New drug clinical trial information collection and analysis device

Abstract

The present invention belongs to the technical field of clinical trials of new drugs, and specifically relates to a device for collecting and analyzing information on clinical trials of new drugs. The device includes an information collection terminal, an information analysis terminal, and a result display terminal; the information collection terminal is used to collect basic information about drugs and basic information about cases, and the information analysis terminal is used to summarize the basic information about drugs and cases, exclude redundant information, and retain valid information related to drugs and cases; the knowledge graph generation module is used to generate a visual knowledge graph from the valid information; the result display terminal is used to display the generated knowledge graph. For the first time, the present invention aims to collect data on the efficacy evaluation of new drugs, constructs a knowledge graph about the efficacy of new drugs, and explains the knowledge graph, which not only realizes the interconnection between clinical experimental data and real-world data, but also visually displays the results of drug research and development, helping researchers and developers understand information such as drug effectiveness and safety.

Description

Technical Field

[0001] The present invention belongs to the technical field of new drug clinical trials, and in particular relates to a new drug clinical trial information collection and analysis device. Background Art

[0002] Clinical trials are a necessary step for new drugs to be brought to market. They refer to research trials conducted on volunteer patients or healthy volunteers. They involve the collection of multiple basic information such as the volunteers' gender, age, and medical history. They also involve changes in the volunteers' vital signs after taking the drugs. By analyzing this complex and extensive information, the information on the drug's effectiveness and side effects is summarized. After multiple rounds of clinical trials, the drug's safety, effectiveness, and dosage are ultimately determined.

[0003] Since the types and quantities of data involved in clinical trials of new drugs are very large, the integrity and accuracy of data collection are directly related to the evaluation of drug efficacy and trial efficiency. In the early days, these huge and complex data were manually counted, calculated and analyzed, which required a lot of manpower and time. With the development of information technology and the application of artificial intelligence, the efficiency and quality of data collection have improved, which has also improved the effectiveness of drug clinical trials.

[0004] Chinese patent CN108830034B discloses a pharmaceutical clinical research and development information processing system and method, the system includes a research and development subject end and a user subject end, the research and development subject end includes a first communication module, a first information receiving analysis and storage module, a first information calling analysis and processing module and a report generation module, the user subject end includes a terminal for collecting, inputting, storing and transmitting. By exchanging information between the research and development subject end and the user subject end, the clinical trial results are fed back to the researchers in real time, which facilitates the researchers to collect efficacy test results and adjust the test plan in time, thereby improving the efficiency of drug research and development.

[0005] However, the display of the analysis results of the above patent is a vague process and not specific enough. In view of the complexity of clinical trial data of new drugs and the complexity of data analysis, data processing requires professional information processing personnel, while drug research and development personnel generally only have a deep understanding and knowledge of the drug development process, and it is difficult to master the complex information processing process. Then, the clinical trial design and result analysis of new drugs require the cooperation of personnel from different fields, which increases the difficulty of communication between technical personnel. If a set of devices that can concretely display data analysis results can be developed, it will bring great convenience to non-computer information professional drug research and development technicians, and also facilitate non-computer information professional medical personnel to understand new drugs. Summary of the invention

[0006] In order to solve the above technical problems, the present invention provides a new drug clinical trial information collection and analysis device, which displays a large amount of new drug clinical trial data analysis results in the form of a knowledge graph, which is intuitive and easy to understand, and can effectively assist the research and development of new drugs.

