System

Abstract

A system is provided.SOLUTION: A system comprising: means for automatically collecting clinical and non-clinical data; means for formatting and converting the collected data into a unified format; means for extracting key information from the formatted data and generating a summary; means for visualizing in a graphical or chart format based on the summary data; means for comparing new data with existing data and providing the results in a report format; means for periodically obtaining up-to-date regulatory information and updating the database; and means for allowing a user to search for necessary data and view summaries and reports using an interface.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The technology of the present disclosure relates to a system. [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] The pharmaceutical application process for new drug development, medical devices, cosmetics, etc. requires a large amount of clinical and non-clinical trial data, efficacy and safety reviews, and pharmaceutical regulatory research. This work requires specialized knowledge, and the personnel costs and work time are major obstacles. This burden is particularly great for small biotechnology companies and venture companies, and there is an urgent need to streamline the pharmaceutical application process. Therefore, there is a need to provide a system that solves these issues and achieves efficiency and cost reduction in the pharmaceutical application process. [Means for solving the problem]

[0005] The present invention provides a system that automatically collects clinical and non-clinical data, formats the collected data, and converts it into a unified format, extracts important information from the formatted data and generates summaries, visualizes the summarized data in graphs and charts, compares new data with existing data and provides the results in report format, periodically obtains the latest pharmaceutical regulatory information and updates the database, and allows users to use an interface to search for required data and view summaries and reports. This system significantly reduces the time and cost required for pharmaceutical applications and improves the efficiency and accuracy of the application process.

[0006] "Clinical data" refers to data from experiments conducted on humans to evaluate the effectiveness and safety of pharmaceuticals, medical devices, etc.

[0007] "Non-clinical data" refers to data resulting from animal experiments and in-vitro experiments conducted to evaluate the efficacy and safety of pharmaceuticals, medical devices, etc.

[0008] An "automated collection means" is a device or method by which a system or machine performs a data collection process without human intervention.

[0009] "Data formatting means" refers to the process or method for converting acquired data into a unified format and making data of different formats into a consistent format.

[0010] "Information extraction methods" are techniques and methods for selecting important and relevant items or elements from large amounts of data.

[0011] A "summary generator" is a technique or method for concisely summarizing large amounts of information into short sentences containing the main points.

[0012] A "visualization tool" is a process or tool that displays data in a visual format, such as a graph or chart, to make it easier to understand.

[0013] A "data comparison tool" is a process or method for comparing newly acquired data with existing data to evaluate and report similarities and differences.

[0014] "Means of providing in report format" refers to the process or method of documenting the analysis results and extracted information and providing them to users in an easy-to-read format.

[0015] "Pharmaceutical regulatory information" refers to information on rules and guidelines regarding the approval, manufacturing, and sale of pharmaceuticals and medical devices established by regulatory authorities in each country.

[0016] A "means of updating a database" is a process or method for constantly incorporating the latest information and data to keep an existing database up to date.

[0017] An "interface" is the means or display by which a user interacts with a system, allowing them to easily find and obtain the information they need.

[0018] A "search tool" is a process or method for finding relevant data or information based on specific keywords or criteria.

[0019] "Means for viewing summaries and reports" refers to functions and methods that allow users to check the generated summaries and analysis reports on the screen and obtain the necessary information. [Brief explanation of the drawings]

[0020] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4]FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0021] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

[0022] First, the terms used in the following description will be explained.

[0023] 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, a 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), and an APU (Accelerated Processing Unit).

[0024] 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.

[0025] 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.

[0026] 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.

[0027] 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."

[0028] [First embodiment]

[0029] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0030] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0031] 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).

[0032] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, 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, the output device 40, and the camera 42 are also connected to the bus 52.

[0033] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0034] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (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. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0035] 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.

[0036] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0037] 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 "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.

[0038] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0039] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0040] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0041] The present invention is a system for streamlining data collection, formatting, summarization, visualization, comparison, and updating of regulatory information in the pharmaceutical application process. This system has the following components and functions.

[0042] System Overview

[0043] The system consists of a server, terminals, and users. The server is primarily responsible for data collection, automatic formatting, summary generation, data visualization, data comparison, and regulatory information updates, while the terminals function as an interface with users. Users use the system to search for and view the necessary data and create the documents required for pharmaceutical applications.

[0044] Program processing

[0045] 1. Data Collection

[0046] Subject: Server

[0047] The server automatically collects data based on pre-defined keywords from publicly available scientific papers, clinical trial databases, drug regulatory agency websites, etc. Specifically, it uses various APIs and web scraping technologies to retrieve the required data and store it in a database.

[0048] Examples:

[0049] The server retrieves the titles, authors, abstracts, and full texts of relevant articles from PubMed using the keyword "cancer treatment" and stores them in a database.

[0050] 2. Data Formatting

[0051] Subject: Server

[0052] The server reviews the collected data and converts it into a unified format, removing duplicates and imputing missing values ​​to ensure data integrity.

[0053] Examples:

[0054] The server analyzes multiple article data retrieved from PubMed, removes duplicate entries, and adds reliable default values ​​for missing entries.

[0055] 3. Data Summary

[0056] Subject: Server

[0057] The server uses natural language processing technology to extract important information from the data and generate a summary of key points (e.g., treatment effects, side effects, etc.).

[0058] Examples:

[0059] The server extracts information about each paper's key keywords, therapeutic effects, and side effects, and provides it as a summary.

[0060] 4. Data Visualization

[0061] Subject: Server

[0062] The server uses tools to generate graphs and charts based on the summarized data to visually display the data.

[0063] Examples:

[0064] The server visualizes the treatment efficacy data as a line graph and displays the frequency of side effects as a bar graph.

[0065] 5. Data Comparison

[0066] Subject: Server

[0067] The server compares the newly acquired data with the existing data, analyzes similarities and differences, and provides the comparison results in the form of a report.

[0068] Examples:

[0069] The server compares data on existing cancer treatments with data on newly acquired treatments and generates a detailed report on differences in treatment effectiveness and side effects.

[0070] 6. Regulatory Updates

[0071] Subject: Server

[0072] The server periodically retrieves the latest regulatory information from the websites of each country's drug regulatory agency and updates the database.

[0073] Examples:

[0074] The server uses web scraping technology to obtain the latest US FDA guidelines and imports them into the database for updating.

[0075] 7. User Interface

[0076] Subject: User

[0077] Users use a web interface to search for the data they need, view summaries and reports, and generate regulatory submissions based on the information provided.

[0078] Examples:

[0079] The user searches for data on "new treatments for lung cancer" in the interface, checks the obtained summary and comparison reports, and downloads a template for pharmaceutical application documents based on the information and fills in the required fields to complete the application.

[0080] In this way, the present invention is a system that automates application procedures and improves accuracy and speed by efficiently collecting, formatting, summarizing, visualizing, and comparing the various data required for the pharmaceutical application process, as well as updating the latest regulatory information.

[0081] The processing flow will be explained below.

[0082] Step 1: Data collection

[0083] Subject: Server

[0084] The server uses pre-configured API keys and web scraping techniques to collect data from scientific papers, clinical trial databases, and drug regulatory agency websites. Specifically, the server uses the PubMed API to retrieve paper titles and abstracts related to "cancer treatment" and stores them in a database in JSON format. It also uses web scraping technology to crawl the FDA's latest guidelines page, retrieves their contents in text format, and stores them.

[0085] Step 2: Data Shaping

[0086] Subject: Server

[0087] The server analyzes the collected data and unifies different data formats. Specifically, it parses JSON-formatted data, removes duplicate data, and fills in missing values ​​with appropriate default values. For example, in the case of paper data, if multiple papers with the same title are obtained, they are merged into one, and if information such as author names or publication dates is missing, it is filled in from reliable external information.

[0088] Step 3: Data Summarization

[0089] Subject: Server

[0090] The server uses natural language processing (NLP) technology to extract important information from the collected data and generate summaries. Specifically, it uses an NLP library (e.g., NLTK or SpaCy) to analyze the main keywords in the paper, as well as descriptions of therapeutic effects and side effects, and generates summaries based on this. The generated summaries are then stored in a database.

[0091] Step 4: Data visualization

[0092] Subject: Server

[0093] The server uses tools to generate graphs and charts based on the summarized data for visualization. For example, Matplotlib is used to generate line graphs showing the effectiveness of each treatment and bar graphs showing the frequency of side effects. The generated graphs and charts are converted into a format that can be viewed by the user interface and displayed on a dashboard in real time.

[0094] Step 5: Data comparison

[0095] Subject: Server

[0096] The server compares the newly acquired data with existing data and analyzes differences and commonalities. Specifically, it cross-references data on existing cancer treatments with data on new treatments, comparing them in terms of treatment effectiveness, side effects, etc. The comparison results are generated in report format, making them easy for users to understand.

[0097] Step 6: Update regulatory information

[0098] Subject: Server

[0099] The server periodically retrieves the latest information from the websites of each country's drug regulatory agency and updates the database. Specifically, it uses a web scraping tool to retrieve the latest guidelines and regulatory changes from the FAA and EMA, stores them in the database, and sends notifications to users when changes are confirmed.

[0100] Step 7: Search and view data

[0101] Subject: User

[0102] Users use the device's web interface to search for and retrieve the data they need. Specifically, they enter keywords such as "new treatments for lung cancer" into the search screen and click the search button to send a request to the server. The server then retrieves relevant summaries and comparison reports from the database and displays them on the user's device.

[0103] Step 8: Prepare regulatory submission documents

[0104] Subject: User

[0105] Users create regulatory application documents based on the displayed summaries and comparison reports. Specifically, they download templates provided by the system and create documents by entering the displayed information. They can also submit the created documents for review by experts, who can make revisions based on the feedback.

[0106] Example 1

[0107] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0108] The current pharmaceutical application process involves a wide range of tasks, including collecting, organizing, summarizing, visualizing, comparing, and updating regulatory information, and is extremely time-consuming and labor-intensive. Furthermore, the quality and consistency of the collected data varies, and the accuracy and efficiency of application document preparation is not fully guaranteed. This leads to delays in the application process and the submission of incorrect information.

[0109] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0110] In this invention, the server includes, in data collection, means for automatically obtaining data from publicly available information sources based on preset keywords, means for reviewing the collected data and checking for format inconsistencies and missing values, means for formatting the collected data, converting it into a unified format, deleting duplicate data, and imputing missing values, means for extracting important information from the collected data using natural language processing technology and generating a summary, means for visualizing the summary data in the form of a graph or chart, means for comparing newly obtained data with existing data, analyzing similarities and differences, and providing the results in the form of a report, means for periodically obtaining the latest regulatory information and updating the database, means for a user to use an interface to search for necessary data and view summaries and reports, and means for providing application document templates based on the search results and for the user to enter the necessary information to create them. This makes it possible to improve the efficiency and accuracy of the pharmaceutical application process.

[0111] "Data harvesting" refers to the automatic retrieval of data from publicly available sources based on pre-defined keywords.

[0112] "Format inconsistency" refers to a situation where collected data exists in different structures or formats.

[0113] "Missing values" refers to a state in which some data is missing or necessary information is not entered.

[0114] "Duplicate data" refers to a situation in which multiple pieces of data with the same content exist.

[0115] "Missing value imputation" refers to filling in missing data based on predictions or reliable sources.

[0116] "Natural language processing" refers to the technology for understanding, analyzing, and generating human language.

[0117] "Extracting important information" refers to extracting only specific, necessary information from collected data.

[0118] A "summary" is a short summary of the main points from a long piece of text or a large amount of data.

[0119] "Visualization" refers to the display of data in a visual format such as a graph or chart.

[0120] "Report format" refers to a format in which the results of analysis and comparison are systematically summarized in documents and charts.

[0121] "Regulatory information" refers to information on norms and standards provided by public institutions, such as laws, regulations, and guidelines.

[0122] "Database updating" refers to adding new information to existing data and keeping it up to date.

[0123] "User interface" refers to the screens and controls that allow a user to interact with a computer system or application.

[0124] "Application template" refers to a template for application documents in a predefined format.

[0125] This invention is a system that streamlines data collection, formatting, summarization, visualization, comparison, and updating of regulatory information in the pharmaceutical application process. This system is mainly composed of a server, terminals, and users. The specific system structure and operation are described in detail below.

[0126] System Configuration

[0127] server:

[0128] The server is responsible for data collection, automatic formatting, summary generation, data visualization, data comparison, and regulatory information updates. The server uses the following hardware and software:

[0129] Hardware: High-performance server (e.g., Xeon-based server)

[0130] Software: Python, natural language processing libraries (e.g., NLTK, SpaCy), data visualization tools (e.g., Matplotlib, D3.js)

[0131] Device:

[0132] The terminal provides the interface through which the user interacts with the system. The terminal uses the following hardware and software:

[0133] Hardware: Personal computers, tablets

[0134] Software: Web browser (e.g., GOOGLE CHROME (registered trademark), Firefox)

[0135] User:

[0136] Users are drug developers and regulatory officials involved in the regulatory application process, who use the system to search for the data they need, view summaries and reports, and prepare application documents.

[0137] How it works

[0138] 1. Data Collection:

[0139] The server collects data based on pre-defined keywords from published scientific papers, clinical trial databases, drug regulatory agency websites, etc., using various API calls and web scraping techniques.

[0140] Examples:

[0141] The server retrieves the titles, authors, abstracts, and full texts of relevant articles from PubMed using the keyword "cancer treatment" and stores them in a database.

[0142] 2. Data Formatting:

[0143] The server reviews the collected data and converts it into a unified format, removing duplicates and imputing missing values ​​to ensure data integrity.

[0144] Examples:

[0145] The server analyzes multiple article data retrieved from PubMed, removes duplicate entries, and adds reliable default values ​​for missing entries.

[0146] 3. Data Summary:

[0147] The server uses natural language processing techniques to extract key information from the data and generate a summary.

[0148] Examples:

[0149] The server extracts information about each paper's key keywords, therapeutic effects, and side effects, and provides it as a summary.

[0150] 4. Data visualization:

[0151] The server uses tools to generate graphs and charts based on the summarized data to visually display the data.

[0152] Examples:

[0153] The server visualizes the treatment efficacy data as a line graph and displays the frequency of side effects as a bar graph.

[0154] 5. Data comparison:

[0155] The server compares the newly acquired data with the existing data, analyzes similarities and differences, and provides the results in a report.

[0156] Examples:

[0157] The server compares data on existing cancer treatments with data on newly acquired treatments and generates a detailed report on differences in treatment effectiveness and side effects.

[0158] 6. Regulatory Updates:

[0159] The server periodically retrieves the latest regulatory information from the websites of each country's drug regulatory agency and updates the database.

[0160] Examples:

[0161] The server uses web scraping technology to obtain the latest US FDA guidelines and imports them into the database for updating.

[0162] 7. User Interface:

[0163] Users use a web interface to search for the data they need, view summaries and reports, and download application templates based on their search results, then fill in the required fields to complete the application.

