Data processing device, data processing method, and data processing program

Abstract

To provide a data processing device, method, and program for generating a protocol or a report of a pharmaceutical development test using a generative AI model.SOLUTION: A data processing device includes: an input section that receives an input of necessary information, which is information necessary for generating a protocol or a report of a pharmaceutical development test, from a user; and a generation section that inputs instruction information including the necessary information and instructing on generation of a specific portion of the protocol or the report to a user-dedicated generative AI model, inputs instruction information instructing on verification of data of the specific portion to a general-purpose generative AI model trained using public data, inputs instruction information instructing on correction of the data of the specific portion including a verification result of the acquired data of the specific portion to the user-dedicated generative AI model, and generates the protocol or the report of the pharmaceutical development test using the acquired data of the specific portion after the correction.SELECTED DRAWING: Figure 4

Description

[Technical Field]

[0001] The technology of the present disclosure relates to a data processing device, a data processing method, and a data processing program. [Background technology]

[0002] Patent document 1 discloses a persona chatbot control method performed by at least one processor, the method including the steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to a description of the chatbot character, encoding the prompt, and inputting the encoded prompt into a language model to generate a chatbot utterance in response to the user utterance. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2022-180282 Summary of the Invention [Problem to be solved by the invention]

[0004] However, existing techniques leave room for improvement in using generative AI models to generate protocols or reports for drug development trials. [Means for solving the problem]

[0005] A first aspect of the technology disclosed herein is a data processing device comprising: an input unit that accepts input of necessary information from a user, which is information necessary for creating a protocol or report for a pharmaceutical development trial; a specific part acquisition unit that inputs instruction information including the necessary information and instructing the creation of a specific part of the protocol or report into a user-dedicated generative AI model trained using confidential data provided by the user, and acquires data for the specific part from the output of the user-dedicated generative AI model; a verification result acquisition unit that inputs instruction information including the acquired data for the specific part and instructing verification of the data for the specific part into a general-purpose generative AI model trained using public data, and acquires verification results for the data for the specific part from the output of the general-purpose generative AI model; a correction result acquisition unit that inputs instruction information including the verification results of the acquired data for the specific part and instructing correction of the data for the specific part into the user-dedicated generative AI model, and acquires the corrected data for the specific part from the output of the user-dedicated generative AI model; and a creation unit that creates a protocol or report for the pharmaceutical development trial using the acquired corrected data for the specific part.

[0006] A second aspect of the technology disclosed herein is a data processing method in which a computer accepts input of necessary information from a user, which is information necessary for creating a protocol or report for a pharmaceutical development trial, inputs instruction information including the necessary information and instructing the creation of a specific portion of the protocol or report into a user-specific generative AI model trained using confidential data provided by the user, acquires data for the specific portion from the output of the user-specific generative AI model, inputs instruction information including the acquired data for the specific portion and instructing verification of the data for the specific portion into a general-purpose generative AI model trained using public data, acquires a verification result for the data for the specific portion from the output of the general-purpose generative AI model, inputs instruction information including the verification result for the acquired data for the specific portion and instructing correction of the data for the specific portion into the user-specific generative AI model, acquires data for the corrected specific portion from the output of the user-specific generative AI model, and executes a process of creating a protocol or report for the pharmaceutical development trial using the acquired corrected data for the specific portion.

[0007] A third aspect of the technology disclosed herein is a data processing program that causes a computer to execute a process of accepting input of necessary information from a user, which is information necessary for creating a protocol or report for a pharmaceutical development trial, inputting instruction information that includes the necessary information and instructs the creation of a specific portion of the protocol or report into a user-specific generative AI model that has been trained using confidential data provided by the user, acquiring data for the specific portion from the output of the user-specific generative AI model, inputting instruction information that includes the acquired data for the specific portion and instructs verification of the data for the specific portion into a general-purpose generative AI model that has been trained using public data, acquiring a verification result for the data for the specific portion from the output of the general-purpose generative AI model, inputting instruction information that includes the verification result for the acquired data for the specific portion and instructs correction of the data for the specific portion into the user-specific generative AI model, acquiring data for the corrected specific portion from the output of the user-specific generative AI model, and creating a protocol or report for the pharmaceutical development trial using the acquired corrected data for the specific portion. [Brief explanation of the drawings]