[0007] The purpose of the present invention is to provide a new drug clinical trial information collection and analysis device, including: an information collection end, an information analysis end and a result display end;

[0008] The information collection terminal includes a research and development data collection terminal and a trial user data collection terminal; the research and development data collection terminal is used to input basic drug information, and the trial user data collection terminal is used to input basic case information of clinical drug trial personnel;

[0009] The information analysis end includes a research and development data archiving module, a trial user data archiving module, an information analysis module, an information troubleshooting module, and a knowledge graph generation module;

[0010] The R&D data archiving module is used to classify the entered basic drug information, correspond it with the drug name, and establish a R&D data table of multiple reaction classification information;

[0011] The trial user data archiving module is used to classify the entered basic case information, correspond it with the drug name, and establish a case data table of multiple reaction classification information;

[0012] The information analysis module is used to associate the data in the research and development data table with the data in the case data table, determine the corresponding relationship between the basic information of the drug and the basic information of the case, and establish multiple associated information data tables to store the corresponding relationship;

[0013] The information screening module is used to exclude redundant data in the associated information data table and retain valid information associated with drugs and cases;

[0014] The knowledge graph generation module is used to generate a visual knowledge graph from the effective information;

[0015] The result display end is used to display the generated knowledge graph.

[0016] Preferably, in the above-mentioned new drug clinical trial information collection and analysis device, the R&D data collection end includes a R&D data entry module and a R&D data storage module; the R&D data entry module is used to directly enter the basic information of the drug, and the R&D data storage module is used to store the entered basic information of the drug.

[0017] Preferably, in the above-mentioned new drug clinical trial information collection and analysis device, the basic drug information is indicators of the drug's research, project establishment, research and development process, physical and chemical properties, and disease-treating mechanism.

[0018] Preferably, in the above-mentioned new drug clinical trial information collection and analysis device, the trial patient data collection end includes a trial patient data entry module and a trial patient data storage module, the trial patient data entry module is used to directly enter the basic information of the case, and the trial patient data storage module is used to store the entered basic information of the case.

[0019] Preferably, in the above-mentioned new drug clinical trial information collection and analysis device, the basic case information refers to the change information of the conventional test indicators of the drug test personnel, the test indicators of the clinical imaging department, and the test indicators of the clinical body fluid and blood testing department before and after the trial of the drug.

[0020] Preferably, in the above-mentioned new drug clinical trial information collection and analysis device, each of the associated information data tables contains a group of basic drug information and a group of basic case information. A group of basic drug information refers to a combination of specific values ​​of different indicators in the basic drug information, and a group of basic case information refers to a combination of specific values ​​of different indicators in the basic case information before and after medication. The corresponding relationship between different indicators in the basic drug information table and different indicators in the basic case information is reflected through the associated information data table.

[0021] Preferably, in the above-mentioned new drug clinical trial information collection and analysis device, the method by which the information screening module excludes redundant data in the associated information data table is as follows:

[0022] The first step is to select the name of the drug to be analyzed and call all the associated information data tables related to the drug name;

[0023] The second step is to divide all indicators in all related information data tables into basic categories and efficacy categories, wherein the basic category includes non-related indicators that have no significant impact on drug indications, and the efficacy category includes directly related indicators that have a significant impact on drug indications;

[0024] The third step is to exclude all irrelevant indicators in the associated information data table to obtain a simplified data table, which is used as effective information to create a knowledge graph.

[0025] Preferably, in the above-mentioned new drug clinical trial information collection and analysis device, the non-related indicators and the directly related indicators are divided according to clinical experience or according to principal component analysis.

[0026] Preferably, in the above-mentioned new drug clinical trial information collection and analysis device, the specific process of the empirical method division is as follows:

[0027] Researchers and developers divide the basic information of drugs into directly relevant indicators and non-relevant indicators based on animal experimental results. Clinical medical staff divide the basic information of cases into directly relevant indicators and non-relevant indicators based on clinical experience. Directly relevant indicators have a significant impact on drug efficacy, while non-relevant indicators have no significant effect on drug efficacy.

[0028] Preferably, in the above-mentioned new drug clinical trial information collection and analysis device, the specific process of the principal component analysis method is as follows:

[0029] The efficacy of the target indication was taken as the dependent variable, and the various indicators of the basic information of the case and the basic information of the drug were taken as the independent variables. SPSS software was used to perform principal component analysis, and finally, several indicators with the highest principal component scores were screened out from the basic information of the case and the basic information of the drug. They were used as directly related indicators, and the remaining indicators were used as non-related indicators.

