Clinical test scheme automatic generation system based on artificial intelligence

Abstract

The invention is suitable for the technical field of artificial intelligence, and provides an artificial intelligence-based clinical test scheme automatic generation system, which comprises a database, a retrieval module, an acquisition module, an extraction module, a storage module, an AI training module and an integration module in communication connection with the AI training module, according to the device, the clinical test design scheme is formulated by extracting clinical value points in the clinical test development plan, the clinical test design scheme can be automatically generated, the research and development direction is comprehensive and accurate, the success rate of clinical tests and the success rate of drug development are increased, and the clinical test development efficiency is improved. The clinical test development time is shortened, and the clinical test cost is reduced; the AI training module is combined with a machine learning model and a deep learning model, so that the system has strong autonomous learning ability, and self behaviors are continuously corrected through big data, so that the system has intelligent decision making and intelligent operation planning ability.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and more specifically, to an artificial intelligence-based automatic generation system for clinical trial protocols. Background Art

[0002] Clinical trials are systematic trials that aim to discover or verify the clinical medicine, pharmacology, and other pharmacodynamic effects, adverse reactions, or absorption, distribution, metabolism, and excretion of an investigational drug in order to determine the efficacy and safety of the drug.

[0003] A drug development plan describes all aspects of the development of a product from the target product profile to post-approval activities. The plan is usually prepared prospectively and updated as the development process proceeds and new information becomes available. The plan typically includes characterization of dosage forms for development, nonclinical studies needed to support product evaluation in human clinical studies and to support product approval, clinical studies to support efficacy and safety data in relevant patient populations, studies in special populations, regional considerations for product commercialization, and post-approval studies.

[0004] Artificial intelligence, abbreviated as AI in English, is an important driving force for a new round of scientific and technological revolution and industrial transformation. It is a new technical science that studies and develops theories, methods, technologies and application systems for simulating, extending and expanding human intelligence.

[0005] At present, in the entire clinical trial industry, clinical trial design plans are generated manually. Insufficient clinical trial planning leads to the failure of clinical trials, reduced efficiency of clinical trials, and ultimately the failure of new drug development. Therefore, it is very necessary to design an automatic generation system for clinical trial plans based on artificial intelligence. Summary of the invention

[0006] In view of the shortcomings of the prior art, the purpose of the present invention is to provide an automatic generation system of clinical trial plans based on artificial intelligence. By extracting clinical value points in the clinical trial development plan to formulate a clinical trial design plan, the clinical trial design plan can be automatically generated, and the research and development direction is comprehensive and accurate, thereby improving the success rate of clinical trials and the success rate of drug development, shortening the time of clinical trial development and reducing the cost of clinical trials; at the same time, through the AI ​​training module combined with the machine learning model and the deep learning model, the success rate of clinical trials and the success rate of drug development are further improved.

[0007] To achieve the above object, the present invention provides the following technical solutions: An artificial intelligence-based automatic generation system for clinical trial plans, including a database, the database is used for clinical trial data information of similar drugs and preclinical trial data information of the drug; a retrieval module, the retrieval module is used to retrieve and enter relevant test data information; a collection module, the collection module is used to obtain standard metadata; an extraction module communicatively connected to the collection module, the extraction module constructs a neural network model according to the data collected by the collection module, and extracts information based on the neural network model; a storage module communicatively connected to the extraction module; an AI training module communicatively connected to the storage module, the AI ​​training module trains the standard metadata of the clinical trial design history plan in the storage module to obtain standard meta-training data; and an integration module communicatively connected to the AI ​​training module, the integration module integrates the standard meta-training data to obtain a clinical trial design plan.

[0008] The present invention is further configured as follows: the standard meta-training data setting needs to extract standard metadata of the clinical trial design plan based on the collection settings related to basic elements and laws and regulations; wherein the laws and regulations include laws and regulations issued by the U.S. Food and Drug Administration and the National Medical Products Administration; in some optional embodiments, the standard metadata of the clinical trial design plan includes variables and variable labels, and the variable labels are composed of strings.

[0009] The present invention is further configured as follows: the neural network model includes a BERT model, a BILSTM model and a CRF model, the output data of the BERT model is input into the BILSTM model, and the data in the BERT model is output into the BILSTM model as the input of the CRF model, wherein the BERT model is used to realize the function of word embedding, and the BILSTM model and the CRF model construct a training neural network and constrain the output results.

[0010] The present invention is further configured as follows: the collection settings include project environment, project information, storage locations of various types of files and test data reading cycles; the project environment and project information are set according to the clinical trial project; a clinical trial design scheme flow chart is established based on the set collection settings and basic elements of the clinical trial design scheme; and the clinical trial design scheme flow chart and laws and regulations are established based on the established clinical trial design scheme flow chart.