[0164] Examples:

[0165] Users can search for data on "new treatments for lung cancer" in the interface, check the resulting summary and comparison reports, and download a template for a pharmaceutical application based on the information, then fill in the required fields to create it.

[0166] Prompt Sentence Examples

[0167] "Collect the latest articles on 'cancer treatment' from PubMed and create summaries that extract key keywords, therapeutic effects, and side effects."

[0168] As described above, this system comprehensively covers the entire process from data collection to preparation of application documents, dramatically improving the efficiency and accuracy of pharmaceutical applications.

[0169] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0170] Step 1: Data collection

[0171] Input: Predefined keywords (e.g., "cancer treatment").

[0172] Specific behavior:

[0173] The server automatically collects data from various data sources (e.g., scientific paper databases, clinical trial databases, regulatory agency websites, etc.) based on configured keywords, using API calls and web scraping techniques.

[0174] Data processing:

[0175] The acquired data is stored in a temporary database.

[0176] Output: The raw data collected.

[0177] Step 2: Data review

[0178] Input: The raw data collected.

[0179] Specific behavior:

[0180] The server reviews the collected data and performs basic analysis to check for formatting inconsistencies and missing values.

[0181] Data processing:

[0182] Standardize data formats and detect inconsistencies and missing values.

[0183] Output: The formatted data.

[0184] Step 3: Data Shaping

[0185] Input: Data with confirmed format.

[0186] Specific behavior:

[0187] The server converts the collected data into a unified format, removes duplicate data, and imputes missing values ​​using data cleansing algorithms.

[0188] Data processing:

[0189] Data is converted, duplicate entries are removed, and missing parts are filled in.

[0190] Output: Clean, formatted data.

[0191] Step 4: Data Summarization

[0192] Input: Clean, formatted data.

[0193] Specific behavior:

[0194] The server uses natural language processing technology to extract important information from the data and generate a summary. It identifies important items (e.g., treatment effects, side effects, etc.) and organizes them into a summary.

[0195] Data processing:

[0196] Natural language processing algorithms extract and summarize information.

[0197] Output: Summary statement.

[0198] Step 5: Data visualization

[0199] Input: Abstract text.

[0200] Specific behavior:

[0201] The server uses tools to generate graphs and charts based on the summarized data and visually display the data, using data visualization tools such as Matplotlib and D3.js.

[0202] Data processing:

[0203] Visualize data using graphs and charting tools.

[0204] Output: Graphs and charts.

[0205] Step 6: Data comparison

[0206] Input: Newly acquired and existing data.

[0207] Specific behavior:

[0208] The server compares the newly acquired data with existing data and analyzes the similarities and differences, using a comparison algorithm to identify differences and commonalities.

[0209] Data processing:

[0210] Compare data and generate analytical results.

[0211] Output: Comparison report.

[0212] Step 7: Update regulatory information

[0213] Input: Latest information obtained from national drug regulatory agency websites.

[0214] Specific behavior:

[0215] The server periodically retrieves the latest regulatory information from regulatory agencies' websites and updates the database using web scraping technology.

[0216] Data processing:

[0217] Integrate acquired regulatory information into existing databases.

[0218] Output: The updated database.

[0219] Step 8: User Interface

[0220] Input: The keyword the user wants to search for (e.g., "new treatment for lung cancer").

[0221] Specific behavior:

[0222] Users use a web interface to search for the data they need, view summaries and reports, and are provided with application templates based on the search results, which users can fill in and complete.

[0223] Data processing:

[0224] View search results and generate reports.

[0225] Output: Data search results, summary, comparison report and application template.

[0226] (Application example 1)

[0227] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0228] In today's world, the number and variety of security incidents are increasing, creating a need for rapid responses. However, the task of collecting, formatting, and summarizing the necessary information from vast amounts of data is extremely time-consuming, making it difficult to respond efficiently. It is also difficult to regularly obtain the latest information on security incidents and take appropriate measures based on that information. Therefore, an effective system that solves these issues is needed.

[0229] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0230] In this invention, the server includes means for automatically collecting clinical and non-clinical data, means for formatting the collected data and converting it into a unified format, means for extracting important information from the formatted data and generating a summary, means for visualizing the summary data in graph or chart format, means for comparing new data with existing data and providing the results in report format, means for periodically obtaining the latest pharmaceutical regulatory information and updating the database, means for a user to use an interface to search for necessary data and view summaries and reports, and means for automatically collecting, formatting, summarizing, visualizing, and comparing security incident data and periodically updating the latest security information. This enables efficient processing of massive amounts of security incident information and enables prompt and appropriate responses.

[0231] "Clinical data" refers to data collected in medical settings regarding patient diagnostic results and treatment.

[0232] "Non-clinical data" refers to data obtained outside of clinical practice, such as experimental results and observational data conducted in a laboratory or research lab.

[0233] "Collection methods" refers to the methods and techniques used to gather specific information, including the process of obtaining the required data from public databases and websites.

[0234] "Format" refers to the process of converting collected data into a unified format and, if necessary, removing duplicate data and completing missing values.

[0235] "Means for generating summaries" refers to the process of extracting important information from data and summarizing it concisely using natural language processing techniques, etc.

[0236] "Visualization means" refers to methods for displaying data in graphs or charts to make it easier to understand visually.

[0237] "Comparative means" refers to the process of matching newly acquired data with existing data and analyzing similarities and differences.

[0238] "Means of providing in report format" refers to a method of reporting comparison results and other important information in the form of documents, electronic files, etc.

[0239] "Means for regularly obtaining and updating regulatory information" refers to the process of regularly collecting the latest guidelines and regulatory information from regulatory agencies in each country and keeping the database up to date.

[0240] "User interface" refers to the screens and operating means that users use to search, view, and manipulate data using a system.

[0241] A "security incident" refers to an accident or incident related to information security, such as unauthorized access or data leakage.

[0242] "Security incident data" refers to data containing information related to security incidents, including specific examples such as attack methods, scope of impact, and countermeasures.

[0243] This invention is a system that streamlines the collection, formatting, summarization, visualization, and comparison of security incident data, as well as the updating of the latest security information. This system consists of a server, terminals, and users. The server is primarily responsible for data collection, automatic formatting, summary generation, data visualization, data comparison, and security information updates, while the terminals function as an interface with the users. Users use the system to search for and view the necessary data and obtain the information necessary for security countermeasures.

[0244] System Overview

[0245] 1. Data Collection

[0246] The server automatically collects data from public security incident databases, specialized blogs, security forums, etc. based on specific keywords (e.g., "phishing attack" or "ransomware"). Specifically, it uses various APIs and web scraping technologies to obtain the necessary data and store it in a database. The technologies used include Python, the Requests library, and BeautifulSoup.

[0247] 2. Data Formatting

[0248] The server reviews the collected data and converts it into a unified format. To ensure data integrity, it removes duplicate data and imputes missing values. MongoDB is used for database management.

[0249] 3. Data Summary

[0250] The server uses natural language processing techniques to extract important information from the data and generate summaries. For example, it uses the NLP library NLTK to extract important keywords from each incident report and provide a summary.

[0251] 4. Data Visualization

[0252] The server uses tools to generate graphs and charts based on the summarized data to visually display the data, using Python's matplotlib library to visualize the distribution of attack techniques as pie charts and frequency of occurrence as histograms.

[0253] 5. Data Comparison

[0254] The server compares the newly acquired data with the existing data, analyzing similarities and differences. The program then presents the results of these comparisons in a report. Statistical analysis libraries such as Scipy are used to collate the old and new data.

[0255] 6. Regulatory Updates

[0256] The server periodically retrieves the latest regulatory information from the websites of security regulatory agencies in each country and updates the database using web scraping technology and APIs.

[0257] 7. User Interface

[0258] Users use a smartphone or a head-mounted display (HMD) to search for the required data and view summaries and reports. The user interface is built using a web browser-based front-end framework (e.g., React).

[0259] Program examples and prompts

[0260] A specific example would be a scenario where a user uses a smartphone or HMD to search for data on the latest phishing attack trends, view the resulting summaries, and refine their company's countermeasures.

[0261] Example prompt sentence:

[0262] "Build a system that collects information on the latest phishing attack trends and provides a formatted summary of the data. Use Python to collect, format, summarize, visualize, compare, and update the latest trend information. Specifically, demonstrate a process for collecting information on phishing attacks from security blogs and forums and generating a summary and comparison report."

[0263] In this way, the present invention functions as a system that efficiently processes security incident data and enables a rapid response.

[0264] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0265] Step 1:

[0266] The server automatically collects data from public security incident databases, specialized blogs, and security forums based on specific security keywords (e.g., "phishing attack" or "ransomware"), using Python and the Requests library to send web requests and BeautifulSoup to parse HTML pages. The input data is the URL of the collected web page, and the output data is the parsed security-related information.

[0267] Step 2:

[0268] The server formats the collected data and converts it into a unified format. Specifically, it removes duplicate entries from the collected data and fills in missing values ​​with reliable default values. MongoDB is used as the database to maintain data integrity. The input data is the collected raw data, and the output data is formatted data in a unified format.

[0269] Step 3:

[0270] The server uses natural language processing (NLP) technology to extract important information from the formatted data and generate a summary. It uses the NLTK library to extract key keywords from the incident report and generate a summary. The input data is the formatted data, and the output data is the summarized information.

[0271] Step 4:

[0272] The server generates graphs and charts based on the summarized data. It uses Python's matplotlib library to visually display the data. Specifically, it displays the distribution of attack techniques as a pie chart and visualizes their occurrence frequency as a histogram. The input data is the summarized data, and the output data is the generated graphs and charts.

[0273] Step 5:

[0274] The server compares the newly acquired data with the existing data and analyzes the similarities and differences. This process uses statistical analysis libraries such as Scipy to collate and analyze the data, and provides the results in the form of a report. The input data is the old and new data, and the output data is a report of the comparison results.

[0275] Step 6:

[0276] The server periodically retrieves the latest regulatory information from the websites of security regulatory agencies in each country and updates the database. Using APIs and web scraping technology, the server collects the latest information and stores it in MongoDB. The input data is regulatory agency website information, and the output data is the updated database information.

[0277] Step 7:

[0278] Users use a smartphone or head-mounted display (HMD) to search for the data they need and view summaries and reports. The user interface is built using a front-end framework such as React to allow users to easily manipulate the data. The input data are search keywords from the user, and the output data are search results and summary reports.

[0279] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0280] The present invention provides a system that streamlines data collection, formatting, summarization, visualization, comparison, and updating of regulatory information in the pharmaceutical application process, and also includes an emotion engine that recognizes user emotions and adjusts interface display and data search results based on the emotions. The system has the following components and functions:

[0281] System Overview

[0282] The system consists of a server, a terminal, and a user. The server is mainly responsible for data collection, automatic formatting, summary generation, data visualization, data comparison, regulatory information updates, and user emotion recognition. The terminal functions as the interface with the user, allowing the user to search and view data through the system and create documents required for drug applications.

[0283] Program processing

[0284] 1. Data Collection

[0285] Subject: Server

[0286] The server uses pre-configured API keys and web scraping techniques to collect data from scientific papers, clinical trial databases, and drug regulatory agency websites. Specifically, the server uses the PubMed API to retrieve paper titles and abstracts related to "cancer treatment" and stores them in the database in JSON format. It also uses web scraping technology to crawl the FDA's latest guidelines page, retrieves and stores the content in text format.

[0287] 2. Data Formatting

[0288] Subject: Server

[0289] The server analyzes the collected data and unifies different data formats. Specifically, it parses JSON-formatted data, removes duplicate data, and fills in missing values ​​with appropriate default values. For example, in the case of paper data, if multiple papers with the same title are obtained, they are merged into one, and if information such as author names or publication dates is missing, it is filled in from reliable external information.

[0290] 3. Data Summary

[0291] Subject: Server

[0292] The server uses natural language processing (NLP) technology to extract important information from the collected data and generate summaries. Specifically, it uses an NLP library (e.g., NLTK or SpaCy) to analyze the main keywords in the paper, as well as descriptions of therapeutic effects and side effects, and generates summaries based on this. The generated summaries are then stored in a database.

[0293] 4. Data Visualization

[0294] Subject: Server

[0295] The server uses tools to generate graphs and charts based on the summarized data for visualization. For example, Matplotlib is used to generate line graphs showing the effectiveness of each treatment and bar graphs showing the frequency of side effects. The generated graphs and charts are converted into a format that can be viewed by the user interface and displayed on a dashboard in real time.

[0296] 5. Data Comparison

[0297] Subject: Server

[0298] The server compares the newly acquired data with existing data and analyzes differences and commonalities. Specifically, it cross-references data on existing cancer treatments with data on new treatments, comparing them in terms of treatment effectiveness, side effects, etc. The comparison results are generated in report format, making them easy for users to understand.

[0299] 6. Regulatory Updates

[0300] Subject: Server

[0301] The server periodically retrieves the latest information from the websites of drug regulatory agencies in each country and updates the database. Specifically, it uses a web scraping tool to retrieve the latest guidelines and regulatory changes from the FDA and EMA, stores them in the database, and sends notifications to users when changes are confirmed.

[0302] 7. Data Search and Browsing

[0303] Subject: User

[0304] Users use the device's web interface to search for and retrieve the data they need. Specifically, they enter keywords such as "new treatments for lung cancer" into the search screen and click the search button to send a request to the server. The server then retrieves relevant summaries and comparison reports from the database and displays them on the user's device.

[0305] 8. Emotion Recognition and Interface Adjustment

[0306] Subject: Server

[0307] The server uses an emotion engine to recognize the user's emotions, for example, by analyzing the user's facial expressions and tone of voice, and then adjusts the priority of data search results and optimizes the interface display.

[0308] Examples:

[0309] If the server detects that a user is feeling stressed while completing a regulatory application, it will simplify the interface and adjust the display to make important information easier to access. If there are too many search results, it will adjust the ranking to prioritize the most relevant data.

[0310] 9. Preparation of pharmaceutical application documents

[0311] Subject: User

[0312] Users create regulatory application documents based on the displayed summaries and comparison reports. Specifically, they download templates provided by the system and create documents by entering the displayed information. They can also submit the created documents for review by experts, who can make revisions based on the feedback.

[0313] In this way, the present invention, equipped with an emotion engine, is a system that not only efficiently collects, formats, summarizes, visualizes, and compares the various data required for the pharmaceutical application process, and updates the latest regulatory information, but also further supports the application process and improves its accuracy and speed by optimizing the interface in consideration of the user's emotions.

[0314] The processing flow will be explained below.

[0315] Step 1: Data collection

[0316] Subject: Server

[0317] The server uses pre-configured API keys and web scraping techniques to collect data from scientific papers, clinical trial databases, and drug regulatory agency websites. Specifically, the server uses the PubMed API to retrieve paper titles and abstracts related to "cancer treatment." The retrieved data is stored in a database in JSON format. It also uses web scraping technology to crawl the FDA's latest guidelines page, retrieve their contents in text format, and store them.