[0030] Compared with the prior art, the present invention has the following beneficial effects:

[0031] 1. For the first time, this invention aims to collect data on the efficacy evaluation of new drugs, constructs a knowledge graph about the efficacy of new drugs, and explains the knowledge graph. It not only realizes the interconnection between clinical experimental data and real-world data, but also visually displays the results of drug research and development, helping researchers to understand information such as drug effectiveness and safety.

[0032] 2. The knowledge graph constructed by the present invention is visualized and concrete, and contains important indicator information related to drug efficacy, such as dosage, usage, efficacy, applicable population, physiological indicators, etc. The present invention selects indicators significantly related to drugs and subjects from a large number of test data, eliminates the influence of redundant data, and simplifies the data processing and data verification process without affecting the authenticity of drug evaluation. We sort these indicators according to their importance and finally draw a knowledge graph representing the relationship between drugs, subjects and indicators, which has high usability and credibility.

[0033] 3. Taking the development of new drugs for breast cancer as an example, the present invention explains the construction process of the knowledge graph, including how to collect and analyze clinical trial data, how to associate with real-world electronic medical records, etc., and realizes the knowledge co-construction, sharing and reuse of clinical trials for the development of new drugs for breast cancer and real-world cases. The knowledge graph is used to realize the interconnection and intercommunication between clinical trial data for the development of new drugs for breast cancer and real-world cases. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 This is a schematic diagram of the new drug clinical trial information collection and analysis device of the present invention;

[0035] Figure 2A knowledge graph 1 produced for Example 1 of the present invention;

[0036] Figure 3 This is the second knowledge graph produced for Example 1 of the present invention. DETAILED DESCRIPTION

[0037] In order to enable those skilled in the art to better understand and implement the technical solution of the present invention, the present invention is further described below in conjunction with specific embodiments and drawings.

[0038] The present invention provides a new drug clinical trial information collection and analysis device, including an information collection terminal, an information analysis terminal and a result display terminal. Figure 1 .

[0039] The information collection end includes a research and development data collection end and a trial user data collection end. The research and development data collection end is used to enter basic drug information; the basic drug information mainly includes indicators such as drug research, project establishment, research and development process, physical and chemical properties, and therapeutic mechanisms, including drug appearance information, physical and chemical information, mechanism information and other information. The drug appearance information includes drug color (white, yellow, pink, etc.), dosage form (tablets, injections, capsules, granules, oral liquids, etc.), properties (powder, liquid, semi-solid, etc.), and auxiliary materials for preparing drugs (starch, sucrose, etc.). The physical and chemical information includes the name of the drug, the chemical structure, melting point, boiling point, etc. of the effective substance in the drug. The mechanism information includes the source of the effective active substance in the drug, the drug processing method, animal experiments and cell experiments, the therapeutic active groups of the effective substance, the drug therapeutic mechanism, the interaction principle between the drug and other drugs, etc. The other information refers to information about the drug other than the above information.

[0040] The R&D data collection end includes a R&D data entry module and a R&D data storage module; the R&D data entry module can be a mobile phone, computer, tablet computer, etc. These devices have a human-computer interaction interface and can manually enter various basic information of drugs; the R&D data storage module is used to store the entered basic information of drugs, which can be stored in the server or the memory provided by the R&D data entry module, so as to facilitate R&D personnel to call and view the basic information of drugs in real time.