[0008] [Figure 1] FIG. 1 is a conceptual diagram illustrating an example of a configuration of a data processing system. [Figure 2] FIG. 2 is a conceptual diagram showing an example of main functions of a data processing device and a user terminal. [Figure 3] An overview of the specific processing is shown below. [Figure 4] 2 shows a schematic functional configuration of a specific processing unit of the data processing device. [Figure 5] 10 is a diagram illustrating an example of an operational flow of specific processing by a data processing device. DETAILED DESCRIPTION OF THE INVENTION

[0009] Hereinafter, exemplary embodiments of a data processing device, a data processing method, and a data processing program according to the techniques of the present disclosure will be described with reference to the accompanying drawings.

[0010] First, the terms used in the following description will be explained.

[0011] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, the processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), an APU (Accelerated Processing Unit), or a TPU (Tensor Processing Unit).

[0012] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0013] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0014] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0015] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0016] FIG. 1 shows an example of the configuration of a data processing system 10 according to the embodiment.

[0017] As shown in Fig. 1, the data processing system 10 includes a data processing device 12 and a user terminal 14. An example of the data processing device 12 is a server. An example of the user terminal 14 is a personal computer or a smartphone. In this embodiment, the data processing device 12 is an example of a "data processing device" according to the technology of the present disclosure.

[0018] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0019] The user terminal 14 includes a computer 36, a reception device 38, an output device 40, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38 and the output device 40 are also connected to the bus 52.

[0020] The reception device 38 is equipped with a keyboard, a mouse, etc., and receives user input. The reception device 38 may also receive user input by touch of an indicator (for example, a pen or a finger) by detecting contact with the touch panel, or may receive user input by voice by detecting the user's voice by a microphone. The control unit 46A transmits data indicating the received user input to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0021] The output device 40 includes a display 40A and a speaker 40B, and presents data to the person 20 by outputting the data in a form of expression that can be perceived by the person 20 (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46.

[0022] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

[0023] FIG. 2 shows an example of the main functions of the data processing device 12 and the user terminal 14.

[0024] As shown in FIG. 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "data processing program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0025] Dedicated generative AI models 58A and 58B and a general-purpose generative AI model 58C are stored in storage 32. Dedicated generative AI models 58A and 58B and a general-purpose generative AI model 58C are used by specific processing unit 290.

[0026] The dedicated generative AI models 58A, 58B and the general-purpose generative AI model 58C are so-called generative AI (Artificial Intelligence). Examples of the dedicated generative AI models 58A, 58B and the general-purpose generative AI model 58C include ChatGPT (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> ) and other generative AI models. The dedicated generative AI models 58A, 58B and the general-purpose generative AI model 58C are obtained by performing deep learning on a neural network. A prompt including an instruction is input to the dedicated generative AI models 58A, 58B and the general-purpose generative AI model 58C, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The dedicated generative AI models 58A, 58B and the general-purpose generative AI model 58C perform inference on the input inference data in accordance with the instruction indicated by the prompt, and output the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0027] The dedicated generative AI models 58A, 58B are additionally trained using confidential data provided by the user.

[0028] For example, the dedicated generative AI model 58A is additionally trained using a dataset including the test plan portion of a clinical trial protocol for a pharmaceutical product, which is provided as confidential data by the user. The dedicated generative AI model 58B is additionally trained using a dataset including the analysis plan portion of a clinical trial protocol for a pharmaceutical product, which is provided by the user. Here, ChatGPT fine-tuning can be used for the additional training.

[0029] The general-purpose generative AI model 58C has been trained using publicly available data. For example, the general-purpose generative AI model 58C has been trained using a dataset containing information on medical guidelines and regulations from various countries, information on past clinical trials, and information on medical papers.

