whether the positive and negative signs of the difference are consistent with the positive and negative variability of the clinical trial data. If they are consistent, rotate the test side toward the center of the reference bottom surface according to the rotation angle. If they are inconsistent, rotate the test side toward the direction away from the reference bottom surface according to the rotation angle.

[0075] When adjusting the angle of the test side, the test side needs to be rotated with the bottom edge as the axis, and the direction of rotation is not arbitrary, but needs to be determined based on the clinical trial data. First, the positive and negative variability of the selected clinical trial data is evaluated, that is, after the drug is administered or the device is applied in the test group, is the ideal state of the clinical trial data positively or negatively changed relative to the control group? After determining the positive and negative variability, the clinical trial data of the test group and the clinical trial data of the control group are subtracted to obtain the difference, and the difference is determined to be positive or negative, and the positive and negative signs of the difference can be obtained. The positive and negative signs are compared with the positive and negative variability. If the changes are consistent, the test side is rotated toward the reference bottom surface. If the changes are inconsistent, the test side is rotated to the other side. For example, when the clinical trial data is required to change positively, if the clinical trial data in the test group changes in the opposite direction, the test side is rotated away from the reference bottom surface, and the rotation angle is determined according to the rotation angle of the difference mapping.

[0076] Preferably, the execution steps of the difference calculation unit 4 include:

[0077] Step S41, respectively obtaining the center of the reference bottom surface, the center of the test side surface, and the center of the top edge of the test side surface;

[0078] Step S42, calculating a first Euclidean distance from the center of the reference bottom surface to the center of the test side surface and a second Euclidean distance from the center of the reference bottom surface to the center of the top edge of the test side surface;

[0079] Step S43: obtaining the difference between the test side surface and the reference bottom surface based on the first Euclidean distance and the second Euclidean distance. The specific steps are: averaging the first Euclidean distance and the second Euclidean distance, and outputting the average value as the difference.

[0080] After obtaining a three-dimensional figure containing a single reference bottom surface and multiple test side surfaces, it is necessary to calculate the difference between each test side surface and the reference bottom surface, where the difference is reflected by calculating the distance. Since the angle of the test side surface has been adjusted, the angle between each test side surface and the reference bottom surface is different, and the calculated difference will also be different. When calculating the difference, three points are first determined, the first is the center of the reference bottom surface, the second is the center of the test side surface, and the third is the center of the top edge of the test side surface. Then, the first Euclidean distance between the center of the reference bottom surface and the center of the test side surface, as well as the second Euclidean distance between the center of the reference bottom surface and the center of the top edge of the test side surface are calculated respectively. After averaging the first Euclidean distance and the second Euclidean distance, the difference can be obtained.

[0081] Preferably, the comprehensive adjustment unit 5 performs the following steps: comparing the difference with the expected difference in the initial clinical trial plan, and adjusting the test conditions of the test group patients based on the comparison results.

[0082] When adjusting the clinical trial plan based on the difference, it can be compared with the expected difference in the clinical trial plan. Then, based on the comparison results, the test conditions of the patients in the test group can be adjusted, such as changing the test environment of the test group, replacing new patients, etc. By collecting data in real time and calculating the difference for adjustment, closed-loop feedback can be achieved to improve the accuracy of clinical trial results.

[0083] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A clinical trial management system based on artificial intelligence, characterized in that: include: The collection unit is used to collect clinical trial data of the control group and the experimental group and perform preprocessing; A simulation surface construction unit, used to construct a reference bottom surface and a test side surface according to the clinical trial data of the control group and the test group respectively; A simulation surface connection unit is used to connect the test side surface to the top of the reference bottom surface and adjust the angle of the test side surface; A difference calculation unit, used to calculate the difference between the test side surface and the reference bottom surface; Comprehensive adjustment unit, used to adjust clinical trial plans based on the degree of variability; The acquisition unit, the simulation surface construction unit, the simulation surface connection unit, the difference calculation unit and the comprehensive adjustment unit are sequentially data-connected, and the simulation surface construction unit is data-connected with the simulation surface connection unit.