[0041] The trial user data collection terminal is used to input the basic case information of clinical drug trial personnel; the basic case information mainly includes the routine test indicators of drug trial personnel (such as gender, age, height, weight, measurements, blood pressure, etc.), medication conditions (such as drug usage and dosage, medication time, other drug combination treatment plans, etc.), clinical imaging test indicators (such as B-ultrasound, color ultrasound, CT, etc.) and clinical body fluid and blood test indicators (such as urine routine, stool routine, blood routine, liver and kidney function indicators, infectious disease indicators, etc.), especially the changes in indicators related to drug indications that need to be input. The change value before and after medication of a specific test indicator can be used as an input indicator. The larger the change value, the greater the impact of the drug on the indicator. At this time, it is necessary to pay attention to the difference between the input indicator and the real-world test indicator. The trial user data collection terminal collects the input indicator, and the input indicator is the difference between the unified test indicators before and after medication. The basic case information must include the information before and after medication of the above indicators, which is convenient for comparing and observing the physiological health status of drug trial personnel after medication. The other information refers to information about the subject's case other than the above information.

[0042] The trial user data collection terminal is similar to the R&D data collection terminal, and includes a trial user data entry module and a trial user data storage module; the trial user data entry module can be a mobile phone, a computer, a tablet computer, etc. These devices have a human-computer interaction interface, and various basic case information can be manually entered; the trial user data storage module can use the server or the memory storage provided by the trial user data entry module to store the entered basic case information, so as to facilitate clinical medical staff to call and view the basic case information in real time.

[0043] The information analysis end is used to aggregate the basic drug information collected by the R&D data collection end and the basic case information collected by the trial user data collection end, so as to realize the interconnection and intercommunication of data in the R&D data collection end and the trial user data collection end. The information analysis end includes a R&D data archiving module, a trial user data archiving module, an information analysis module, an information troubleshooting module, and a knowledge graph generation module.

[0044] The R&D data archiving module is used to classify the entered basic drug information, correspond it with the drug name, and establish multiple R&D data tables with reaction classification information to facilitate future calls and traceability. For example, there is a R&D data table to record a drug name and its basic information, and the R&D data table corresponds to the name of the person who entered the information, the date of entry, etc. Through this R&D data table, the drug name and other corresponding basic drug information, the name of the person who entered the information, the date of entry, etc. are associated, and the R&D data table is numbered; for different drug names, establish their own corresponding R&D data tables, one R&D data table corresponds to one number, and when you need to call the content in the R&D data table, you only need to enter the R&D data table number through the R&D data entry module.

[0045] The trial user data archiving module is used to classify the entered basic case information, correspond it with the drug name, and establish multiple case data tables of reaction classification information for easy future call and traceability. For example, there is a case data table to record a case name and its basic information, and the case data table corresponds to the name of a person who enters the information, the entry date, etc. Through this case data table, the case name and other corresponding case basic information, the name of the person who enters the information, the entry date, etc. are associated, and the case data table is numbered; for different case names, corresponding case data tables are established. If there are duplicate names, an identifier is added after the name to distinguish different drug trial user cases. One case data table corresponds to one number. When the content in the case data table needs to be called, it is only necessary to enter the table number through the trial user data entry module.

[0046] The information analysis module is used to associate the research and development data table with the case data table, give the corresponding relationship between the basic information of the drug and the basic information of the case, and establish multiple associated information data tables to save these corresponding relationships. Each associated information data table contains a set of basic information of drugs and a set of basic information of cases. A set of basic information of drugs refers to the combination of specific values ​​of different indicators in the basic information of drugs. For different drugs, even if it is the same indicator, its specific value may be different, such as for the dosage form indicator, some are injections and some are tablets; a set of basic information of cases refers to the combination of specific values ​​of different indicators before and after medication in the basic information of cases, such as for the same blood sugar indicator of the same diabetes case, its blood sugar value before and after medication may be different. The corresponding relationship between different indicators in the basic information of drugs and different indicators in the basic information of cases is reflected by the associated information data table. For example, there is a corresponding relationship between the dosage of a drug and the weight of the case. As the weight gradually increases, the dosage of the drug gradually increases; for another example, the dosage form of a drug is related to the age of the case. Children can use obstetrics and powder particles, and adults can use oral tablets and capsules.