[0030] The dedicated generation AI models 58A and 58B can only be used by the user, and the general-purpose generation AI model 58C can be used by all users.

[0031] In the user terminal 14, the processor 46 performs the reception output processing. The storage 50 stores a reception output program 62. The reception output program 62 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 62 from the storage 50 and executes the read reception output program 62 on the RAM 48. The reception output processing is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 62 executed on the RAM 48.

[0032] Next, the processing of the specific processing unit 290 when the data processing device 12 performs specific processing to create a clinical trial protocol for a pharmaceutical product will be described.

[0033] In the specific processing of this embodiment, as shown in Figure 3, the results of the test plan portion of the implementation plan for a clinical trial of a pharmaceutical product, created using a dedicated generative AI model 58A, are verified using a general-purpose generative AI model 58C, and then the results are repeatedly corrected using the dedicated generative AI model 58A according to the verification results.

[0034] Furthermore, for the analysis plan portion of the clinical trial protocol for a pharmaceutical product, the results created using the dedicated generative AI model 58B are verified using the general-purpose generative AI model 58C, and then, depending on the verification results, the results are corrected using the dedicated generative AI model 58B. The trial plan portion and the analysis plan portion are examples of specific portions.

[0035] In addition, the ethical and other sections of the protocol for clinical trials of pharmaceuticals are created using the general-purpose generative AI model 58C. Note that the ethical and other sections are examples of sections that are different from specific sections.

[0036] Then, a general-purpose generative AI model 58C is used to create a clinical trial protocol for a pharmaceutical product by integrating each part of the clinical trial protocol.

[0037] As shown in FIG. 4, the identification processing unit 290 includes an input unit 291, a specific portion acquisition unit 292, a verification result acquisition unit 293, a correction result acquisition unit 294, a different portion acquisition unit 295, and a creation unit 296.

[0038] The input unit 291 acquires user input received by the user terminal 14. Specifically, necessary information, which is information required to create a protocol or report for a drug development test, is acquired as user input received by the user terminal 14. The necessary information includes at least the purpose of the clinical trial, the target disease, and information on the treatment method or drug to be administered.

[0039] The specific part acquisition unit 292 inputs instruction information, which includes necessary information and instructs the creation of the test plan part of the implementation plan or report, to the dedicated generation AI model 58A, and acquires data for the test plan part from the output of the dedicated generation AI model 58A. For example, "Please create the test plan part of the implementation plan of the pharmaceutical development test based on the necessary information below. ##Required Information## Clinical trial objective: Target disease: XXX The instruction information "Information on treatment or medication to be administered: XXX" is input into the dedicated generative AI model 58A.

[0040] In addition, the specific part acquisition unit 292 inputs instruction information containing the necessary information and instructing the creation of the analysis plan part of the implementation plan or report into the dedicated generation AI model 58B, and acquires data for the test plan part from the output of the dedicated generation AI model 58B.

[0041] The verification result acquisition unit 293 inputs instruction information, which includes the acquired data of the test plan portion and instructs the general-purpose generative AI model 58C to verify the data of the test plan portion, and acquires the verification result of the data of the test plan portion from the output of the general-purpose generative AI model 58C. For example, "The following is the test plan portion of the implementation plan for the drug development test. Please verify whether it is appropriate as an implementation plan for the drug development test. ##Study Plan Section of the Drug Development Study Protocol## The instruction information "〇〇〇" is input into the general-purpose generation AI model 58C.

[0042] In addition, the verification result acquisition unit 293 inputs instruction information that includes the acquired data of the analysis plan portion and instructs verification of the data of the analysis plan portion into the general-purpose generation AI model 58C, and acquires the verification results for the data of the analysis plan portion from the output of the general-purpose generation AI model 58C.

[0043] The correction result acquisition unit 294 inputs instruction information, which includes the verification results for the acquired data of the test plan portion and instructs correction of the data of the test plan portion, into the dedicated generative AI model 58A, and acquires the data of the test plan portion after correction from the output of the dedicated generative AI model 58A. For example, "Please correct the test plan portion of the implementation plan of the pharmaceutical development test based on the following verification results. ##Test results## The instruction information "〇〇〇" is input into the dedicated generation AI model 58A.