[0011] The present invention is further configured as follows: the AI ​​training module includes a ROBOT unit and an AI+RPA unit; the ROBOT unit is connected to the AI+RPA unit via an engine.

[0012] The present invention is further configured as follows: the AI ​​training module also includes a control program; the AI ​​control module is electrically connected to the extraction module; the AI ​​control module is used to execute the control program to achieve the information extraction based on the neural network model as described above.

[0013] The present invention is further configured as follows: the integration module is used to retrieve and integrate standard meta-training data to obtain a clinical trial design plan, based on industry experience, industry domestic and international regulations and the summarized "669" principle, namely the six major processes of clinical trials, the six major elements of clinical trial design and the nine minor elements of clinical trial design. The six major processes of clinical trials include planning, design, execution, analysis, reporting and interpretation.

[0014] The present invention is further configured as follows: the six major elements of the clinical trial design include research population, intervention, control group, response variable, method for reducing or evaluating bias, and statistical analysis; the nine minor elements of the clinical trial design include research population, intervention, control group, response variable, method for reducing or evaluating bias, statistical analysis, research purpose, research type, and data source, and the standard metadata of the clinical trial design plan is integrated to obtain the clinical trial design plan.

[0015] The advantages of the present invention are: The present invention formulates clinical trial design plans by extracting clinical value points in clinical trial development plans. It can automatically generate clinical trial design plans with comprehensive and accurate research and development directions, thereby improving the success rates of clinical trials and drug development, shortening the time of clinical trial development and reducing clinical trial costs. At the same time, through the AI ​​training module combined with machine learning models and deep learning models, it has strong autonomous learning capabilities, and continuously corrects its own behavior through big data, thereby having intelligent decision-making and intelligent operations capabilities, further improving the success rates of clinical trials and drug development. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 The present invention is a flowchart of an artificial intelligence-based automatic generation system for clinical trial protocols.

[0017] Figure 2 This is a composition diagram of the AI ​​training module of the present invention. DETAILED DESCRIPTION

[0018] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application may be combined with each other. The present invention will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0019] It should be noted that, unless otherwise specified, all technical and scientific terms used in this application have the same meanings as commonly understood by ordinary technicians in the technical field to which this application belongs.

[0020] In the present invention, unless otherwise specified, the directions used, such as "up" and "down", usually refer to the directions shown in the drawings, or to the vertical, perpendicular or gravity directions; similarly, for ease of understanding and description, "left" and "right" usually refer to the left and right shown in the drawings; "inside" and "outside" refer to the inside and outside relative to the outline of each component itself, but the above-mentioned directions are not used to limit the present invention.

[0021] For example, see Figure 1-2 , the present invention provides the following technical solutions: A clinical trial protocol automatic generation system based on artificial intelligence, specifically, includes a database, and the database is used for clinical trial data information of similar drugs and preclinical trial data information of the drug.

[0022] Retrieval module, the retrieval module is used to retrieve and enter relevant test data information.

[0023] Collection module,The collection module is used to obtain standard metadata.

[0024] An extraction module is communicatively connected to the acquisition module, the extraction module constructs a neural network model according to the data collected by the acquisition module, and extracts information based on the neural network model; and a storage module is communicatively connected to the extraction module.

[0025] An AI training module is communicatively connected to the storage module, and the AI ​​training module trains the standard metadata of the clinical trial design history scheme in the storage module to obtain standard metadata training data.

[0026] and an integration module that communicates with the AI ​​training module, which integrates the standard meta-training data to derive a clinical trial design plan.

[0027] Working principle of the first embodiment: By extracting clinical value points from clinical trial development plans to formulate clinical trial design plans, clinical trial design plans can be automatically generated with comprehensive and accurate research and development directions, which improves the success rate of clinical trials and drug development, shortens the time of clinical trial development and reduces clinical trial costs; at the same time, through the AI ​​training module combined with machine learning models and deep learning models, it has strong autonomous learning capabilities, and continuously corrects its own behavior through big data, thus having intelligent decision-making and intelligent operations capabilities, further improving the success rate of clinical trials and drug development.

[0028] For example 2, please refer to Figure 1-2, this embodiment 2 makes the following improvements on the basis of embodiment 1. Specifically, the standard metadata training data setting needs to extract the standard metadata of the clinical trial design plan according to the collection settings related to the basic elements and laws and regulations; the laws and regulations include the laws and regulations issued by the U.S. Food and Drug Administration and the National Medical Products Administration; in some optional embodiments, the standard metadata of the clinical trial design plan includes variables and variable labels, and the variable labels are composed of strings.