[0047] The information screening module is used to exclude redundant data in the associated information data table and retain valid information with a high degree of association with drugs and cases, as follows:

[0048] The first step is to select the name of the drug to be analyzed, and the information search module calls all the associated information data tables related to the drug name;

[0049] In the second step, the information screening module divides all indicators in all related information data tables, including indicators in the basic information of the drug and indicators in the basic information of the case, and divides these indicators into basic categories and efficacy categories; the basic category contains non-related indicators that have no significant effect on the drug indications, and the efficacy category contains directly related indicators that have a significant effect on the drug indications.

[0050] It should be noted that which of the above indicators are directly related to the drug and which are not related to the drug are divided according to clinical experience or according to principal component analysis.

[0051] The specific process of the empirical method division is as follows:

[0052] (1) Obtained from the R&D department, the R&D personnel divide the basic information of the drug into directly relevant indicators and non-relevant indicators based on the results of animal experiments. For example, after animal experiments, it is found that the drug dosage form has no significant effect on the drug efficacy, but the drug dosage has a significant effect on the drug efficacy. In this case, the drug dosage form is regarded as a non-relevant indicator, and the drug dosage is regarded as a relevant indicator. It should be noted that the above-mentioned significant or insignificant effects can be analyzed through significance analysis. For example, through significance analysis, the indicators with an effect at the level of P < 0.05 or P < 0.01 are identified and regarded as indicators with significant effects. Then the remaining indicators related to the drug can be regarded as non-relevant indicators.

[0053] (2) Clinical medical staff divide the basic information of the case into directly relevant indicators and non-relevant indicators based on clinical experience. Since the indicators corresponding to each disease have recognized rules, the credibility of the division based on clinical experience is also relatively high. For example, if the clinical trial is a drug for the treatment of hypertension, then the directly relevant indicators can be age, blood pressure, history of hypertension, etc., and the non-relevant indicators can be gender, etc.

[0054] The specific process of principal component analysis is as follows:

[0055] In the first step, since the purpose of drug research and development is to develop drugs with target indications, the efficacy of the target indications is used as the dependent variable, and the indicators of the basic information of the case and the basic information of the drug are used as independent variables, and principal component analysis is performed using SPSS software. Finally, the absolute values ​​of the principal component scores (some principal component scores are positive, and some are negative. This is because the impact directions are different, which correspond to the positive efficacy and the negative efficacy in the present invention. The positive efficacy means that the disease is improved, and the negative efficacy means that the disease is worsened or the side effects are large) are selected from the basic information of the case and the basic information of the drug. The highest indicators are used as directly related indicators, and the remaining indicators are used as non-related indicators.

[0056] In the second step, since each indicator in the basic information of cases and the basic information of drugs may reflect the efficacy of drugs to different degrees, even for the same indication and the same drug, the contents of various basic information indicators of drugs and basic information indicators of cases are different. We take these different contents but belonging to the same indicator as independent variables. Therefore, in order to eliminate the interference of human factors in empirical division, the principal component analysis method can be used to analyze each data that has a potential impact on drug efficacy to improve the credibility of the results.

[0057] The third step is to exclude all irrelevant indicators in the related information data table to obtain a simplified data table, which is used as effective information to create a knowledge graph using the simplified data table that records directly related indicators;

[0058] Preferably, for the directly related indicators and non-related indicators divided by the principal component analysis method, a check is performed before excluding all non-related indicators in the association information data table, as follows:

[0059] Randomly call a non-correlated indicator and use it and the directly related indicators that are screened out as independent variables, and use the efficacy of the target indication as the dependent variable. Then repeat the principal component analysis method in the above "second step" once, and rank all the independent variables by scores. If it is found that the randomly called indicator is ranked last, it means that the choice of the directly related indicator is correct.

[0060] Repeat the verification at least 3 times, and randomly call indicators that are different from the previous ones. If the randomly called indicators are ranked last each time, it means that the information screening process is correct, and the directly related indicators selected will be used to draw the knowledge graph.