2. The clinical trial management system based on artificial intelligence according to claim 1, characterized in that: The execution steps of the acquisition unit include: Step S11: extract key features from the clinical trial protocol, input the key features into the trained neural network, and process the neural network to obtain the indicators to be monitored; Step S12: Acquire clinical trial data of the control group and the experimental group according to the indicators to be monitored, and perform data cleaning, denoising, and standardization on the clinical trial data.

3. The clinical trial management system based on artificial intelligence according to claim 2, characterized in that: The training steps of the neural network are as follows: obtaining a large number of historical clinical trial plans and historical monitoring indicators collected during the trial, extracting features from the historical clinical trial plans and obtaining historical key features, treating the historical key features and the corresponding historical monitoring indicators as a set of data, obtaining a training set and a test set by dividing them, training the neural network with the training set, and testing it with the test set after the training is completed.

4. The clinical trial management system based on artificial intelligence according to claim 1, characterized in that: The execution steps of the simulation surface construction unit include: Step S21: randomly select a clinical trial data as a construction index, and map the construction index of the control group to the area data of the control group; Step S22: construct a reference base of a regular polygon according to the number of experimental groups, adjust the size of the reference base based on the area data of the control group, and obtain the side length of the reference base; Step S23: construct a rectangular test side surface for each test group, with the bottom side of the test side surface being consistent with the side length of the reference bottom surface, mapping the construction index of the test group to the test group area data, and adjusting the height of the test side surface according to the test group area data; Step S24: Repeat steps S21-S23 until all clinical trial data are selected, and take the reference bottom surface and the test side surface corresponding to each construction indicator as a group.

5. The artificial intelligence-based clinical trial management system according to claim 4, characterized in that: If the number of control groups is plural, the clinical trial data of the control groups will be averaged.

6. The clinical trial management system based on artificial intelligence according to claim 1, characterized in that: The execution steps of the simulation surface connection unit include: Step S31: establishing a virtual multi-dimensional space, and mapping the reference bottom surface and the test side surface into the virtual multi-dimensional space; Step S32: align the bottom edge of each test side surface with the edge of the reference bottom surface, and keep each test side surface perpendicular to the reference bottom surface; Step S33: Compare the same clinical trial data of the test group and the control group, obtain the difference, and adjust the angle of the test side based on the difference.

7. The artificial intelligence-based clinical trial management system according to claim 6, characterized in that: The specific steps of step S33 are: Step S331: Evaluate the positive and negative variability of the selected clinical trial data, subtract the clinical trial data of the experimental group from the clinical trial data of the control group, obtain the difference, and map the difference to a rotation angle; Step S332: Determine whether the positive and negative signs of the difference are consistent with the positive and negative variability of the clinical trial data. If they are consistent, rotate the test side toward the center of the reference bottom surface according to the rotation angle. If they are inconsistent, rotate the test side toward the direction away from the reference bottom surface according to the rotation angle.

8. The artificial intelligence-based clinical trial management system according to claim 1, characterized in that: The execution steps of the differential calculation unit include: Step S41, respectively obtaining the center of the reference bottom surface, the center of the test side surface, and the center of the top edge of the test side surface; Step S42, calculating a first Euclidean distance from the center of the reference bottom surface to the center of the test side surface and a second Euclidean distance from the center of the reference bottom surface to the center of the top edge of the test side surface; Step S43: Obtain the difference between the test side surface and the reference bottom surface based on the first Euclidean distance and the second Euclidean distance.

9. The artificial intelligence-based clinical trial management system according to claim 8, characterized in that: The specific steps of step S43 are: averaging the first Euclidean distance and the second Euclidean distance, and outputting the average value as the difference.

10. The clinical trial management system based on artificial intelligence according to claim 1, characterized in that: The execution steps of the comprehensive adjustment unit are: comparing the difference with the expected difference in the initial clinical trial plan, and adjusting the test conditions of the test group patients based on the comparison results.

Images