[0318] Step 2: Data Shaping

[0319] Subject: Server

[0320] The server analyzes the collected data and unifies different data formats. Specifically, it parses JSON-formatted data, removes duplicate data, and fills in missing values ​​with appropriate default values. For example, if multiple papers with the same title are retrieved, they are merged into one, and if missing information such as author names or publication dates is provided, it is filled in using reliable external information.

[0321] Step 3: Data Summarization

[0322] Subject: Server

[0323] The server uses natural language processing (NLP) technology to extract important information from the collected data and generate summaries. Specifically, it uses NLP libraries (e.g., NLTK and SpaCy) to analyze each paper's main keywords, therapeutic effects, and side effects. Based on the extracted information, it automatically generates summaries and stores them in a database.

[0324] Step 4: Data visualization

[0325] Subject: Server

[0326] The server uses tools to generate graphs and charts based on the summarized data for visualization. Specifically, Matplotlib is used to generate line graphs showing the effectiveness of each treatment and bar graphs showing the frequency of side effects. The generated graphs and charts are converted into a format that can be viewed by the user interface and displayed on a dashboard in real time.

[0327] Step 5: Data comparison

[0328] Subject: Server

[0329] The server compares the newly acquired data with existing data and analyzes differences and commonalities. Specifically, it cross-references data on existing cancer treatments with data on new treatments to evaluate differences in treatment effectiveness and side effects. The comparison results are generated in a detailed report format that users can easily understand.

[0330] Step 6: Update regulatory information

[0331] Subject: Server

[0332] The server periodically retrieves the latest information from the websites of each country's drug regulatory agency and updates the database. Specifically, it uses a web scraping tool to retrieve the latest guidelines and regulatory changes from the FDA and EMA, and stores them in the database. If any changes are confirmed, the server notifies the user.

[0333] Step 7: Emotion Recognition

[0334] Subject: Server

[0335] The server uses an emotion engine to recognize the user's emotions. Specifically, it analyzes the user's facial expressions, voice tone, input speed, etc. to evaluate the user's emotions (e.g., stress, satisfaction). The results of emotion recognition are used to adjust the interface and prioritize search results.

[0336] Step 8: Interface Adjustment

[0337] Subject: Server

[0338] The server adjusts the interface display content based on the user's emotions recognized by the emotion engine. Specifically, if the user is feeling stressed, the interface will be simplified and important information will be highlighted. The server will also prioritize the display of content that matches the user's interests.

[0339] Step 9: Search and view data

[0340] Subject: User

[0341] Users use the device's web interface to search for and retrieve the data they need. Specifically, they enter keywords such as "new treatments for lung cancer" into the search screen and click the search button to send a request to the server. The server then retrieves relevant summaries and comparison reports from the database and displays them on the user's device.

[0342] Step 10: Prepare regulatory application documents

[0343] Subject: User

[0344] Users create regulatory application documents based on the displayed summaries and comparison reports. Specifically, they download templates provided by the system and create documents by entering the displayed information. They then submit the created documents for review by experts, who provide feedback and make revisions.

[0345] In this way, the present invention, equipped with an emotion engine, is a system that not only efficiently collects, formats, summarizes, visualizes, and compares the various data required for the pharmaceutical application process, and updates the latest regulatory information, but also further supports the application process and improves its accuracy and speed by optimizing the interface in consideration of the user's emotions.

[0346] Example 2

[0347] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0348] The pharmaceutical application process requires a huge amount of data collection, complex data formatting, detailed summarization, effective data visualization, rigorous data comparison, and the latest regulatory information updates, making it extremely time-consuming and labor-intensive. Furthermore, the lack of user-friendly interface adjustments can lead to reduced work efficiency. Therefore, there is a need for a system that can efficiently complete pharmaceutical application procedures without causing stress to users.

[0349] The specific processing by the specific processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for automatically collecting clinical data and non-clinical data, means for formatting the collected data and converting it into a unified format, means for extracting important information from the formatted data and generating a summary, means for visualizing the summary data in the form of graphs or charts, means for comparing new data with existing data and providing the results in the form of a report, means for periodically obtaining the latest pharmaceutical regulatory information and updating the database, means for a user to use an interface to search for necessary data and view summaries and reports, means for analyzing user emotions using an emotion engine and adjusting the interface display and search results, and means for generating prompt sentences to input to the generative AI model. This enables efficient data collection, data formatting, summarization, visualization, comparison, and updating of the latest regulatory information in the pharmaceutical application process, and further enables significant improvement in work efficiency and user experience through interface adjustment based on user emotions.

[0350] "Data Collection Instruments" refers to processes or mechanisms for the automated collection of clinical and non-clinical data.

[0351] "Data formatting means" refers to the process or method of analyzing collected data and converting it into a unified format.

[0352] "Data summarization means" refers to a mechanism or process that extracts important information from formatted data and generates a summary.

[0353] "Data visualization methods" refers to the process of visualizing summarized data in the form of graphs or charts.

[0354] "Data comparison means" refers to a method or technology that compares new data with existing data and provides the results of the comparison in a report format.

[0355] "Regulatory information update means" refers to the process or method of periodically obtaining the latest pharmaceutical regulatory information and updating the database.

[0356] "Data retrieval means" refers to the ability for users to use the interface to retrieve required data and view summaries and reports.

[0357] "Emotion analysis means" refers to a mechanism that uses an emotion engine to analyze user emotions and adjusts interface display and search results based on the results.

[0358] "Prompt generation means" refers to the process or method for generating prompt sentences to be input into a generative AI model.

[0359] "Automatic means" refers to processes or techniques that do not require manual intervention and are operated automatically by the system.

[0360] "User interface" refers to the screens and methods by which a user interacts with a system and searches and views data.

[0361] The present invention is a system that streamlines data collection, formatting, summarization, visualization, comparison, and updating of regulatory information in the pharmaceutical application process. Furthermore, it aims to improve operability by incorporating an emotion engine that recognizes user emotions and adjusts interface display and data search results based on the emotions. The system of the present invention has the following specific configuration:

[0362] System Components

[0363] 1. Server

[0364] The server is the central part of the system and performs the following functions:

[0365] Data collection: Using PubMed API and web scraping techniques, we collect data from scientific articles, clinical trial databases, and drug regulatory agency websites. For example, we use PubMed API to retrieve titles, authors, and abstracts of articles related to "cancer treatment" and store them in a database in JSON format.

[0366] Data Transformation: Analyze the collected data and convert it into a unified format, specifically removing duplicate entries in the JSON data and imputing missing values ​​from external reliable data.

[0367] Data summarization: Using natural language processing (NLP) techniques, we extract important information from the collected data and create summaries. For example, we use NLP libraries (e.g., NLTK and SpaCy) to analyze the main keywords in papers and descriptions of treatment effects and side effects.

[0368] Data visualization: Visualize the data using tools that generate graphs and charts (e.g., Matplotlib) based on the summarized data. The generated graphs and charts are displayed in real time on a dashboard.

[0369] Data comparison: Compare newly acquired data with existing data and analyze differences and commonalities. Generate a report of the comparison results so that users can easily understand them.

[0370] Regulatory information updates: Regularly retrieve the latest information from the websites of national drug regulatory agencies and update the database. For example, use web scraping tools to retrieve the latest guidelines from the FDA and EMA.

[0371] Emotion recognition: An emotion engine is used to analyze emotions from a user's facial expressions and tone of voice to adjust interface display and data search result priorities.

[0372] 2. Terminal

[0373] The terminal acts as an interface with the user:

[0374] Data search and browsing: Users use a web interface to search for and retrieve the data they need. For example, they enter keywords such as "new treatments for lung cancer" into the search screen and click the search button to send a request to the server. The server retrieves relevant summaries and comparison reports from the database and displays them on the terminal.

[0375] Interface Adaptation: If the emotion engine detects stress in the user, it will automatically simplify the interface and adjust to make important information easier to access.

[0376] 3. Users

[0377] Users use the system to:

[0378] Data retrieval: Users can search for the required data through a web interface and view summaries and reports.

[0379] Creating regulatory application documents: Users create regulatory application documents based on the displayed summary and comparison reports. They download templates provided by the system and enter the required information. They submit the completed documents for expert review and make revisions based on the feedback.

[0380] Specific examples

[0381] 1. The server uses the PubMed API to search for papers on "lung cancer treatments" and retrieves their titles, authors, and abstracts. Then, it uses a web crawler to parse the FDA's "New Treatment Guidelines" page and saves the content in text format.

[0382] 2. The server eliminates duplicates from the collected data and completes missing author names, publication dates, etc. from external databases.

[0383] 3. Using natural language processing technology, generate a summary from the collected papers: "This new drug improved the survival rate of lung cancer patients by 20%, but nausea was reported as a side effect."

[0384] 4. Use Matplotlib to generate a line graph showing the effectiveness of "Lung Cancer Treatment A" and a bar graph showing the frequency of side effects. Display these graphs on a dashboard.

[0385] 5. Compare the effectiveness and side effects of "existing treatment A" and "new treatment B" and generate a comparative report showing that new treatment B is slightly less effective but has fewer side effects.

[0386] 6. Use a web crawler to regularly check for the latest EMA guidelines and store new regulatory information in a database.

[0387] 7. If users are experiencing stress while completing regulatory submissions, simplify the interface, make important information more accessible, and adjust the ranking of search results to surface more relevant information.

[0388] Example prompts to input to the generative AI model

[0389] "Collect the latest research data on lung cancer treatments, summarize and compare them, and use them to draft regulatory applications."

[0390] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0391] Step 1:

[0392] Data collection

[0393] The server collects data using the PubMed API and a web scraping tool. Specifically, it retrieves the titles, authors, and abstracts of papers related to "lung cancer treatment" from the PubMed API. It also uses the web scraping tool to crawl the FDA's latest guidelines page and saves the content in text format.

[0394] Input: Keyword "lung cancer treatment", API key, website URL

[0395] Output: Paper data (JSON format), guideline text data

[0396] Specific operation: The server collects paper data on "lung cancer treatment" from the PubMed API, and at the same time uses a web crawler to retrieve and save the contents of the FDA's latest guidelines page.

[0397] Step 2:

[0398] Data Formatting

[0399] The server parses the collected JSON-formatted data and unifies different formats, specifically removing duplicate data and completing missing values ​​from external information.

[0400] Input: Collected paper data (JSON format), guideline text data

[0401] Output: Formatted data (Uniform JSON format)

[0402] What it does: The server reads a file such as "Lung Cancer Treatments.pdf," merges duplicate entries, and completes missing author names and publication dates.

[0403] Step 3:

[0404] Data Summary

[0405] The server uses natural language processing (NLP) technology to analyze the collected data and extract important information. It uses NLP libraries (NLTK and SpaCy) to analyze the main keywords of the papers, as well as descriptions of therapeutic effects and side effects, and generates summaries.

[0406] Input: Formatted data (Uniform JSON format)

[0407] Output: Summary data (text format)

[0408] Specific operation: The server analyzes the abstract of a paper on "lung cancer treatment" using SpaCy, generates a summary, and saves it.

[0409] Step 4:

[0410] Data Visualization

[0411] The server generates graphs and charts based on the summarized data, using Matplotlib to graphically display the effectiveness of each treatment and the frequency of side effects.

[0412] Input: Summary data (text format)

[0413] Output: Graphs and charts (image format)

[0414] Specific operation: Based on the summarized data, the server uses Matplotlib to create a line graph showing the effectiveness of "Lung Cancer Treatment A" and a bar graph showing the frequency of side effects.

[0415] Step 5:

[0416] Data comparison

[0417] The server cross-references the newly acquired data with the existing data, analyzes the differences and commonalities between them, and generates a report of the comparison results.

[0418] Input: Summary data (text format), existing data

[0419] Output: Comparison report (text format)

[0420] Specific operation: The server cross-references the effectiveness and side effects of "existing treatment A" and "new treatment B" and generates a report stating that treatment B is slightly less effective but has fewer side effects.

[0421] Step 6:

[0422] Regulatory updates

[0423] The server periodically scrapes the websites of each country's drug regulatory agency to collect the latest information and update the database.

[0424] Input: Regulatory agency website URL

[0425] Output: Latest regulatory information (text format)

[0426] What it does: The server uses a web crawler to crawl the latest FDA and EMA guideline pages and stores the necessary information in a database.

[0427] Step 7:

[0428] Data Search and Viewing

[0429] Users use the device's web interface to search for and retrieve the data they need. By entering search keywords and clicking the search button, the request is sent to the server, which retrieves the relevant data and displays it on the user's device.

[0430] Input: Search keyword

[0431] Output: Relevant data (summary and comparison reports)

[0432] Specific operation: When a user enters "new treatments for lung cancer" into the search box and performs a search, the server displays relevant summaries and reports on the device.

[0433] Step 8:

[0434] Emotion Recognition and Interface Adjustment

[0435] The server uses an emotion engine to analyze the user's emotions, including facial expressions and tone of voice, and adjusts the interface display and search result priorities based on the results.

[0436] Input: User's facial expression data, voice data

[0437] Output: Tailored interface, prioritized search results

[0438] What it does: The server collects data from the user's webcam and microphone and simplifies the interface if the user is stressed.

[0439] Step 9:

[0440] Preparation of pharmaceutical application documents

[0441] Users create regulatory application documents based on the displayed summary and comparison reports. They download templates provided by the system, enter the required information, and submit the completed documents for expert review, where they can make revisions based on the feedback.

[0442] Inputs: Summary data, comparison reports, templates

[0443] Output: Pharmaceutical application documents

[0444] What it does: A user downloads a template from the system, copies information from the summary and comparison reports to complete the application, and then submits it to an expert for feedback and makes any necessary revisions.

[0445] (Application example 2)

[0446] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0447] Conventional systems do not efficiently collect, organize, summarize, visualize, and compare clinical and non-clinical data, or update and search regulatory information. Furthermore, they do not allow for interface adjustments that take into account users' emotions and stress levels, or optimization of productivity and work environments. This makes it difficult for managers to grasp the status of workers in real time and provide an optimal work environment.

[0448] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0449] In this invention, the server includes a means for automatically collecting clinical and non-clinical data, a means for formatting the data and converting it into a unified format, and a means for extracting important information and generating summaries. This allows for efficient processing of various types of data. It also includes a means for collecting biometric information from workers, analyzing their emotions and stress levels, a means for adjusting the interface display content based on their emotions and stress levels, and a means for visualizing production data and optimizing production efficiency and the work environment. This allows managers to understand the status of workers in real time and provide a comfortable and efficient work environment.

[0450] "Clinical Data" refers to patient medical information collected in a medical setting, including data related to diagnosis, treatment, and follow-up.

[0451] "Non-clinical data" refers to research data obtained outside of clinical trials, including data obtained from animal experiments, cell experiments, etc.

[0452] "Data transformation" refers to the process of analyzing collected data and unifying different formats and forms.

[0453] "Unified format" refers to a format that converts data obtained from different sources into a consistent format.

[0454] "Important information extraction" refers to the process of selecting the necessary data from a large amount of data according to a specific purpose or requirement.

[0455] A "summary" is a document or text that succinctly summarizes the main points or essence of the collected data.