[0061] The knowledge graph generation module is used to generate a visual knowledge graph from the effective information. The knowledge graph is in the form of a two-dimensional graph or a three-dimensional graph. The two-dimensional graph is a statistical graph including multiple peak spectra, or a statistical graph including multiple pixel points.

[0062] The result display terminal is used to display the generated knowledge graph, and a display or printer can be selected to achieve this function. By comparing the knowledge graphs of different drugs, researchers can obtain their safety and effectiveness after application to the human body. In particular, when the main components displayed in the knowledge graph contain indicators of drug side effects that are harmful to the human body, even if it has a high therapeutic effect, it may not be the first choice for clinical use.

[0063] The working principle of the present invention is explained below by taking the clinical trial of breast cancer drugs as an example.

[0064] Example 1

[0065] A device for collecting and analyzing information on clinical trials of new drugs comprises an information collecting end, an information analyzing end and a result display end.

[0066] The information collection end includes a research and development data collection end and a trial user data collection end. The research and development data collection end is used to input basic drug information; the basic drug information includes drug appearance information, physical and chemical information and mechanism information. The drug appearance information includes drug color (white), dosage form (tablets), and auxiliary materials (starch) for preparing drugs. The physical and chemical information includes the drug name (anti-breast cancer drug A), and the mechanism information includes drug treatment method (oral), and target (breast cancer cell line MCF-7). There are a total of 6 indicators in the basic drug information.

[0067] The R&D data collection end includes a R&D data entry module and a R&D data storage module; the R&D data entry module is a computer; the R&D data storage module is used to store the entered basic information of the drug, and it uses a cloud server.

[0068] The trial user data collection terminal is used to input the basic case information of clinical drug trial personnel; the basic case information includes conventional detection indicators (gender, age, height, weight, blood pressure), drug dosage (three times a day, 2g each time, continuous use for half a month), clinical imaging detection indicators (such as breast color Doppler ultrasound) and urine and blood test indicators (blood routine, liver function, kidney function, in order to simply and intuitively explain the principle of the present invention, the blood routine, liver function, kidney function of the blood test indicators are summarized as one indicator). Except for gender, age and height, the remaining indicators need to input the information difference before and after medication. There are 11 indicators in the basic case information.

[0069] The trial user data collection terminal includes a trial user data entry module and a trial user data storage module; the trial user data entry module is a computer; the trial user data storage module uses the same cloud server as the R&D data storage module.

[0070] The information analysis end includes a research and development data archiving module, a trial user data archiving module, an information analysis module, an information troubleshooting module, and a knowledge graph generation module.

[0071] The R&D data archiving module is used to classify and correspond the entered basic drug information, and establish multiple R&D data tables with reaction classification information and corresponding relationships. This embodiment associates 6 indicators of the basic drug information, the name of the person who entered the information, the date of entry, etc., one indicator is one category, and the name of the person who entered the basic drug information and the date of entry are a set of corresponding relationship data.

[0072] The trial user data archiving module is used to classify and correspond the entered basic case information, and establish multiple case data tables with reaction classification information and corresponding relationships. In this embodiment, 11 indicators of the basic case information, the name of the person entering the information, the entry date, etc. are associated, one indicator is one category, and the name of the person entering the basic drug information and the entry date are a set of corresponding relationship data.

[0073] The information analysis module is used to associate the research and development data table with the case data table, provide the corresponding relationship between the basic information of the drug and the basic information of the case, and establish multiple associated information data tables to save these corresponding relationships.

[0074] In this embodiment, the six indicators of the basic information of the drug are used as the vertical column titles, the 11 indicators of the basic information of the case are used as the horizontal column titles, and the size of the breast cancer tumor mass (indicating the efficacy of the drug) is used as the specific value of the indicator. An Excel table is prepared as a related information data table. For different cases or drugs, the size of the breast cancer tumor mass is different, so the efficacy represented is also different.