[0044] In addition, the correction result acquisition unit 294 inputs instruction information, which includes the verification results for the acquired data of the analysis plan portion and instructs correction of the data of the analysis plan portion, into the dedicated generation AI model 58B, and acquires the data of the corrected analysis plan portion from the output of the dedicated generation AI model 58B.

[0045] The acquisition of corrected data by the specific part acquisition unit 292 and the acquisition of verification results by the verification result acquisition unit 293 are repeated until there are no more items to be corrected as a result of the verification of the data of the test plan part.

[0046] Furthermore, the acquisition of corrected data by the specific part acquisition unit 292 and the acquisition of verification results by the verification result acquisition unit 293 are repeated until there are no more items to be corrected as a result of the verification of the data in the analysis plan part.

[0047] The other part acquisition unit 295 inputs instruction information containing necessary information and instructing the creation of the ethical part and other parts of the implementation plan for the pharmaceutical development trial into the general-purpose generative AI model 58C, and acquires data on the ethical part and other parts of the implementation plan for the pharmaceutical development trial from the output of the general-purpose generative AI model 58C.

[0048] The preparation unit 296 prepares a protocol for the pharmaceutical development trial using the acquired data for the revised trial plan portion, the acquired data for the revised analysis plan portion, and the acquired data for the ethics portion and other portions.

[0049] Specifically, the creation unit 296 inputs instruction information into the general-purpose generative AI model 58C that instructs the creation of an entire implementation plan for the pharmaceutical development trial, including the acquired data for the revised test plan portion, the acquired data for the revised analysis plan portion, and the acquired data for the ethics portion and other portions, and integrates the data, and obtains the creation result of the implementation plan for the pharmaceutical development trial from the output of the general-purpose generative AI model 58C.

[0050] The creation unit 296 also transmits the results of the creation of the pharmaceutical development test protocol to the user terminal 14. In the user terminal 14, the control unit 46A causes the output device 40 to output the results of the creation of the pharmaceutical development test protocol. The reception device 38 acquires user input regarding the results of the creation of the pharmaceutical development test protocol. The control unit 46A transmits the user input acquired by the reception device 38 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the user input.

[0051] Next, the operation of the data processing system 10 will be described.

[0052] An example of the flow of the specific processing by the data processing device 12 will be described with reference to Fig. 5. The flow of the specific processing shown in Fig. 5 is an example of the "data processing method" according to the technique of the present disclosure.

[0053] In step S100, the input unit 291 acquires, as user input received at the user terminal 14, necessary information that is information required to create a protocol or report for a pharmaceutical development test.

[0054] In step S102, the specific part acquisition unit 292 inputs instruction information containing the necessary information and instructing the creation of the test plan part of the implementation plan or report into the dedicated generation AI model 58A, and acquires data for the test plan part from the output of the dedicated generation AI model 58A.

[0055] In addition, the specific part acquisition unit 292 inputs instruction information containing the necessary information and instructing the creation of the analysis plan part of the implementation plan or report into the dedicated generation AI model 58B, and acquires data for the test plan part from the output of the dedicated generation AI model 58B.

[0056] In step S104, the verification result acquisition unit 293 inputs instruction information including the acquired data of the test plan part and instructing verification of the data of the test plan part into the general-purpose generation AI model 58C, and acquires the verification results for the data of the test plan part from the output of the general-purpose generation AI model 58C.

[0057] In addition, the verification result acquisition unit 293 inputs instruction information that includes the acquired data of the analysis plan portion and instructs verification of the data of the analysis plan portion into the general-purpose generation AI model 58C, and acquires the verification results for the data of the analysis plan portion from the output of the general-purpose generation AI model 58C.