[0456] "Worker biometric information" refers to data about a worker's physical condition, such as heart rate, body temperature, and stress level.

[0457] Analyzing "emotions and stress levels" refers to the process of assessing a worker's emotional state and stress level from collected biometric information.

[0458] "Adjusting the interface display content" refers to the process of changing the system display content and operation method according to the user's condition to provide the optimal user experience.

[0459] "Production data" refers to data related to the manufacturing process of products in factories and production lines, including operating hours, failure rates, production efficiency, etc.

[0460] "Production efficiency" refers to an indicator that shows how much product can be produced using a certain amount of resources (time, manpower, materials).

[0461] "Working environment" refers to the conditions in the place where workers perform their work, including temperature, humidity, noise level, etc.

[0462] The present invention provides a system for achieving efficient data management and optimizing the working environment at a production site. The system includes the following hardware and software.

[0463] Hardware

[0464] IoT devices and sensors: installed to collect temperature, humidity, and vibration data.

[0465] Wearable devices: Worn to collect workers' biometric information (heart rate, stress levels, etc.).

[0466] Smart glasses: worn by workers, they display the information they need in real time while they work.

[0467] software

[0468] Server: Responsible for data collection, formatting, analysis, visualization, comparison, and regulatory information updates.

[0469] Data Collection API: Collects data from sensors and wearable devices using the MQTT protocol or Bluetooth communication.

[0470] Parsing libraries: Use natural language processing libraries such as NLTK or SpaCy to generate summaries of your data.

[0471] Visualization tools: Use Matplotlib and Plotly to visualize data in the form of graphs and charts.

[0472] Machine learning model: TENSORFLOW® is used to analyze workers' emotions and stress levels.

[0473] System Details

[0474] 1. Data Collection:

[0475] The server collects data in real time from IoT devices and sensors, capturing temperature, humidity, and vibration data using the MQTT protocol and storing them in a database. It also collects biometric data from workers' smart glasses and wearable devices via Bluetooth and transmits it to the server.

[0476] 2. Data Formatting:

[0477] The server parses the collected data and converts it into a standard format, cleaning it using Python's Pandas library, removing duplicates, and imputing missing values.

[0478] 3. Data Summary:

[0479] Use natural language processing libraries (e.g., NLTK, SpaCy) to extract important information from the collected data and generate summaries. For example, analyze temperature fluctuations obtained from multiple sensor data and generate a summary.

[0480] 4. Data visualization:

[0481] The server generates graphs and charts in real time based on the collected data, using Matplotlib and Plotly to visually display the operation status of the production line and the stress levels of workers.

[0482] 5. Data comparison:

[0483] The server compares and analyzes newly acquired data with existing data to identify anomalies and patterns, particularly by comparing machine operating hours and failure rates with past data.

[0484] 6. Regulatory Updates:

[0485] Regularly obtain information on each country's occupational safety regulations and environmental standards using a web scraping tool (e.g., BeautifulSoup) and update the database. If there are any changes in regulations, notify the administrator.

[0486] 7. Emotion Recognition and Interface Adjustment:

[0487] The server analyzes the worker's emotions and stress levels from facial expressions and biometric data. It uses TensorFlow to classify emotions and adjusts the interface display if it recognizes a high level of stress. For example, if a worker's stress level is high, it changes the work priority and displays an alert suggesting a break.

[0488] 8. Optimizing production efficiency and working environment:

[0489] The server analyzes the collected production data and optimizes production efficiency and the working environment, thereby improving productivity in the factory and providing a comfortable working environment for workers.

[0490] Specific examples

[0491] For example, when Worker A is wearing smart glasses while working, the glasses measure his heart rate and stress level using a data collection API sent from the server. Using TensorFlow, if the data determines that Worker A's stress level is high, the server changes the work priority and immediately displays an alert recommending a break. Also, if the factory temperature or humidity data exceeds the standard, the server simultaneously sends an alert and issues instructions for improving the environment.

[0492] Example prompts for generative AI models

[0493] Based on Worker A's latest biometric data, generate a response plan for when stress levels are high.

[0494] Working environment data: Temperature: 28°C, Humidity: 60%, Vibration: Low

[0495] Vital Data: Heart rate: 80 bpm, Stress level: High

[0496] In this way, the system can maintain production efficiency while ensuring worker health and comfort.

[0497] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0498] Step 1:

[0499] The server collects environmental data (temperature, humidity, vibration) from IoT devices and sensors.

[0500] Input: Data sent from each sensor via MQTT protocol.

[0501] Processing: The server parses the received data and stores it in the database.

[0502] Output: Formatted environmental data (Temperature: 25°C, Humidity: 60%, Vibration: Low).

[0503] Step 2:

[0504] The server collects biometric information (heart rate, stress level) from smart glasses or wearable devices worn by workers.

[0505] Input: Real-time biometric data transmitted via Bluetooth.

[0506] Processing: The server receives the biometric data and stores it in a database.

[0507] Output: Formatted biometric data (heart rate: 80 bpm, stress level: high).

[0508] Step 3:

[0509] The server analyzes the collected environmental and biological data and converts it into a unified format.

[0510] Input: Environmental and biological data stored in a database.

[0511] Processing: Using Python's Pandas library, the data is cleaned, missing values ​​are imputed, and it is converted into a standard format.

[0512] Output: A uniformly formatted dataset.

[0513] Step 4:

[0514] The server extracts important information from the dataset and generates a summary.

[0515] Input: A uniformly formatted dataset.

[0516] Processing: Use NLTK or SpaCy to analyze the data and generate summaries.

[0517] Output: Summarized data.

[0518] Step 5:

[0519] The server visualizes the summarized data and displays it on the factory dashboard.

[0520] Input: Summarized data.

[0521] Processing: Visualize the data in the form of graphs and charts using Matplotlib and Plotly.

[0522] Output: Graphs and charts displayed on a dashboard.

[0523] Step 6:

[0524] The server compares the newly acquired data with past data to detect any anomalies.

[0525] Input: Newly acquired data and existing datasets.

[0526] Processing: Use Pandas and SciPy to compare data and identify anomalies and patterns.

[0527] Output: Anomaly detection results report.

[0528] Step 7:

[0529] The server regularly obtains each country's occupational safety regulations and environmental standards using a web scraping tool and updates the database.

[0530] Input: Websites that provide regulatory information.

[0531] Processing: Use BeautifulSoup to perform web scraping and store new regulatory information in a database.

[0532] Output: Updated regulatory information.

[0533] Step 8:

[0534] The server analyzes the worker's biometric data and adjusts the interface display based on their emotions and stress levels.

[0535] Input: Worker's biometric data.

[0536] Processing: TensorFlow is used to analyze biometric data and classify emotions and stress levels, while simultaneously adjusting the interface display.

[0537] Output: The adjusted interface display.

[0538] Step 9:

[0539] The server analyzes production data and optimizes production efficiency and the working environment.

[0540] Input: Production data and work environment data.

[0541] Processing: Using data analysis tools, optimization algorithms are applied to evaluate production efficiency.

[0542] Output: Improvement suggestions and alert messages.

[0543] For example, if Worker A's biometric data is analyzed and it is determined that his stress level is high, an alert will be displayed immediately recommending that he take a break. Also, if the temperature or humidity in the factory exceeds the standard, instructions to improve the environment will be automatically sent.

[0544] Example prompts for generative AI models

[0545] Based on Worker A's latest biometric data, generate a response plan for when stress levels are high.

[0546] Working environment data: Temperature: 28°C, Humidity: 60%, Vibration: Low

[0547] Vital Data: Heart rate: 80 bpm, Stress level: High

[0548] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0549] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs 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.

[0550] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0551] [Second embodiment]

[0552] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0553] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0554] 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).

[0555] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, 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 microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0556] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0557] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0558] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0559] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0560] The specific processing program 56 is an example of a "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 processing 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.

[0561] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0562] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0563] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0564] The present invention is a system for streamlining data collection, formatting, summarization, visualization, comparison, and updating of regulatory information in the pharmaceutical application process. This system has the following components and functions.

[0565] System Overview

[0566] The system consists of a server, terminals, and users. The server is primarily responsible for data collection, automatic formatting, summary generation, data visualization, data comparison, and regulatory information updates, while the terminals function as an interface with users. Users use the system to search for and view the necessary data and create the documents required for pharmaceutical applications.

[0567] Program processing

[0568] 1. Data Collection

[0569] Subject: Server

[0570] The server automatically collects data based on pre-defined keywords from publicly available scientific papers, clinical trial databases, drug regulatory agency websites, etc. Specifically, it uses various APIs and web scraping technologies to retrieve the required data and store it in a database.

[0571] Examples:

[0572] The server retrieves the titles, authors, abstracts, and full texts of relevant articles from PubMed using the keyword "cancer treatment" and stores them in a database.

[0573] 2. Data Formatting

[0574] Subject: Server

[0575] The server reviews the collected data and converts it into a unified format, removing duplicates and imputing missing values ​​to ensure data integrity.

[0576] Examples:

[0577] The server analyzes multiple article data retrieved from PubMed, removes duplicate entries, and adds reliable default values ​​for missing entries.

[0578] 3. Data Summary

[0579] Subject: Server

[0580] The server uses natural language processing technology to extract important information from the data and generate a summary of key points (e.g., treatment effects, side effects, etc.).

[0581] Examples:

[0582] The server extracts information about each paper's key keywords, therapeutic effects, and side effects, and provides it as a summary.

[0583] 4. Data Visualization

[0584] Subject: Server

[0585] The server uses tools to generate graphs and charts based on the summarized data to visually display the data.

[0586] Examples:

[0587] The server visualizes the treatment efficacy data as a line graph and displays the frequency of side effects as a bar graph.

[0588] 5. Data Comparison

[0589] Subject: Server

[0590] The server compares the newly acquired data with the existing data, analyzes similarities and differences, and provides the comparison results in the form of a report.

[0591] Examples:

[0592] The server compares data on existing cancer treatments with data on newly acquired treatments and generates a detailed report on differences in treatment effectiveness and side effects.

[0593] 6. Regulatory Updates

[0594] Subject: Server

[0595] The server periodically retrieves the latest regulatory information from the websites of each country's drug regulatory agency and updates the database.

[0596] Examples:

[0597] The server uses web scraping technology to obtain the latest US FDA guidelines and imports them into the database for updating.

[0598] 7. User Interface

[0599] Subject: User

[0600] Users use a web interface to search for the data they need, view summaries and reports, and generate regulatory submissions based on the information provided.

[0601] Examples:

[0602] The user searches for data on "new treatments for lung cancer" in the interface, checks the obtained summary and comparison reports, and downloads a template for pharmaceutical application documents based on the information and fills in the required fields to complete the application.

[0603] In this way, the present invention is a system that automates application procedures and improves accuracy and speed by efficiently collecting, formatting, summarizing, visualizing, and comparing the various data required for the pharmaceutical application process, as well as updating the latest regulatory information.

[0604] The processing flow will be explained below.

[0605] Step 1: Data collection

[0606] Subject: Server

[0607] The server uses pre-configured API keys and web scraping techniques to collect data from scientific papers, clinical trial databases, and drug regulatory agency websites. Specifically, the server uses the PubMed API to retrieve paper titles and abstracts related to "cancer treatment" and stores them in a database in JSON format. It also uses web scraping technology to crawl the FDA's latest guidelines page, retrieves their contents in text format, and stores them.

[0608] Step 2: Data Shaping

[0609] Subject: Server

[0610] The server analyzes the collected data and unifies different data formats. Specifically, it parses JSON-formatted data, removes duplicate data, and fills in missing values ​​with appropriate default values. For example, in the case of paper data, if multiple papers with the same title are obtained, they are merged into one, and if information such as author names or publication dates is missing, it is filled in from reliable external information.

[0611] Step 3: Data Summarization

[0612] Subject: Server

[0613] The server uses natural language processing (NLP) technology to extract important information from the collected data and generate summaries. Specifically, it uses an NLP library (e.g., NLTK or SpaCy) to analyze the main keywords in the paper, as well as descriptions of therapeutic effects and side effects, and generates summaries based on this. The generated summaries are then stored in a database.

[0614] Step 4: Data visualization

[0615] Subject: Server

[0616] The server uses tools to generate graphs and charts based on the summarized data for visualization. For example, Matplotlib is used to generate line graphs showing the effectiveness of each treatment and bar graphs showing the frequency of side effects. The generated graphs and charts are converted into a format that can be viewed by the user interface and displayed on a dashboard in real time.

[0617] Step 5: Data comparison

[0618] Subject: Server

[0619] The server compares the newly acquired data with existing data and analyzes differences and commonalities. Specifically, it cross-references data on existing cancer treatments with data on new treatments, comparing them in terms of treatment effectiveness, side effects, etc. The comparison results are generated in report format, making them easy for users to understand.

[0620] Step 6: Update regulatory information

[0621] Subject: Server

[0622] The server periodically retrieves the latest information from the websites of each country's drug regulatory agency and updates the database. Specifically, it uses a web scraping tool to retrieve the latest guidelines and regulatory changes from the FAA and EMA, stores them in the database, and sends notifications to users when changes are confirmed.

[0623] Step 7: Search and view data

[0624] Subject: User

[0625] Users use the device's web interface to search for and retrieve the data they need. Specifically, they enter keywords such as "new treatments for lung cancer" into the search screen and click the search button to send a request to the server. The server then retrieves relevant summaries and comparison reports from the database and displays them on the user's device.

[0626] Step 8: Prepare regulatory submission documents

[0627] Subject: User

[0628] Users create regulatory application documents based on the displayed summaries and comparison reports. Specifically, they download templates provided by the system and create documents by entering the displayed information. They can also submit the created documents for review by experts, who can make revisions based on the feedback.

[0629] Example 1

[0630] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0631] The current pharmaceutical application process involves a wide range of tasks, including collecting, organizing, summarizing, visualizing, comparing, and updating regulatory information, and is extremely time-consuming and labor-intensive. Furthermore, the quality and consistency of the collected data varies, and the accuracy and efficiency of application document preparation is not fully guaranteed. This leads to delays in the application process and the submission of incorrect information.

[0632] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0633] In this invention, the server includes, in data collection, means for automatically obtaining data from publicly available information sources based on preset keywords, means for reviewing the collected data and checking for format inconsistencies and missing values, means for formatting the collected data, converting it into a unified format, deleting duplicate data, and imputing missing values, means for extracting important information from the collected data using natural language processing technology and generating a summary, means for visualizing the summary data in the form of a graph or chart, means for comparing newly obtained data with existing data, analyzing similarities and differences, and providing the results in the form of a report, means for periodically obtaining the latest regulatory information and updating the database, means for a user to use an interface to search for necessary data and view summaries and reports, and means for providing application document templates based on the search results and for the user to enter the necessary information to create them. This makes it possible to improve the efficiency and accuracy of the pharmaceutical application process.

[0634] "Data harvesting" refers to the automatic retrieval of data from publicly available sources based on pre-defined keywords.

[0635] "Format inconsistency" refers to a situation where collected data exists in different structures or formats.