[0075] The information screening module is used to exclude redundant data in the associated information data table and retain valid information with a high degree of association with drugs and cases, as follows:

[0076] The first step is to select the name of the drug to be analyzed, and the information search module calls all the associated information data tables related to the drug name;

[0077] In the second step, the information screening module divides all indicators in all related information data tables, including indicators in the basic information of the drug and indicators in the basic information of the case, into basic categories and efficacy categories; the basic category contains non-related indicators that have no significant effect on the drug indications, and the efficacy category contains directly related indicators that have a significant effect on the drug indications. All non-related indicators in the related information data table are excluded to obtain a data simplification table, which is used as effective information to create a knowledge graph using the data simplification table that records the directly related indicators.

[0078] This embodiment adopts principal component analysis method for division, and the directly related indicators finally screened out include treatment method, drug dosage, target, breast ultrasound, blood routine, body weight, liver function, and kidney function, and the remaining indicators are non-related indicators.

[0079] The knowledge graph generation module is used to generate a two-dimensional knowledge graph in a visual table form from the valid information, and the result display terminal is used to display the generated knowledge graph, see Figure 2 . Figure 2 The importance of directly related indicators is arranged horizontally or vertically from large to small. The horizontal direction represents the indicators of basic case information, and the vertical direction represents the indicators of basic drug information. The influence trend of these indicators on tumor efficacy can be seen.

[0080] Alternatively, we use the knowledge graph generation module to generate a visual radial two-dimensional knowledge graph, see Figure 3 , distribute the drug names and directly related indicators according to quadrants and importance, such as Figure 3 The upper right quadrant is a significant influencing factor, involving the relationship between the drug, the dosage and treatment of the drug, the target of action (indications), breast ultrasound (indications of efficacy), and other information. The upper left quadrant is a general influencing factor, involving the relationship between the dosage and treatment of the drug and blood routine indicators. The lower left quadrant is a side effect influencing factor, involving the interaction between the dosage and treatment of the drug and liver function and kidney function. The lower right quadrant is basically no influencing factor, involving the interaction between the dosage and treatment of the drug and body weight. Figure 3 In the figure, the longer the guide line between the drug name and the indicator name, the greater the influence.

[0081] It should be noted that this embodiment is only for explaining the working principle of the present invention, so blood routine is regarded as an indicator. In fact, blood routine should include multiple detection indicators. In addition, the non-related indicators and directly related indicators finally obtained in this embodiment are also simulated data, and should not be directly used as the knowledge map of a certain breast cancer drug.

[0082] It should be noted that for a certain specific disease, there are multiple medication terms and efficacy terms. The standardization of medication terms can promote the standardized collection of data and facilitate the statistical analysis of data. In order to realize the sharing and exchange of drug clinical trial data management system and real-world data, following the corresponding standards and specifications is the basis. In order for the device of the present invention to correctly understand the data, the meaning of the data must be expressed in an accurate and consistent form. Medical terminology and ontology are recognized standards for formalizing (encoding) data in the industry, with the aim of avoiding misunderstanding or ambiguity of the data by computers. CDISC is a standard currently used by the FDA and other countries including China in the process of drug clinical trials. SNOMED CT and LOINC are the most widely used concept expression standards in electronic medical records. Therefore, the knowledge graph constructed by the present invention can be coordinated with CDISC, SNOMED CT and LOINC standards, including concepts such as breast cancer drug treatment efficacy entities, entity attributes, and instance coding. It aims to ensure consistent expression of data.