[0029] In order to ensure that the standard metadata of the clinical trial design plan is of high quality, complete and accurate, specifically, before obtaining the standard metadata of the clinical trial design plan, the basic elements of the clinical trial design plan are formulated; among which, the basic elements of the formulated clinical trial design plan include innovations in basic research and drug action mechanisms, precision treatment points, dynamic changes in treatment needs, improvements in drug safety, improvements in treatment experience and convenience, and patient needs.

[0030] The collection settings include project environment, project information, storage locations of various files and test data reading cycle. The project environment and project information are set according to the clinical trial project. According to the set collection settings and the basic elements of the clinical trial design plan, a clinical trial design plan flow chart is established. According to the established clinical trial design plan flow chart and laws and regulations, the standard metadata of the clinical trial design plan is extracted.

[0031] The neural network models include the BERT model, the BILSTM model and the CRF model. The output data of the BERT model is input into the BILSTM model, and the data in the BERT model is output into the BILSTM model as the input of the CRF model. The BERT model is used to realize the word embedding function, and the BILSTM model and the CRF model construct a training neural network and constrain the output results.

[0032] The BERT model is an architecture that can be used for many downstream tasks, such as answering questions, classification, NER, etc. The working principle of the BERT model is the same as that of most Imagenet deep learning models. First, the BERT model is trained on a large corpus (Masked LM task), and then the model for our own task is fine-tuned by adding some extra layers at the end. The model can be classification, question answering, or NER, etc.; BiLSTM (Bidirectional Long Short-Term Memory Network) is a special recurrent neural network (RNN) that can process sequence data and maintain long-term memory. Unlike traditional RNN models, BiLSTM considers both past and future information, allowing the model to better capture contextual relationships in sequence data; Conditional Random Field (CRF) is a basic model in natural language processing, widely used in scenarios such as word segmentation, entity recognition, and part-of-speech tagging. With the popularity of deep learning, models such as BILSTM+CRF, BERT+CRF, TRANSFORMER+CRF, etc. have gradually appeared, and the effect has been significantly improved in these tagging scenarios.

[0033] The collection settings include project environment, project information, storage locations of various files and test data reading cycles. The project environment and project information are set according to the clinical trial project. According to the set collection settings and the basic elements of the clinical trial design plan, a clinical trial design plan flow chart is established. According to the established clinical trial design plan flow chart and laws and regulations.

[0034] The AI ​​training module includes a ROBOT unit and an AI+RPA unit; the ROBOT unit and the AI+RPA unit are connected through an engine; the AI ​​training module also includes a control program; the AI ​​control module and the extraction module are electrically connected; the AI ​​control module is used to execute the control program to realize the information extraction based on the neural network model as above.

[0035] The deep learning model uses the CNN+FPN network structure for feature extraction, and then uses the RPN network structure to generate multiple table candidate areas, performs table classification, table boundary regression and table line segmentation prediction on each table candidate area, and finally obtains all lines of all wired tables and all lines of all wireless tables on the current image; the AI ​​control module includes the ROBOT unit and the AI+RPA unit; the ROBOT unit is connected to the AI+RPA unit through an engine; the output end of the data preprocessing module is communicated with the ROBOT unit; the AI ​​control module combines machine learning models and deep learning models, and has strong autonomous learning capabilities. Its OCR, NLP, speech recognition and other technologies enable the RPA unit to have cognitive capabilities, and can continuously correct its own behavior through big data, thereby having intelligent decision-making and intelligent operations capabilities.

[0036] The integration module is used to retrieve and integrate standard meta-training data to obtain a clinical trial design plan based on industry experience, domestic and international industry regulations, and the summarized "669" principle, namely the six major processes of clinical trials, the six major elements of clinical trial design, and the nine minor elements of clinical trial design. The six major processes of clinical trials include planning, design, execution, analysis, reporting, and interpretation.

[0037] The six major elements of clinical trial design include study population, intervention, control group, response variable, methods to reduce or assess bias, and statistical analysis; the nine minor elements of clinical trial design include study population, intervention, control group, response variable, methods to reduce or assess bias, statistical analysis, research purpose, research type, and data source. The standard metadata of the clinical trial design plan is integrated to derive the clinical trial design plan.