[0636] "Missing values" refers to a state in which some data is missing or necessary information is not entered.

[0637] "Duplicate data" refers to a situation in which multiple pieces of data with the same content exist.

[0638] "Missing value imputation" refers to filling in missing data based on predictions or reliable sources.

[0639] "Natural language processing" refers to the technology for understanding, analyzing, and generating human language.

[0640] "Extracting important information" refers to extracting only specific, necessary information from collected data.

[0641] A "summary" is a short summary of the main points from a long piece of text or a large amount of data.

[0642] "Visualization" refers to the display of data in a visual format such as a graph or chart.

[0643] "Report format" refers to a format in which the results of analysis and comparison are systematically summarized in documents and charts.

[0644] "Regulatory information" refers to information on norms and standards provided by public institutions, such as laws, regulations, and guidelines.

[0645] "Database updating" refers to adding new information to existing data and keeping it up to date.

[0646] "User interface" refers to the screens and controls that allow a user to interact with a computer system or application.

[0647] "Application template" refers to a template for application documents in a predefined format.

[0648] This invention is a system that streamlines data collection, formatting, summarization, visualization, comparison, and updating of regulatory information in the pharmaceutical application process. This system is mainly composed of a server, terminals, and users. The specific system structure and operation are described in detail below.

[0649] System Configuration

[0650] server:

[0651] The server is responsible for data collection, automatic formatting, summary generation, data visualization, data comparison, and regulatory information updates. The server uses the following hardware and software:

[0652] Hardware: High-performance server (e.g., Xeon-based server)

[0653] Software: Python, natural language processing libraries (e.g., NLTK, SpaCy), data visualization tools (e.g., Matplotlib, D3.js)

[0654] Device:

[0655] The terminal provides the interface through which the user interacts with the system. The terminal uses the following hardware and software:

[0656] Hardware: Personal computers, tablets

[0657] Software: Web browser (e.g., Google Chrome, Firefox)

[0658] User:

[0659] Users are drug developers and regulatory officials involved in the regulatory application process, who use the system to search for the data they need, view summaries and reports, and prepare application documents.

[0660] How it works

[0661] 1. Data Collection:

[0662] The server collects data based on pre-defined keywords from published scientific papers, clinical trial databases, drug regulatory agency websites, etc., using various API calls and web scraping techniques.

[0663] Examples:

[0664] The server retrieves the titles, authors, abstracts, and full texts of relevant articles from PubMed using the keyword "cancer treatment" and stores them in a database.

[0665] 2. Data Formatting:

[0666] The server reviews the collected data and converts it into a unified format, removing duplicates and imputing missing values ​​to ensure data integrity.

[0667] Examples:

[0668] The server analyzes multiple article data retrieved from PubMed, removes duplicate entries, and adds reliable default values ​​for missing entries.

[0669] 3. Data Summary:

[0670] The server uses natural language processing techniques to extract key information from the data and generate a summary.

[0671] Examples:

[0672] The server extracts information about each paper's key keywords, therapeutic effects, and side effects, and provides it as a summary.

[0673] 4. Data visualization:

[0674] The server uses tools to generate graphs and charts based on the summarized data to visually display the data.

[0675] Examples:

[0676] The server visualizes the treatment efficacy data as a line graph and displays the frequency of side effects as a bar graph.

[0677] 5. Data comparison:

[0678] The server compares the newly acquired data with the existing data, analyzes similarities and differences, and provides the results in a report.

[0679] Examples:

[0680] The server compares data on existing cancer treatments with data on newly acquired treatments and generates a detailed report on differences in treatment effectiveness and side effects.

[0681] 6. Regulatory Updates:

[0682] The server periodically retrieves the latest regulatory information from the websites of each country's drug regulatory agency and updates the database.

[0683] Examples:

[0684] The server uses web scraping technology to obtain the latest US FDA guidelines and imports them into the database for updating.

[0685] 7. User Interface:

[0686] Users use a web interface to search for the data they need, view summaries and reports, and download application templates based on their search results, then fill in the required fields to complete the application.

[0687] Examples:

[0688] Users can search for data on "new treatments for lung cancer" in the interface, check the resulting summary and comparison reports, and download a template for a pharmaceutical application based on the information, then fill in the required fields to create it.

[0689] Prompt Sentence Examples

[0690] "Collect the latest articles on 'cancer treatment' from PubMed and create summaries that extract key keywords, therapeutic effects, and side effects."

[0691] As described above, this system comprehensively covers the entire process from data collection to preparation of application documents, dramatically improving the efficiency and accuracy of pharmaceutical applications.

[0692] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0693] Step 1: Data collection

[0694] Input: Predefined keywords (e.g., "cancer treatment").

[0695] Specific behavior:

[0696] The server automatically collects data from various data sources (e.g., scientific paper databases, clinical trial databases, regulatory agency websites, etc.) based on configured keywords, using API calls and web scraping techniques.

[0697] Data processing:

[0698] The acquired data is stored in a temporary database.

[0699] Output: The raw data collected.

[0700] Step 2: Data review

[0701] Input: The raw data collected.

[0702] Specific behavior:

[0703] The server reviews the collected data and performs basic analysis to check for formatting inconsistencies and missing values.

[0704] Data processing:

[0705] Standardize data formats and detect inconsistencies and missing values.

[0706] Output: The formatted data.

[0707] Step 3: Data Shaping

[0708] Input: Data with confirmed format.

[0709] Specific behavior:

[0710] The server converts the collected data into a unified format, removes duplicate data, and imputes missing values ​​using data cleansing algorithms.

[0711] Data processing:

[0712] Data is converted, duplicate entries are removed, and missing parts are filled in.

[0713] Output: Clean, formatted data.

[0714] Step 4: Data Summarization

[0715] Input: Clean, formatted data.

[0716] Specific behavior:

[0717] The server uses natural language processing technology to extract important information from the data and generate a summary. It identifies important items (e.g., treatment effects, side effects, etc.) and organizes them into a summary.

[0718] Data processing:

[0719] Natural language processing algorithms extract and summarize information.

[0720] Output: Summary statement.

[0721] Step 5: Data visualization

[0722] Input: Abstract text.

[0723] Specific behavior:

[0724] The server uses tools to generate graphs and charts based on the summarized data and visually display the data, using data visualization tools such as Matplotlib and D3.js.

[0725] Data processing:

[0726] Visualize data using graphs and charting tools.

[0727] Output: Graphs and charts.

[0728] Step 6: Data comparison

[0729] Input: Newly acquired and existing data.

[0730] Specific behavior:

[0731] The server compares the newly acquired data with existing data and analyzes the similarities and differences, using a comparison algorithm to identify differences and commonalities.

[0732] Data processing:

[0733] Compare data and generate analytical results.

[0734] Output: Comparison report.

[0735] Step 7: Update regulatory information

[0736] Input: Latest information obtained from national drug regulatory agency websites.

[0737] Specific behavior:

[0738] The server periodically retrieves the latest regulatory information from regulatory agencies' websites and updates the database using web scraping technology.

[0739] Data processing:

[0740] Integrate acquired regulatory information into existing databases.

[0741] Output: The updated database.

[0742] Step 8: User Interface

[0743] Input: The keyword the user wants to search for (e.g., "new treatment for lung cancer").

[0744] Specific behavior:

[0745] Users use a web interface to search for the data they need, view summaries and reports, and are provided with application templates based on the search results, which users can fill in and complete.

[0746] Data processing:

[0747] View search results and generate reports.

[0748] Output: Data search results, summary, comparison report and application template.

[0749] (Application example 1)

[0750] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0751] In today's world, the number and variety of security incidents are increasing, creating a need for rapid responses. However, the task of collecting, formatting, and summarizing the necessary information from vast amounts of data is extremely time-consuming, making it difficult to respond efficiently. It is also difficult to regularly obtain the latest information on security incidents and take appropriate measures based on that information. Therefore, an effective system that solves these issues is needed.

[0752] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0753] In this invention, the server includes means for automatically collecting clinical and non-clinical data, means for formatting the collected data and converting it into a unified format, means for extracting important information from the formatted data and generating a summary, means for visualizing the summary data in graph or chart format, means for comparing new data with existing data and providing the results in report format, means for periodically obtaining the latest pharmaceutical regulatory information and updating the database, means for a user to use an interface to search for necessary data and view summaries and reports, and means for automatically collecting, formatting, summarizing, visualizing, and comparing security incident data and periodically updating the latest security information. This enables efficient processing of massive amounts of security incident information and enables prompt and appropriate responses.

[0754] "Clinical data" refers to data collected in medical settings regarding patient diagnostic results and treatment.

[0755] "Non-clinical data" refers to data obtained outside of clinical practice, such as experimental results and observational data conducted in a laboratory or research lab.

[0756] "Collection methods" refers to the methods and techniques used to gather specific information, including the process of obtaining the required data from public databases and websites.

[0757] "Format" refers to the process of converting collected data into a unified format and, if necessary, removing duplicate data and completing missing values.

[0758] "Means for generating summaries" refers to the process of extracting important information from data and summarizing it concisely using natural language processing techniques, etc.

[0759] "Visualization means" refers to methods for displaying data in graphs or charts to make it easier to understand visually.

[0760] "Comparative means" refers to the process of matching newly acquired data with existing data and analyzing similarities and differences.

[0761] "Means of providing in report format" refers to a method of reporting comparison results and other important information in the form of documents, electronic files, etc.

[0762] "Means for regularly obtaining and updating regulatory information" refers to the process of regularly collecting the latest guidelines and regulatory information from regulatory agencies in each country and keeping the database up to date.

[0763] "User interface" refers to the screens and operating means that users use to search, view, and manipulate data using a system.

[0764] A "security incident" refers to an accident or incident related to information security, such as unauthorized access or data leakage.

[0765] "Security incident data" refers to data containing information related to security incidents, including specific examples such as attack methods, scope of impact, and countermeasures.

[0766] This invention is a system that streamlines the collection, formatting, summarization, visualization, and comparison of security incident data, as well as the updating of the latest security information. This system consists of a server, terminals, and users. The server is primarily responsible for data collection, automatic formatting, summary generation, data visualization, data comparison, and security information updates, while the terminals function as an interface with the users. Users use the system to search for and view the necessary data and obtain the information necessary for security countermeasures.

[0767] System Overview

[0768] 1. Data Collection

[0769] The server automatically collects data from public security incident databases, specialized blogs, security forums, etc. based on specific keywords (e.g., "phishing attack" or "ransomware"). Specifically, it uses various APIs and web scraping technologies to obtain the necessary data and store it in a database. The technologies used include Python, the Requests library, and BeautifulSoup.

[0770] 2. Data Formatting

[0771] The server reviews the collected data and converts it into a unified format. To ensure data integrity, it removes duplicate data and imputes missing values. MongoDB is used for database management.

[0772] 3. Data Summary

[0773] The server uses natural language processing techniques to extract important information from the data and generate summaries. For example, it uses the NLP library NLTK to extract important keywords from each incident report and provide a summary.

[0774] 4. Data Visualization

[0775] The server uses tools to generate graphs and charts based on the summarized data to visually display the data, using Python's matplotlib library to visualize the distribution of attack techniques as pie charts and frequency of occurrence as histograms.

[0776] 5. Data Comparison

[0777] The server compares the newly acquired data with the existing data, analyzing similarities and differences. The program then presents the results of these comparisons in a report. Statistical analysis libraries such as Scipy are used to collate the old and new data.

[0778] 6. Regulatory Updates

[0779] The server periodically retrieves the latest regulatory information from the websites of security regulatory agencies in each country and updates the database using web scraping technology and APIs.

[0780] 7. User Interface

[0781] Users use a smartphone or a head-mounted display (HMD) to search for the required data and view summaries and reports. The user interface is built using a web browser-based front-end framework (e.g., React).

[0782] Program examples and prompts

[0783] A specific example would be a scenario where a user uses a smartphone or HMD to search for data on the latest phishing attack trends, view the resulting summaries, and refine their company's countermeasures.

[0784] Example prompt sentence:

[0785] "Build a system that collects information on the latest phishing attack trends and provides a formatted summary of the data. Use Python to collect, format, summarize, visualize, compare, and update the latest trend information. Specifically, demonstrate a process for collecting information on phishing attacks from security blogs and forums and generating a summary and comparison report."

[0786] In this way, the present invention functions as a system that efficiently processes security incident data and enables a rapid response.

[0787] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0788] Step 1:

[0789] The server automatically collects data from public security incident databases, specialized blogs, and security forums based on specific security keywords (e.g., "phishing attack" or "ransomware"), using Python and the Requests library to send web requests and BeautifulSoup to parse HTML pages. The input data is the URL of the collected web page, and the output data is the parsed security-related information.

[0790] Step 2:

[0791] The server formats the collected data and converts it into a unified format. Specifically, it removes duplicate entries from the collected data and fills in missing values ​​with reliable default values. MongoDB is used as the database to maintain data integrity. The input data is the collected raw data, and the output data is formatted data in a unified format.

[0792] Step 3:

[0793] The server uses natural language processing (NLP) technology to extract important information from the formatted data and generate a summary. It uses the NLTK library to extract key keywords from the incident report and generate a summary. The input data is the formatted data, and the output data is the summarized information.

[0794] Step 4:

[0795] The server generates graphs and charts based on the summarized data. It uses Python's matplotlib library to visually display the data. Specifically, it displays the distribution of attack techniques as a pie chart and visualizes their occurrence frequency as a histogram. The input data is the summarized data, and the output data is the generated graphs and charts.

[0796] Step 5:

[0797] The server compares the newly acquired data with the existing data and analyzes the similarities and differences. This process uses statistical analysis libraries such as Scipy to collate and analyze the data, and provides the results in the form of a report. The input data is the old and new data, and the output data is a report of the comparison results.

[0798] Step 6:

[0799] The server periodically retrieves the latest regulatory information from the websites of security regulatory agencies in each country and updates the database. Using APIs and web scraping technology, the server collects the latest information and stores it in MongoDB. The input data is regulatory agency website information, and the output data is the updated database information.

[0800] Step 7:

[0801] Users use a smartphone or head-mounted display (HMD) to search for the data they need and view summaries and reports. The user interface is built using a front-end framework such as React to allow users to easily manipulate the data. The input data are search keywords from the user, and the output data are search results and summary reports.

[0802] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0803] The present invention provides a system that streamlines data collection, formatting, summarization, visualization, comparison, and updating of regulatory information in the pharmaceutical application process, and also includes an emotion engine that recognizes user emotions and adjusts interface display and data search results based on the emotions. The system has the following components and functions:

[0804] System Overview

[0805] The system consists of a server, a terminal, and a user. The server is mainly responsible for data collection, automatic formatting, summary generation, data visualization, data comparison, regulatory information updates, and user emotion recognition. The terminal functions as the interface with the user, allowing the user to search and view data through the system and create documents required for drug applications.