[0083] It should be noted that when the present invention involves a numerical range, it should be understood that the two endpoints of each numerical range and any value between the two endpoints can be selected. Since the steps and methods used are the same as those in the embodiments, in order to avoid redundancy, the present invention describes a preferred embodiment. Although the preferred embodiments of the present invention have been described, those skilled in the art may make additional changes and modifications to these embodiments once they know the basic creative concept. Therefore, the attached claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0084] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. New drug clinical trial information collection and analysis device, It is characterized in that include: Information collection end, information analysis end and result display end; The information collection terminal includes a research and development data collection terminal and a trial user data collection terminal; the research and development data collection terminal is used to input basic drug information, and the trial user data collection terminal is used to input basic case information of clinical drug trial personnel; The information analysis end includes a research and development data archiving module, a trial user data archiving module, an information analysis module, an information troubleshooting module, and a knowledge graph generation module; The R&D data archiving module is used to classify the entered basic drug information, correspond it with the drug name, and establish a R&D data table of multiple reaction classification information; The trial user data archiving module is used to classify the entered basic case information, correspond it with the drug name, and establish a case data table of multiple reaction classification information; The information analysis module is used to associate the data in the research and development data table with the data in the case data table, determine the corresponding relationship between the basic information of the drug and the basic information of the case, and establish multiple associated information data tables to store the corresponding relationship; The information screening module is used to exclude redundant data in the associated information data table and retain valid information associated with drugs and cases; The knowledge graph generation module is used to generate a visual knowledge graph from the effective information; The result display terminal is used to display the generated knowledge graph; Each of the associated information data tables contains a set of basic drug information and a set of basic case information. A set of basic drug information refers to a combination of specific values ​​of different indicators in the basic drug information, and a set of basic case information refers to a combination of specific values ​​of different indicators in the basic case information before and after medication. The associated information data table is used to reflect the correspondence between different indicators in the basic drug information table and different indicators in the basic case information. The method for the information checking module to exclude redundant data in the associated information data table is as follows: The first step is to select the name of the drug to be analyzed and call all the associated information data tables related to the drug name; The second step is to divide all indicators in all related information data tables into basic categories and efficacy categories, wherein the basic category includes non-related indicators that have no significant impact on drug indications, and the efficacy category includes directly related indicators that have a significant impact on drug indications; The third step is to exclude all irrelevant indicators in the associated information data table to obtain a simplified data table, which is used as effective information to create a knowledge graph; The non-related indicators and the directly related indicators are divided according to clinical experience or according to principal component analysis.

2. The new drug clinical trial information collection and analysis device according to claim 1, It is characterized in that The R&D data collection end includes a R&D data entry module and a R&D data storage module; the R&D data entry module is used to directly enter basic drug information, and the R&D data storage module is used to store the entered basic drug information.

3. The new drug clinical trial information collection and analysis device according to claim 2, It is characterized in that The basic information of the drug is indicators of the drug's research, project establishment, research and development process, physical and chemical properties, and therapeutic mechanism.

4. The new drug clinical trial information collection and analysis device according to claim 1, It is characterized in that The trial user data collection terminal includes a trial user data entry module and a trial user data storage module. The trial user data entry module is used to directly enter basic case information, and the trial user data storage module is used to store the entered basic case information.

5. The new drug clinical trial information collection and analysis device according to claim 4, It is characterized in that The basic case information refers to the changes in routine test indicators, clinical imaging test indicators, and clinical body fluid and blood testing indicators of drug trial personnel before and after the drug trial.

6. The new drug clinical trial information collection and analysis device according to claim 1, It is characterized in that The specific process of the empirical method division is as follows: Researchers and developers divide the basic information of drugs into directly relevant indicators and non-relevant indicators based on animal experimental results. Clinical medical staff divide the basic information of cases into directly relevant indicators and non-relevant indicators based on clinical experience. Directly relevant indicators have a significant impact on drug efficacy, while non-relevant indicators have no significant effect on drug efficacy.

7. The new drug clinical trial information collection and analysis device according to claim 1, It is characterized in that The specific process of principal component analysis is as follows: The efficacy of the target indication was taken as the dependent variable, and the various indicators of the basic information of the case and the basic information of the drug were taken as the independent variables. SPSS software was used to perform principal component analysis, and finally, several indicators with the highest principal component scores were screened out from the basic information of the case and the basic information of the drug. They were used as directly related indicators, and the remaining indicators were used as non-related indicators.

Images