[0038] The specific working principle of the second embodiment is as follows: First, the clinical value points in the clinical trial development plan are obtained, and then the basic elements of the clinical trial design plan are determined based on the clinical value points in the clinical trial development plan. Based on the basic elements of the clinical trial design plan, a clinical trial design plan flow chart is formulated. Based on the clinical trial design plan flow chart and laws and regulations, the standard metadata of the clinical trial design plan is obtained. By integrating the standard metadata of the clinical trial design plan, the clinical trial design plan is obtained. According to the clinical trial design plan, a clinical trial design plan database structure description file is obtained. According to the clinical trial design plan, a clinical trial design plan database is established according to the clinical trial design plan database structure description file, and the clinical trial design plan is stored in the clinical trial design plan database; and through the AI ​​control module combined with the machine learning model and the deep learning model, it has a strong autonomous learning ability, can continuously correct its own behavior through big data, so as to have intelligent decision-making and intelligent operation capabilities, thereby improving the success rate of clinical trials and the success rate of drug development.

[0039] Obviously, the above-described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0040] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present application. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprise" and / or "include" are used in this specification, it indicates the presence of features, steps, operations, devices, components and / or combinations thereof.

[0041] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the numbers used in this way can be interchanged where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein.

[0042] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

[0043] The above is only a preferred embodiment of the present invention, and the protection scope of the present invention is not limited to the above embodiments. All technical solutions under the concept of the present invention belong to the protection scope of the present invention. It should be pointed out that for ordinary technicians in this technical field, some improvements and modifications without departing from the principle of the present invention should also be regarded as the protection scope of the present invention.

Claims

1. An artificial intelligence-based automatic generation system for clinical trial protocols, characterized by: Including a database, the database is used for clinical trial data information of similar drugs and preclinical trial data information of the drug; A retrieval module, which is used to retrieve and enter relevant test data information; A collection module, wherein the collection module is used to obtain standard metadata; an extraction module connected in communication with the acquisition module, the extraction module constructing a neural network model according to the data collected by the acquisition module and extracting information based on the neural network model; a storage module connected in communication with the extraction module; An AI training module in communication with the storage module, wherein the AI ​​training module trains the standard metadata of the clinical trial design history scheme in the storage module to obtain standard metadata training data; and an integration module that is communicatively connected to the AI ​​training module, wherein the integration module integrates the standard meta-training data to obtain a clinical trial design plan.

2. The artificial intelligence-based automatic generation system for clinical trial protocols according to claim 1, characterized in that: The standard metadata training data setting needs to extract standard metadata of clinical trial design schemes according to the collection settings related to basic elements and laws and regulations; The laws and regulations include those issued by the U.S. Food and Drug Administration and the National Medical Products Administration; in some optional embodiments, the standard metadata of the clinical trial design plan includes variables and variable labels, and the variable labels are composed of strings.

3. The clinical trial protocol automatic generation system based on artificial intelligence according to claim 2, characterized in that: The neural network model includes a BERT model, a BILSTM model and a CRF model. The output data of the BERT model is input into the BILSTM model, and the data in the BERT model is output into the BILSTM model as the input of the CRF model. The BERT model is used to realize the function of word embedding. The BILSTM model and the CRF model construct a training neural network and constrain the output results.

4. The clinical trial protocol automatic generation system based on artificial intelligence according to claim 3, characterized in that: The collection settings include project environment, project information, storage locations of various files and test data reading cycles. The project environment and project information are set according to the clinical trial project. Based on the set collection settings and basic elements of the clinical trial design plan, a clinical trial design plan flow chart is established. Based on the established clinical trial design plan flow chart and laws and regulations.

5. The clinical trial protocol automatic generation system based on artificial intelligence according to claim 4, characterized in that: The AI ​​training module includes a ROBOT unit and an AI+RPA unit; the ROBOT unit is connected to the AI+RPA unit via an engine.

6. The artificial intelligence-based automatic generation system for clinical trial protocols according to claim 5, characterized in that: The AI ​​training module also includes a control program; the AI ​​control module is electrically connected to the extraction module; the AI ​​control module is used to execute the control program to achieve information extraction based on the neural network model as described above.

7. The artificial intelligence-based automatic generation system for clinical trial protocols according to claim 6, characterized in that: The integration module is used to retrieve and integrate standard meta-training data to obtain a clinical trial design plan based on industry experience, domestic and international industry regulations, and the summarized "669" principle, namely the six major processes of clinical trials, the six major elements of clinical trial design, and the nine minor elements of clinical trial design. The six major processes of clinical trials include planning, design, execution, analysis, reporting, and interpretation.

8. The artificial intelligence-based automatic generation system for clinical trial protocols according to claim 7, characterized in that: The six major elements of clinical trial design include study population, intervention, control group, response variable, methods to reduce or evaluate bias, and statistical analysis; the nine minor elements of clinical trial design include study population, intervention, control group, response variable, methods to reduce or evaluate bias, statistical analysis, research purpose, research type, and data source. The standard metadata of the clinical trial design plan is integrated to obtain the clinical trial design plan.

Images