[0806] Program processing

[0807] 1. Data Collection

[0808] Subject: Server

[0809] The server uses pre-configured API keys and web scraping techniques to collect data from scientific papers, clinical trial databases, and drug regulatory agency websites. Specifically, the server uses the PubMed API to retrieve paper titles and abstracts related to "cancer treatment" and stores them in the database in JSON format. It also uses web scraping technology to crawl the FDA's latest guidelines page, retrieves and stores the content in text format.

[0810] 2. Data Formatting

[0811] Subject: Server

[0812] The server analyzes the collected data and unifies different data formats. Specifically, it parses JSON-formatted data, removes duplicate data, and fills in missing values ​​with appropriate default values. For example, in the case of paper data, if multiple papers with the same title are obtained, they are merged into one, and if information such as author names or publication dates is missing, it is filled in from reliable external information.

[0813] 3. Data Summary

[0814] Subject: Server

[0815] The server uses natural language processing (NLP) technology to extract important information from the collected data and generate summaries. Specifically, it uses an NLP library (e.g., NLTK or SpaCy) to analyze the main keywords in the paper, as well as descriptions of therapeutic effects and side effects, and generates summaries based on this. The generated summaries are then stored in a database.

[0816] 4. Data Visualization

[0817] Subject: Server

[0818] The server uses tools to generate graphs and charts based on the summarized data for visualization. For example, Matplotlib is used to generate line graphs showing the effectiveness of each treatment and bar graphs showing the frequency of side effects. The generated graphs and charts are converted into a format that can be viewed by the user interface and displayed on a dashboard in real time.

[0819] 5. Data Comparison

[0820] Subject: Server

[0821] The server compares the newly acquired data with existing data and analyzes differences and commonalities. Specifically, it cross-references data on existing cancer treatments with data on new treatments, comparing them in terms of treatment effectiveness, side effects, etc. The comparison results are generated in report format, making them easy for users to understand.

[0822] 6. Regulatory Updates

[0823] Subject: Server

[0824] The server periodically retrieves the latest information from the websites of drug regulatory agencies in each country and updates the database. Specifically, it uses a web scraping tool to retrieve the latest guidelines and regulatory changes from the FDA and EMA, stores them in the database, and sends notifications to users when changes are confirmed.

[0825] 7. Data Search and Browsing

[0826] Subject: User

[0827] Users use the device's web interface to search for and retrieve the data they need. Specifically, they enter keywords such as "new treatments for lung cancer" into the search screen and click the search button to send a request to the server. The server then retrieves relevant summaries and comparison reports from the database and displays them on the user's device.

[0828] 8. Emotion Recognition and Interface Adjustment

[0829] Subject: Server

[0830] The server uses an emotion engine to recognize the user's emotions, for example, by analyzing the user's facial expressions and tone of voice, and then adjusts the priority of data search results and optimizes the interface display.

[0831] Examples:

[0832] If the server detects that a user is feeling stressed while completing a regulatory application, it will simplify the interface and adjust the display to make important information easier to access. If there are too many search results, it will adjust the ranking to prioritize the most relevant data.

[0833] 9. Preparation of pharmaceutical application documents

[0834] Subject: User

[0835] Users create regulatory application documents based on the displayed summaries and comparison reports. Specifically, they download templates provided by the system and create documents by entering the displayed information. They can also submit the created documents for review by experts, who can make revisions based on the feedback.

[0836] In this way, the present invention, equipped with an emotion engine, is a system that not only efficiently collects, formats, summarizes, visualizes, and compares the various data required for the pharmaceutical application process, and updates the latest regulatory information, but also further supports the application process and improves its accuracy and speed by optimizing the interface in consideration of the user's emotions.

[0837] The processing flow will be explained below.

[0838] Step 1: Data collection

[0839] Subject: Server

[0840] The server uses pre-configured API keys and web scraping techniques to collect data from scientific papers, clinical trial databases, and drug regulatory agency websites. Specifically, the server uses the PubMed API to retrieve paper titles and abstracts related to "cancer treatment." The retrieved data is stored in a database in JSON format. It also uses web scraping technology to crawl the FDA's latest guidelines page, retrieve their contents in text format, and store them.

[0841] Step 2: Data Shaping

[0842] Subject: Server

[0843] The server analyzes the collected data and unifies different data formats. Specifically, it parses JSON-formatted data, removes duplicate data, and fills in missing values ​​with appropriate default values. For example, if multiple papers with the same title are retrieved, they are merged into one, and if missing information such as author names or publication dates is provided, it is filled in using reliable external information.

[0844] Step 3: Data Summarization

[0845] Subject: Server

[0846] The server uses natural language processing (NLP) technology to extract important information from the collected data and generate summaries. Specifically, it uses NLP libraries (e.g., NLTK and SpaCy) to analyze each paper's main keywords, therapeutic effects, and side effects. Based on the extracted information, it automatically generates summaries and stores them in a database.

[0847] Step 4: Data visualization

[0848] Subject: Server

[0849] The server uses tools to generate graphs and charts based on the summarized data for visualization. Specifically, Matplotlib is used to generate line graphs showing the effectiveness of each treatment and bar graphs showing the frequency of side effects. The generated graphs and charts are converted into a format that can be viewed by the user interface and displayed on a dashboard in real time.

[0850] Step 5: Data comparison

[0851] Subject: Server

[0852] The server compares the newly acquired data with existing data and analyzes differences and commonalities. Specifically, it cross-references data on existing cancer treatments with data on new treatments to evaluate differences in treatment effectiveness and side effects. The comparison results are generated in a detailed report format that users can easily understand.

[0853] Step 6: Update regulatory information

[0854] Subject: Server

[0855] The server periodically retrieves the latest information from the websites of each country's drug regulatory agency and updates the database. Specifically, it uses a web scraping tool to retrieve the latest guidelines and regulatory changes from the FDA and EMA, and stores them in the database. If any changes are confirmed, the server notifies the user.

[0856] Step 7: Emotion Recognition

[0857] Subject: Server

[0858] The server uses an emotion engine to recognize the user's emotions. Specifically, it analyzes the user's facial expressions, voice tone, input speed, etc. to evaluate the user's emotions (e.g., stress, satisfaction). The results of emotion recognition are used to adjust the interface and prioritize search results.

[0859] Step 8: Interface Adjustment

[0860] Subject: Server

[0861] The server adjusts the interface display content based on the user's emotions recognized by the emotion engine. Specifically, if the user is feeling stressed, the interface will be simplified and important information will be highlighted. The server will also prioritize the display of content that matches the user's interests.

[0862] Step 9: Search and view data

[0863] Subject: User

[0864] Users use the device's web interface to search for and retrieve the data they need. Specifically, they enter keywords such as "new treatments for lung cancer" into the search screen and click the search button to send a request to the server. The server then retrieves relevant summaries and comparison reports from the database and displays them on the user's device.

[0865] Step 10: Prepare regulatory application documents

[0866] Subject: User

[0867] Users create regulatory application documents based on the displayed summaries and comparison reports. Specifically, they download templates provided by the system and create documents by entering the displayed information. They then submit the created documents for review by experts, who provide feedback and make revisions.

[0868] In this way, the present invention, equipped with an emotion engine, is a system that not only efficiently collects, formats, summarizes, visualizes, and compares the various data required for the pharmaceutical application process, and updates the latest regulatory information, but also further supports the application process and improves its accuracy and speed by optimizing the interface in consideration of the user's emotions.

[0869] Example 2

[0870] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0871] The pharmaceutical application process requires a huge amount of data collection, complex data formatting, detailed summarization, effective data visualization, rigorous data comparison, and the latest regulatory information updates, making it extremely time-consuming and labor-intensive. Furthermore, the lack of user-friendly interface adjustments can lead to reduced work efficiency. Therefore, there is a need for a system that can efficiently complete pharmaceutical application procedures without causing stress to users.

[0872] The specific processing by the specific processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for automatically collecting clinical data and non-clinical data, means for formatting the collected data and converting it into a unified format, means for extracting important information from the formatted data and generating a summary, means for visualizing the summary data in the form of graphs or charts, means for comparing new data with existing data and providing the results in the form of a report, means for periodically obtaining the latest pharmaceutical regulatory information and updating the database, means for a user to use an interface to search for necessary data and view summaries and reports, means for analyzing user emotions using an emotion engine and adjusting the interface display and search results, and means for generating prompt sentences to input to the generative AI model. This enables efficient data collection, data formatting, summarization, visualization, comparison, and updating of the latest regulatory information in the pharmaceutical application process, and further enables significant improvement in work efficiency and user experience through interface adjustment based on user emotions.

[0873] "Data Collection Instruments" refers to processes or mechanisms for the automated collection of clinical and non-clinical data.

[0874] "Data formatting means" refers to the process or method of analyzing collected data and converting it into a unified format.

[0875] "Data summarization means" refers to a mechanism or process that extracts important information from formatted data and generates a summary.

[0876] "Data visualization methods" refers to the process of visualizing summarized data in the form of graphs or charts.

[0877] "Data comparison means" refers to a method or technology that compares new data with existing data and provides the results of the comparison in a report format.

[0878] "Regulatory information update means" refers to the process or method of periodically obtaining the latest pharmaceutical regulatory information and updating the database.

[0879] "Data retrieval means" refers to the ability for users to use the interface to retrieve required data and view summaries and reports.

[0880] "Emotion analysis means" refers to a mechanism that uses an emotion engine to analyze user emotions and adjusts interface display and search results based on the results.

[0881] "Prompt generation means" refers to the process or method for generating prompt sentences to be input into a generative AI model.

[0882] "Automatic means" refers to processes or techniques that do not require manual intervention and are operated automatically by the system.

[0883] "User interface" refers to the screens and methods by which a user interacts with a system and searches and views data.

[0884] The present invention is a system that streamlines data collection, formatting, summarization, visualization, comparison, and updating of regulatory information in the pharmaceutical application process. Furthermore, it aims to improve operability by incorporating an emotion engine that recognizes user emotions and adjusts interface display and data search results based on the emotions. The system of the present invention has the following specific configuration:

[0885] System Components

[0886] 1. Server

[0887] The server is the central part of the system and performs the following functions:

[0888] Data collection: Using PubMed API and web scraping techniques, we collect data from scientific articles, clinical trial databases, and drug regulatory agency websites. For example, we use PubMed API to retrieve titles, authors, and abstracts of articles related to "cancer treatment" and store them in a database in JSON format.

[0889] Data Transformation: Analyze the collected data and convert it into a unified format, specifically removing duplicate entries in the JSON data and imputing missing values ​​from external reliable data.

[0890] Data summarization: Using natural language processing (NLP) techniques, we extract important information from the collected data and create summaries. For example, we use NLP libraries (e.g., NLTK and SpaCy) to analyze the main keywords in papers and descriptions of treatment effects and side effects.

[0891] Data visualization: Visualize the data using tools that generate graphs and charts (e.g., Matplotlib) based on the summarized data. The generated graphs and charts are displayed in real time on a dashboard.

[0892] Data comparison: Compare newly acquired data with existing data and analyze differences and commonalities. Generate a report of the comparison results so that users can easily understand them.

[0893] Regulatory information updates: Regularly retrieve the latest information from the websites of national drug regulatory agencies and update the database. For example, use web scraping tools to retrieve the latest guidelines from the FDA and EMA.

[0894] Emotion recognition: An emotion engine is used to analyze emotions from a user's facial expressions and tone of voice to adjust interface display and data search result priorities.

[0895] 2. Terminal

[0896] The terminal acts as an interface with the user:

[0897] Data search and browsing: Users use a web interface to search for and retrieve the data they need. For example, they enter keywords such as "new treatments for lung cancer" into the search screen and click the search button to send a request to the server. The server retrieves relevant summaries and comparison reports from the database and displays them on the terminal.

[0898] Interface Adaptation: If the emotion engine detects stress in the user, it will automatically simplify the interface and adjust to make important information easier to access.

[0899] 3. Users

[0900] Users use the system to:

[0901] Data retrieval: Users can search for the required data through a web interface and view summaries and reports.

[0902] Creating regulatory application documents: Users create regulatory application documents based on the displayed summary and comparison reports. They download templates provided by the system and enter the required information. They submit the completed documents for expert review and make revisions based on the feedback.

[0903] Specific examples

[0904] 1. The server uses the PubMed API to search for papers on "lung cancer treatments" and retrieves their titles, authors, and abstracts. Then, it uses a web crawler to parse the FDA's "New Treatment Guidelines" page and saves the content in text format.

[0905] 2. The server eliminates duplicates from the collected data and completes missing author names, publication dates, etc. from external databases.

[0906] 3. Using natural language processing technology, generate a summary from the collected papers: "This new drug improved the survival rate of lung cancer patients by 20%, but nausea was reported as a side effect."

[0907] 4. Use Matplotlib to generate a line graph showing the effectiveness of "Lung Cancer Treatment A" and a bar graph showing the frequency of side effects. Display these graphs on a dashboard.

[0908] 5. Compare the effectiveness and side effects of "existing treatment A" and "new treatment B" and generate a comparative report showing that new treatment B is slightly less effective but has fewer side effects.

[0909] 6. Use a web crawler to regularly check for the latest EMA guidelines and store new regulatory information in a database.

[0910] 7. If users are experiencing stress while completing regulatory submissions, simplify the interface, make important information more accessible, and adjust the ranking of search results to surface more relevant information.

[0911] Example prompts to input to the generative AI model

[0912] "Collect the latest research data on lung cancer treatments, summarize and compare them, and use them to draft regulatory applications."

[0913] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0914] Step 1:

[0915] Data collection

[0916] The server collects data using the PubMed API and a web scraping tool. Specifically, it retrieves the titles, authors, and abstracts of papers related to "lung cancer treatment" from the PubMed API. It also uses the web scraping tool to crawl the FDA's latest guidelines page and saves the content in text format.

[0917] Input: Keyword "lung cancer treatment", API key, website URL

[0918] Output: Paper data (JSON format), guideline text data

[0919] Specific operation: The server collects paper data on "lung cancer treatment" from the PubMed API, and at the same time uses a web crawler to retrieve and save the contents of the FDA's latest guidelines page.

[0920] Step 2:

[0921] Data Formatting

[0922] The server parses the collected JSON-formatted data and unifies different formats, specifically removing duplicate data and completing missing values ​​from external information.

[0923] Input: Collected paper data (JSON format), guideline text data

[0924] Output: Formatted data (Uniform JSON format)

[0925] What it does: The server reads a file such as "Lung Cancer Treatments.pdf," merges duplicate entries, and completes missing author names and publication dates.

[0926] Step 3:

[0927] Data Summary

[0928] The server uses natural language processing (NLP) technology to analyze the collected data and extract important information. It uses NLP libraries (NLTK and SpaCy) to analyze the main keywords of the papers, as well as descriptions of therapeutic effects and side effects, and generates summaries.

[0929] Input: Formatted data (Uniform JSON format)

[0930] Output: Summary data (text format)

[0931] Specific operation: The server analyzes the abstract of a paper on "lung cancer treatment" using SpaCy, generates a summary, and saves it.

[0932] Step 4:

[0933] Data Visualization

[0934] The server generates graphs and charts based on the summarized data, using Matplotlib to graphically display the effectiveness of each treatment and the frequency of side effects.