[0058] In step S106, it is determined whether or not there is any item to be corrected based on the verification result acquired in step S104. For example, if the acquired verification result includes indications of items to be corrected (step S106; Yes), the data processing device 12 proceeds to step S108. On the other hand, if the acquired verification result does not include indications of items to be corrected (step S106; No), the data processing device 12 proceeds to step S110.

[0059] In step S108, if the verification results for the data in the test plan portion include indications of items that need to be corrected, the correction result acquisition unit 294 inputs instruction information that includes the verification results for the acquired data in the test plan portion and instructs correction of the data in the test plan portion into the dedicated generation AI model 58A, and acquires the corrected data in the test plan portion from the output of the dedicated generation AI model 58A.

[0060] In addition, if the verification results for the data in the analysis plan section include indications of items that need to be corrected, the correction result acquisition unit 294 inputs instruction information that includes the verification results for the acquired data in the analysis plan section and instructs correction of the data in the analysis plan section into the dedicated generation AI model 58B, and acquires the corrected data for the analysis plan section from the output of the dedicated generation AI model 58B.

[0061] Then, the data processing device 12 returns to step S104.

[0062] In step S110, the other part acquisition unit 295 inputs instruction information containing the necessary information and instructing the creation of the ethical part and other parts of the implementation plan for the pharmaceutical development trial into the general-purpose generative AI model 58C, and acquires data on the ethical part and other parts of the implementation plan for the pharmaceutical development trial from the output of the general-purpose generative AI model 58C.

[0063] In step S112, the creation unit 296 integrates the data, including the acquired data for the revised test plan portion, the revised analysis plan portion, and the acquired data for the ethics portion and other portions, and inputs instruction information that instructs the creation of the entire implementation plan for the drug development trial into the general-purpose generative AI model 58C, and obtains the creation results of the implementation plan for the drug development trial from the output of the general-purpose generative AI model 58C.

[0064] In step S114, the creation unit 296 transmits the results of creating the implementation plan for the pharmaceutical development test to the user terminal 14, and the specification process ends.

[0065] As described above, according to the information processing system of this embodiment, a user-specific generative AI model trained using confidential data provided by the user and a general-purpose generative AI model trained using public data are used to create a drug development trial implementation plan or report, thereby making it possible to appropriately create a drug development trial implementation plan or report taking into account the confidentiality of the data.

[0066] In addition, by verifying data for specific parts of the drug development trial plan or report between a user-specific generative AI model and a general-purpose generative AI model, the drug development trial plan or report can be created more appropriately.

[0067] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[0068] In the above embodiment, an example was given in which a specific portion of a protocol or report for a drug development trial is verified between a user-specific generative AI model and a general-purpose generative AI model. However, a protocol or report for a drug development trial may be created while verifying data in portions other than the specific portion. For example, data on the ethical section and other portions of the protocol for a drug development trial are obtained from the output of the general-purpose generative AI model 58C, and instruction information including the obtained ethical and other data and instructing verification of the ethical and other data is input to the general-purpose generative AI model 58C, and verification results for the ethical and other data are obtained from the output of the general-purpose generative AI model 58C. If the obtained verification results include indications for correction, instruction information including the verification results for the obtained ethical and other data and instructing correction of the ethical and other data is input to the general-purpose generative AI model 58C, and the corrected ethical and other data is obtained from the output of the general-purpose generative AI model 58C.

[0069] In the above embodiment, an example was given in which a specific process is performed by one computer 22, but the technology disclosed herein is not limited to this, and distributed processing of the specific process may be performed by multiple computers including computer 22.

[0070] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[0071] Alternatively, the specific processing program 56 may be stored in a storage device such as a server connected to the data processing device 12 via the network 54, and the specific processing program 56 may be downloaded and installed on the computer 22 in response to a request from the data processing device 12.