[0935] Input: Summary data (text format)

[0936] Output: Graphs and charts (image format)

[0937] Specific operation: Based on the summarized data, the server uses Matplotlib to create a line graph showing the effectiveness of "Lung Cancer Treatment A" and a bar graph showing the frequency of side effects.

[0938] Step 5:

[0939] Data comparison

[0940] The server cross-references the newly acquired data with the existing data, analyzes the differences and commonalities between them, and generates a report of the comparison results.

[0941] Input: Summary data (text format), existing data

[0942] Output: Comparison report (text format)

[0943] Specific operation: The server cross-references the effectiveness and side effects of "existing treatment A" and "new treatment B" and generates a report stating that treatment B is slightly less effective but has fewer side effects.

[0944] Step 6:

[0945] Regulatory updates

[0946] The server periodically scrapes the websites of each country's drug regulatory agency to collect the latest information and update the database.

[0947] Input: Regulatory agency website URL

[0948] Output: Latest regulatory information (text format)

[0949] What it does: The server uses a web crawler to crawl the latest FDA and EMA guideline pages and stores the necessary information in a database.

[0950] Step 7:

[0951] Data Search and Viewing

[0952] Users use the device's web interface to search for and retrieve the data they need. By entering search keywords and clicking the search button, the request is sent to the server, which retrieves the relevant data and displays it on the user's device.

[0953] Input: Search keyword

[0954] Output: Relevant data (summary and comparison reports)

[0955] Specific operation: When a user enters "new treatments for lung cancer" into the search box and performs a search, the server displays relevant summaries and reports on the device.

[0956] Step 8:

[0957] Emotion Recognition and Interface Adjustment

[0958] The server uses an emotion engine to analyze the user's emotions, including facial expressions and tone of voice, and adjusts the interface display and search result priorities based on the results.

[0959] Input: User's facial expression data, voice data

[0960] Output: Tailored interface, prioritized search results

[0961] What it does: The server collects data from the user's webcam and microphone and simplifies the interface if the user is stressed.

[0962] Step 9:

[0963] Preparation of pharmaceutical application documents

[0964] Users create regulatory application documents based on the displayed summary and comparison reports. They download templates provided by the system, enter the required information, and submit the completed documents for expert review, where they can make revisions based on the feedback.

[0965] Inputs: Summary data, comparison reports, templates

[0966] Output: Pharmaceutical application documents

[0967] What it does: A user downloads a template from the system, copies information from the summary and comparison reports to complete the application, and then submits it to an expert for feedback and makes any necessary revisions.

[0968] (Application example 2)

[0969] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0970] Conventional systems do not efficiently collect, organize, summarize, visualize, and compare clinical and non-clinical data, or update and search regulatory information. Furthermore, they do not allow for interface adjustments that take into account users' emotions and stress levels, or optimization of productivity and work environments. This makes it difficult for managers to grasp the status of workers in real time and provide an optimal work environment.

[0971] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0972] In this invention, the server includes a means for automatically collecting clinical and non-clinical data, a means for formatting the data and converting it into a unified format, and a means for extracting important information and generating summaries. This allows for efficient processing of various types of data. It also includes a means for collecting biometric information from workers, analyzing their emotions and stress levels, a means for adjusting the interface display content based on their emotions and stress levels, and a means for visualizing production data and optimizing production efficiency and the work environment. This allows managers to understand the status of workers in real time and provide a comfortable and efficient work environment.

[0973] "Clinical Data" refers to patient medical information collected in a medical setting, including data related to diagnosis, treatment, and follow-up.

[0974] "Non-clinical data" refers to research data obtained outside of clinical trials, including data obtained from animal experiments, cell experiments, etc.

[0975] "Data transformation" refers to the process of analyzing collected data and unifying different formats and forms.

[0976] "Unified format" refers to a format that converts data obtained from different sources into a consistent format.

[0977] "Important information extraction" refers to the process of selecting the necessary data from a large amount of data according to a specific purpose or requirement.

[0978] A "summary" is a document or text that succinctly summarizes the main points or essence of the collected data.

[0979] "Worker biometric information" refers to data about a worker's physical condition, such as heart rate, body temperature, and stress level.

[0980] Analyzing "emotions and stress levels" refers to the process of assessing a worker's emotional state and stress level from collected biometric information.

[0981] "Adjusting the interface display content" refers to the process of changing the system display content and operation method according to the user's condition to provide the optimal user experience.

[0982] "Production data" refers to data related to the manufacturing process of products in factories and production lines, including operating hours, failure rates, production efficiency, etc.

[0983] "Production efficiency" refers to an indicator that shows how much product can be produced using a certain amount of resources (time, manpower, materials).

[0984] "Working environment" refers to the conditions in the place where workers perform their work, including temperature, humidity, noise level, etc.

[0985] The present invention provides a system for achieving efficient data management and optimizing the working environment at a production site. The system includes the following hardware and software.

[0986] Hardware

[0987] IoT devices and sensors: installed to collect temperature, humidity, and vibration data.

[0988] Wearable devices: Worn to collect workers' biometric information (heart rate, stress levels, etc.).

[0989] Smart glasses: worn by workers, they display the information they need in real time while they work.

[0990] software

[0991] Server: Responsible for data collection, formatting, analysis, visualization, comparison, and regulatory information updates.

[0992] Data Collection API: Collects data from sensors and wearable devices using the MQTT protocol or Bluetooth communication.

[0993] Parsing libraries: Use natural language processing libraries such as NLTK or SpaCy to generate summaries of your data.

[0994] Visualization tools: Use Matplotlib and Plotly to visualize data in the form of graphs and charts.

[0995] Machine learning model: Using TensorFlow to analyze worker emotions and stress levels.

[0996] System Details

[0997] 1. Data Collection:

[0998] The server collects data in real time from IoT devices and sensors, capturing temperature, humidity, and vibration data using the MQTT protocol and storing them in a database. It also collects biometric data from workers' smart glasses and wearable devices via Bluetooth and transmits it to the server.

[0999] 2. Data Formatting:

[1000] The server parses the collected data and converts it into a standard format, cleaning it using Python's Pandas library, removing duplicates, and imputing missing values.

[1001] 3. Data Summary:

[1002] Use natural language processing libraries (e.g., NLTK, SpaCy) to extract important information from the collected data and generate summaries. For example, analyze temperature fluctuations obtained from multiple sensor data and generate a summary.

[1003] 4. Data visualization:

[1004] The server generates graphs and charts in real time based on the collected data, using Matplotlib and Plotly to visually display the operation status of the production line and the stress levels of workers.

[1005] 5. Data comparison:

[1006] The server compares and analyzes newly acquired data with existing data to identify anomalies and patterns, particularly by comparing machine operating hours and failure rates with past data.

[1007] 6. Regulatory Updates:

[1008] Regularly obtain information on each country's occupational safety regulations and environmental standards using a web scraping tool (e.g., BeautifulSoup) and update the database. If there are any changes in regulations, notify the administrator.

[1009] 7. Emotion Recognition and Interface Adjustment:

[1010] The server analyzes the worker's emotions and stress levels from facial expressions and biometric data. It uses TensorFlow to classify emotions and adjusts the interface display if it recognizes a high level of stress. For example, if a worker's stress level is high, it changes the work priority and displays an alert suggesting a break.

[1011] 8. Optimizing production efficiency and working environment:

[1012] The server analyzes the collected production data and optimizes production efficiency and the working environment, thereby improving productivity in the factory and providing a comfortable working environment for workers.

[1013] Specific examples

[1014] For example, when Worker A is wearing smart glasses while working, the glasses measure his heart rate and stress level using a data collection API sent from the server. Using TensorFlow, if the data determines that Worker A's stress level is high, the server changes the work priority and immediately displays an alert recommending a break. Also, if the factory temperature or humidity data exceeds the standard, the server simultaneously sends an alert and issues instructions for improving the environment.

[1015] Example prompts for generative AI models

[1016] Based on Worker A's latest biometric data, generate a response plan for when stress levels are high.

[1017] Working environment data: Temperature: 28°C, Humidity: 60%, Vibration: Low

[1018] Vital Data: Heart rate: 80 bpm, Stress level: High

[1019] In this way, the system can maintain production efficiency while ensuring worker health and comfort.

[1020] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1021] Step 1:

[1022] The server collects environmental data (temperature, humidity, vibration) from IoT devices and sensors.

[1023] Input: Data sent from each sensor via MQTT protocol.

[1024] Processing: The server parses the received data and stores it in the database.

[1025] Output: Formatted environmental data (Temperature: 25°C, Humidity: 60%, Vibration: Low).

[1026] Step 2:

[1027] The server collects biometric information (heart rate, stress level) from smart glasses or wearable devices worn by workers.

[1028] Input: Real-time biometric data transmitted via Bluetooth.

[1029] Processing: The server receives the biometric data and stores it in a database.

[1030] Output: Formatted biometric data (heart rate: 80 bpm, stress level: high).

[1031] Step 3:

[1032] The server analyzes the collected environmental and biological data and converts it into a unified format.

[1033] Input: Environmental and biological data stored in a database.

[1034] Processing: Using Python's Pandas library, the data is cleaned, missing values ​​are imputed, and it is converted into a standard format.

[1035] Output: A uniformly formatted dataset.

[1036] Step 4:

[1037] The server extracts important information from the dataset and generates a summary.

[1038] Input: A uniformly formatted dataset.

[1039] Processing: Use NLTK or SpaCy to analyze the data and generate summaries.

[1040] Output: Summarized data.

[1041] Step 5:

[1042] The server visualizes the summarized data and displays it on the factory dashboard.

[1043] Input: Summarized data.

[1044] Processing: Visualize the data in the form of graphs and charts using Matplotlib and Plotly.

[1045] Output: Graphs and charts displayed on a dashboard.

[1046] Step 6:

[1047] The server compares the newly acquired data with past data to detect any anomalies.

[1048] Input: Newly acquired data and existing datasets.

[1049] Processing: Use Pandas and SciPy to compare data and identify anomalies and patterns.

[1050] Output: Anomaly detection results report.

[1051] Step 7:

[1052] The server regularly obtains each country's occupational safety regulations and environmental standards using a web scraping tool and updates the database.

[1053] Input: Websites that provide regulatory information.

[1054] Processing: Use BeautifulSoup to perform web scraping and store new regulatory information in a database.

[1055] Output: Updated regulatory information.

[1056] Step 8:

[1057] The server analyzes the worker's biometric data and adjusts the interface display based on their emotions and stress levels.

[1058] Input: Worker's biometric data.

[1059] Processing: TensorFlow is used to analyze biometric data and classify emotions and stress levels, while simultaneously adjusting the interface display.

[1060] Output: The adjusted interface display.

[1061] Step 9:

[1062] The server analyzes production data and optimizes production efficiency and the working environment.

[1063] Input: Production data and work environment data.

[1064] Processing: Using data analysis tools, optimization algorithms are applied to evaluate production efficiency.

[1065] Output: Improvement suggestions and alert messages.

[1066] For example, if Worker A's biometric data is analyzed and it is determined that his stress level is high, an alert will be displayed immediately recommending that he take a break. Also, if the temperature or humidity in the factory exceeds the standard, instructions to improve the environment will be automatically sent.

[1067] Example prompts for generative AI models

[1068] Based on Worker A's latest biometric data, generate a response plan for when stress levels are high.

[1069] Working environment data: Temperature: 28°C, Humidity: 60%, Vibration: Low

[1070] Vital Data: Heart rate: 80 bpm, Stress level: High

[1071] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1072] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs 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.

[1073] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[1074] [Third embodiment]

[1075] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[1076] 5, the data processing system 310 includes the data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[1077] 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).

[1078] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. 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 microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[1079] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1080] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1081] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1082] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1083] The specific processing program 56 is an example of a "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 processing 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.

[1084] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1085] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1086] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[1087] The present invention is a system for streamlining data collection, formatting, summarization, visualization, comparison, and updating of regulatory information in the pharmaceutical application process. This system has the following components and functions.

[1088] System Overview

[1089] The system consists of a server, terminals, and users. The server is primarily responsible for data collection, automatic formatting, summary generation, data visualization, data comparison, and regulatory information updates, while the terminals function as an interface with users. Users use the system to search for and view the necessary data and create the documents required for pharmaceutical applications.

[1090] Program processing

[1091] 1. Data Collection

[1092] Subject: Server

[1093] The server automatically collects data based on pre-defined keywords from publicly available scientific papers, clinical trial databases, drug regulatory agency websites, etc. Specifically, it uses various APIs and web scraping technologies to retrieve the required data and store it in a database.

[1094] Examples:

[1095] The server retrieves the titles, authors, abstracts, and full texts of relevant articles from PubMed using the keyword "cancer treatment" and stores them in a database.

[1096] 2. Data Formatting

[1097] Subject: Server

[1098] The server reviews the collected data and converts it into a unified format, removing duplicates and imputing missing values ​​to ensure data integrity.

[1099] Examples:

[1100] The server analyzes multiple article data retrieved from PubMed, removes duplicate entries, and adds reliable default values ​​for missing entries.

[1101] 3. Data Summary

[1102] Subject: Server

[1103] The server uses natural language processing technology to extract important information from the data and generate a summary of key points (e.g., treatment effects, side effects, etc.).

[1104] Examples:

[1105] The server extracts information about each paper's key keywords, therapeutic effects, and side effects, and provides it as a summary.

[1106] 4. Data Visualization

[1107] Subject: Server

[1108] The server uses tools to generate graphs and charts based on the summarized data to visually display the data.

[1109] Examples:

[1110] The server visualizes the treatment efficacy data as a line graph and displays the frequency of side effects as a bar graph.

[1111] 5. Data Comparison

[1112] Subject: Server

[1113] The server compares the newly acquired data with the existing data, analyzes similarities and differences, and provides the comparison results in the form of a report.

[1114] Examples:

[1115] The server compares data on existing cancer treatments with data on newly acquired treatments and generates a detailed report on differences in treatment effectiveness and side effects.

[1116] 6. Regulatory Updates

[1117] Subject: Server

[1118] The server periodically retrieves the latest regulatory information from the websites of each country's drug regulatory agency and updates the database.

[1119] Examples:

[1120] The server uses web scraping technology to obtain the latest US FDA guidelines and imports them into the database for updating.

[1121] 7. User Interface

[1122] Subject: User

[1123] Users use a web interface to search for the data they need, view summaries and reports, and generate regulatory submissions based on the information provided.

[1124] Examples:

[1125] The user searches for data on "new treatments for lung cancer" in the interface, checks the obtained summary and comparison reports, and downloads a template for pharmaceutical application documents based on the information and fills in the required fields to complete the application.

[1126] In this way, the present invention is a system that automates application procedures and improves accuracy and speed by efficiently collecting, formatting, summarizing, visualizing, and comparing the various data required for the pharmaceutical application process, as well as updating the latest regulatory information.

[1127] The processing flow will be explained below.