[0072] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[0073] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[0074] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[0075] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[0076] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[0077] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[0078] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference. [Explanation of symbols]

[0079] 10 Data Processing System 12 Data Processing Device 14 User terminal 58A, 58B dedicated generation model 290 Special Processing Department 291 Input section 292 Specific part acquisition part 293 Verification result acquisition unit 294 Correction result acquisition part 295 Other part acquisition part 296 Creation Department

Claims

1. an input unit that accepts input of necessary information from a user, which is information necessary for creating a protocol or report for a drug development test; a specific part acquisition unit that inputs instruction information, including the necessary information, instructing the creation of a specific part of an implementation plan or report, into a user-dedicated generative AI model trained using confidential data provided by the user, and acquires data of the specific part from the output of the user-dedicated generative AI model; a verification result acquisition unit that inputs instruction information, including the acquired specific portion of data and instructing verification of the specific portion of data, into a general-purpose generative AI model trained using public data, and acquires a verification result for the specific portion of data from the output of the general-purpose generative AI model; a correction result acquisition unit that inputs instruction information, which includes a verification result for the acquired specific portion of data and instructs correction of the specific portion of data, into the user-dedicated generative AI model, and acquires the corrected specific portion of data from the output of the user-dedicated generative AI model; a creation unit that creates a protocol or report of the pharmaceutical development test using the acquired corrected data of the specific portion; 2. A data processing device comprising:

2. The system further includes an other part acquisition unit that inputs instruction information, which includes the necessary information and instructs the creation of a part of the implementation plan or report that is different from the specific part, into the general-purpose generative AI model, and acquires data of the part that is different from the specific part from the output of the general-purpose generative AI model; The data processing device described in claim 1, wherein the creation unit inputs instruction information into the general-purpose generative AI model, the instruction information including the acquired data of the specific portion after the correction and the acquired data of a portion different from the specific portion, and instructs the creation of the entire implementation plan or report by integrating the data, and obtains the creation results of the implementation plan or report of the pharmaceutical development test from the output of the general-purpose generative AI model.

3. The public data is at least one of information on medical guidelines and regulations, information on past clinical trials, and information on medical papers; The data processing device according to claim 1 , wherein the general-purpose generative AI model is a model trained using the public data.

4. the input unit accepts input of necessary information that is information necessary for creating a clinical trial plan or report of a pharmaceutical product as the pharmaceutical development test; 2. The data processing device according to claim 1, wherein the necessary information includes at least the purpose of the clinical trial, the target disease, and information on the treatment method or the drug to be administered.

5. The computer Accepting input of necessary information from a user, which is information necessary for preparing a protocol or report for a drug development study; Inputting instruction information including the necessary information and instructing the creation of a specific part of an implementation plan or report into a user-specific generative AI model trained using confidential data provided by the user, and obtaining data for the specific part from the output of the user-specific generative AI model; inputting instruction information including the acquired specific portion of data and instructing verification of the specific portion of data into a general-purpose generative AI model trained using public data, and obtaining a verification result for the specific portion of data from the output of the general-purpose generative AI model; inputting instruction information, which includes a verification result for the acquired specific portion of data and instructs correction of the specific portion of data, into the user-dedicated generative AI model, and obtaining the corrected specific portion of data from the output of the user-dedicated generative AI model; The acquired corrected data of the specific portion is used to prepare a protocol or report for the pharmaceutical development test. The data processing method that performs the processing.

6. On the computer, Accepting input of necessary information from a user, which is information necessary for preparing a protocol or report for a drug development study; Inputting instruction information including the necessary information and instructing the creation of a specific part of an implementation plan or report into a user-specific generative AI model trained using confidential data provided by the user, and obtaining data for the specific part from the output of the user-specific generative AI model; inputting instruction information including the acquired specific portion of data and instructing verification of the specific portion of data into a general-purpose generative AI model trained using public data, and obtaining a verification result for the specific portion of data from the output of the general-purpose generative AI model; inputting instruction information, which includes a verification result for the acquired specific portion of data and instructs correction of the specific portion of data, into the user-dedicated generative AI model, and obtaining the corrected specific portion of data from the output of the user-dedicated generative AI model; The acquired corrected data of the specific portion is used to prepare a protocol or report for the pharmaceutical development test. A data processing program for executing processing.

Images