[1128] Step 1: Data collection

[1129] Subject: Server

[1130] The server uses pre-configured API keys and web scraping techniques to collect data from scientific papers, clinical trial databases, and drug regulatory agency websites. Specifically, the server uses the PubMed API to retrieve paper titles and abstracts related to "cancer treatment" and stores them in a database in JSON format. It also uses web scraping technology to crawl the FDA's latest guidelines page, retrieves their contents in text format, and stores them.

[1131] Step 2: Data Shaping

[1132] Subject: Server

[1133] The server analyzes the collected data and unifies different data formats. Specifically, it parses JSON-formatted data, removes duplicate data, and fills in missing values ​​with appropriate default values. For example, in the case of paper data, if multiple papers with the same title are obtained, they are merged into one, and if information such as author names or publication dates is missing, it is filled in from reliable external information.

[1134] Step 3: Data Summarization

[1135] Subject: Server

[1136] The server uses natural language processing (NLP) technology to extract important information from the collected data and generate summaries. Specifically, it uses an NLP library (e.g., NLTK or SpaCy) to analyze the main keywords in the paper, as well as descriptions of therapeutic effects and side effects, and generates summaries based on this. The generated summaries are then stored in a database.

[1137] Step 4: Data visualization

[1138] Subject: Server

[1139] The server uses tools to generate graphs and charts based on the summarized data for visualization. For example, Matplotlib is used to generate line graphs showing the effectiveness of each treatment and bar graphs showing the frequency of side effects. The generated graphs and charts are converted into a format that can be viewed by the user interface and displayed on a dashboard in real time.

[1140] Step 5: Data comparison

[1141] Subject: Server

[1142] The server compares the newly acquired data with existing data and analyzes differences and commonalities. Specifically, it cross-references data on existing cancer treatments with data on new treatments, comparing them in terms of treatment effectiveness, side effects, etc. The comparison results are generated in report format, making them easy for users to understand.

[1143] Step 6: Update regulatory information

[1144] Subject: Server

[1145] The server periodically retrieves the latest information from the websites of each country's drug regulatory agency and updates the database. Specifically, it uses a web scraping tool to retrieve the latest guidelines and regulatory changes from the FAA and EMA, stores them in the database, and sends notifications to users when changes are confirmed.

[1146] Step 7: Search and view data

[1147] Subject: User

[1148] Users use the device's web interface to search for and retrieve the data they need. Specifically, they enter keywords such as "new treatments for lung cancer" into the search screen and click the search button to send a request to the server. The server then retrieves relevant summaries and comparison reports from the database and displays them on the user's device.

[1149] Step 8: Prepare regulatory submission documents

[1150] Subject: User

[1151] Users create regulatory application documents based on the displayed summaries and comparison reports. Specifically, they download templates provided by the system and create documents by entering the displayed information. They can also submit the created documents for review by experts, who can make revisions based on the feedback.

[1152] Example 1

[1153] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1154] The current pharmaceutical application process involves a wide range of tasks, including collecting, organizing, summarizing, visualizing, comparing, and updating regulatory information, and is extremely time-consuming and labor-intensive. Furthermore, the quality and consistency of the collected data varies, and the accuracy and efficiency of application document preparation is not fully guaranteed. This leads to delays in the application process and the submission of incorrect information.

[1155] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1156] In this invention, the server includes, in data collection, means for automatically obtaining data from publicly available information sources based on preset keywords, means for reviewing the collected data and checking for format inconsistencies and missing values, means for formatting the collected data, converting it into a unified format, deleting duplicate data, and imputing missing values, means for extracting important information from the collected data using natural language processing technology and generating a summary, means for visualizing the summary data in the form of a graph or chart, means for comparing newly obtained data with existing data, analyzing similarities and differences, and providing the results in the form of a report, means for periodically obtaining the latest regulatory information and updating the database, means for a user to use an interface to search for necessary data and view summaries and reports, and means for providing application document templates based on the search results and for the user to enter the necessary information to create them. This makes it possible to improve the efficiency and accuracy of the pharmaceutical application process.

[1157] "Data harvesting" refers to the automatic retrieval of data from publicly available sources based on pre-defined keywords.

[1158] "Format inconsistency" refers to a situation where collected data exists in different structures or formats.

[1159] "Missing values" refers to a state in which some data is missing or necessary information is not entered.

[1160] "Duplicate data" refers to a situation in which multiple pieces of data with the same content exist.

[1161] "Missing value imputation" refers to filling in missing data based on predictions or reliable sources.

[1162] "Natural language processing" refers to the technology for understanding, analyzing, and generating human language.

[1163] "Extracting important information" refers to extracting only specific, necessary information from collected data.

[1164] A "summary" is a short summary of the main points from a long piece of text or a large amount of data.

[1165] "Visualization" refers to the display of data in a visual format such as a graph or chart.

[1166] "Report format" refers to a format in which the results of analysis and comparison are systematically summarized in documents and charts.

[1167] "Regulatory information" refers to information on norms and standards provided by public institutions, such as laws, regulations, and guidelines.

[1168] "Database updating" refers to adding new information to existing data and keeping it up to date.

[1169] "User interface" refers to the screens and controls that allow a user to interact with a computer system or application.

[1170] "Application template" refers to a template for application documents in a predefined format.

[1171] This invention is a system that streamlines data collection, formatting, summarization, visualization, comparison, and updating of regulatory information in the pharmaceutical application process. This system is mainly composed of a server, terminals, and users. The specific system structure and operation are described in detail below.

[1172] System Configuration

[1173] server:

[1174] The server is responsible for data collection, automatic formatting, summary generation, data visualization, data comparison, and regulatory information updates. The server uses the following hardware and software:

[1175] Hardware: High-performance server (e.g., Xeon-based server)

[1176] Software: Python, natural language processing libraries (e.g., NLTK, SpaCy), data visualization tools (e.g., Matplotlib, D3.js)

[1177] Device:

[1178] The terminal provides the interface through which the user interacts with the system. The terminal uses the following hardware and software:

[1179] Hardware: Personal computers, tablets

[1180] Software: Web browser (e.g. Google Chrome, Firefox)

[1181] User:

[1182] Users are drug developers and regulatory officials involved in the regulatory application process, who use the system to search for the data they need, view summaries and reports, and prepare application documents.

[1183] How it works

[1184] 1. Data Collection:

[1185] The server collects data based on pre-defined keywords from published scientific papers, clinical trial databases, drug regulatory agency websites, etc., using various API calls and web scraping techniques.

[1186] Examples:

[1187] The server retrieves the titles, authors, abstracts, and full texts of relevant articles from PubMed using the keyword "cancer treatment" and stores them in a database.

[1188] 2. Data Formatting:

[1189] The server reviews the collected data and converts it into a unified format, removing duplicates and imputing missing values ​​to ensure data integrity.

[1190] Examples:

[1191] The server analyzes multiple article data retrieved from PubMed, removes duplicate entries, and adds reliable default values ​​for missing entries.

[1192] 3. Data Summary:

[1193] The server uses natural language processing techniques to extract key information from the data and generate a summary.

[1194] Examples:

[1195] The server extracts information about each paper's key keywords, therapeutic effects, and side effects, and provides it as a summary.

[1196] 4. Data visualization:

[1197] The server uses tools to generate graphs and charts based on the summarized data to visually display the data.

[1198] Examples:

[1199] The server visualizes the treatment efficacy data as a line graph and displays the frequency of side effects as a bar graph.

[1200] 5. Data comparison:

[1201] The server compares the newly acquired data with the existing data, analyzes similarities and differences, and provides the results in a report.

[1202] Examples:

[1203] The server compares data on existing cancer treatments with data on newly acquired treatments and generates a detailed report on differences in treatment effectiveness and side effects.

[1204] 6. Regulatory Updates:

[1205] The server periodically retrieves the latest regulatory information from the websites of each country's drug regulatory agency and updates the database.

[1206] Examples:

[1207] The server uses web scraping technology to obtain the latest US FDA guidelines and imports them into the database for updating.

[1208] 7. User Interface:

[1209] Users use a web interface to search for the data they need, view summaries and reports, and download application templates based on their search results, then fill in the required fields to complete the application.

[1210] Examples:

[1211] Users can search for data on "new treatments for lung cancer" in the interface, check the resulting summary and comparison reports, and download a template for a pharmaceutical application based on the information, then fill in the required fields to create it.

[1212] Prompt Sentence Examples

[1213] "Collect the latest articles on 'cancer treatment' from PubMed and create summaries that extract key keywords, therapeutic effects, and side effects."

[1214] As described above, this system comprehensively covers the entire process from data collection to preparation of application documents, dramatically improving the efficiency and accuracy of pharmaceutical applications.

[1215] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1216] Step 1: Data collection

[1217] Input: Predefined keywords (e.g., "cancer treatment").

[1218] Specific behavior:

[1219] The server automatically collects data from various data sources (e.g., scientific paper databases, clinical trial databases, regulatory agency websites, etc.) based on configured keywords, using API calls and web scraping techniques.

[1220] Data processing:

[1221] The acquired data is stored in a temporary database.

[1222] Output: The raw data collected.

[1223] Step 2: Data review

[1224] Input: The raw data collected.

[1225] Specific behavior:

[1226] The server reviews the collected data and performs basic analysis to check for formatting inconsistencies and missing values.

[1227] Data processing:

[1228] Standardize data formats and detect inconsistencies and missing values.

[1229] Output: The formatted data.

[1230] Step 3: Data Shaping

[1231] Input: Data with confirmed format.

[1232] Specific behavior:

[1233] The server converts the collected data into a unified format, removes duplicate data, and imputes missing values ​​using data cleansing algorithms.

[1234] Data processing:

[1235] Data is converted, duplicate entries are removed, and missing parts are filled in.

[1236] Output: Clean, formatted data.

[1237] Step 4: Data Summarization

[1238] Input: Clean, formatted data.

[1239] Specific behavior:

[1240] The server uses natural language processing technology to extract important information from the data and generate a summary. It identifies important items (e.g., treatment effects, side effects, etc.) and organizes them into a summary.

[1241] Data processing:

[1242] Natural language processing algorithms extract and summarize information.

[1243] Output: Summary statement.

[1244] Step 5: Data visualization

[1245] Input: Abstract text.

[1246] Specific behavior:

[1247] The server uses tools to generate graphs and charts based on the summarized data and visually display the data, using data visualization tools such as Matplotlib and D3.js.

[1248] Data processing:

[1249] Visualize data using graphs and charting tools.

[1250] Output: Graphs and charts.

[1251] Step 6: Data comparison

[1252] Input: Newly acquired and existing data.

[1253] Specific behavior:

[1254] The server compares the newly acquired data with existing data and analyzes the similarities and differences, using a comparison algorithm to identify differences and commonalities.

[1255] Data processing:

[1256] Compare data and generate analytical results.

[1257] Output: Comparison report.

[1258] Step 7: Update regulatory information

[1259] Input: Latest information obtained from national drug regulatory agency websites.

[1260] Specific behavior:

[1261] The server periodically retrieves the latest regulatory information from regulatory agencies' websites and updates the database using web scraping technology.

[1262] Data processing:

[1263] Integrate acquired regulatory information into existing databases.

[1264] Output: The updated database.

[1265] Step 8: User Interface

[1266] Input: The keyword the user wants to search for (e.g., "new treatment for lung cancer").

[1267] Specific behavior:

[1268] Users use a web interface to search for the data they need, view summaries and reports, and are provided with application templates based on the search results, which users can fill in and complete.

[1269] Data processing:

[1270] View search results and generate reports.

[1271] Output: Data search results, summary, comparison report and application template.

[1272] (Application example 1)

[1273] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1274] In today's world, the number and variety of security incidents are increasing, creating a need for rapid responses. However, the task of collecting, formatting, and summarizing the necessary information from vast amounts of data is extremely time-consuming, making it difficult to respond efficiently. It is also difficult to regularly obtain the latest information on security incidents and take appropriate measures based on that information. Therefore, an effective system that solves these issues is needed.

[1275] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1276] In this invention, the server includes means for automatically collecting clinical and non-clinical data, means for formatting the collected data and converting it into a unified format, means for extracting important information from the formatted data and generating a summary, means for visualizing the summary data in graph or chart format, means for comparing new data with existing data and providing the results in report format, means for periodically obtaining the latest pharmaceutical regulatory information and updating the database, means for a user to use an interface to search for necessary data and view summaries and reports, and means for automatically collecting, formatting, summarizing, visualizing, and comparing security incident data and periodically updating the latest security information. This enables efficient processing of massive amounts of security incident information and enables prompt and appropriate responses.

[1277] "Clinical data" refers to data collected in medical settings regarding patient diagnostic results and treatment.

[1278] "Non-clinical data" refers to data obtained outside of clinical practice, such as experimental results and observational data conducted in a laboratory or research lab.

[1279] "Collection methods" refers to the methods and techniques used to gather specific information, including the process of obtaining the required data from public databases and websites.

[1280] "Format" refers to the process of converting collected data into a unified format and, if necessary, removing duplicate data and completing missing values.

[1281] "Means for generating summaries" refers to the process of extracting important information from data and summarizing it concisely using natural language processing techniques, etc.

[1282] "Visualization means" refers to methods for displaying data in graphs or charts to make it easier to understand visually.

[1283] "Comparative means" refers to the process of matching newly acquired data with existing data and analyzing similarities and differences.

[1284] "Means of providing in report format" refers to a method of reporting comparison results and other important information in the form of documents, electronic files, etc.

[1285] "Means for regularly obtaining and updating regulatory information" refers to the process of regularly collecting the latest guidelines and regulatory information from regulatory agencies in each country and keeping the database up to date.

[1286] "User interface" refers to the screens and operating means that users use to search, view, and manipulate data using a system.

[1287] A "security incident" refers to an accident or incident related to information security, such as unauthorized access or data leakage.

[1288] "Security incident data" refers to data containing information related to security incidents, including specific examples such as attack methods, scope of impact, and countermeasures.

[1289] This invention is a system that streamlines the collection, formatting, summarization, visualization, and comparison of security incident data, as well as the updating of the latest security information. This system consists of a server, terminals, and users. The server is primarily responsible for data collection, automatic formatting, summary generation, data visualization, data comparison, and security information updates, while the terminals function as an interface with the users. Users use the system to search for and view the necessary data and obtain the information necessary for security countermeasures.

[1290] System Overview

[1291] 1. Data Collection

[1292] The server automatically collects data from public security incident databases, specialized blogs, security forums, etc. based on specific keywords (e.g., "phishing attack" or "ransomware"). Specifically, it uses various APIs and web scraping technologies to obtain the necessary data and store it in a database. The technologies used include Python, the Requests library, and BeautifulSoup.

[1293] 2. Data Formatting

[1294] The server reviews the collected data and converts it into a unified format. To ensure data integrity, it removes duplicate data and imputes missing values. MongoDB is used for database management.

[1295] 3. Data Summary

[1296] The server uses natural language processing techniques to extract important information from the data and generate summaries. For example, it uses the NLP library NLTK to extract important keywords from each incident report and provide a summary.

[1297] 4. Data Visualization

[1298] The server uses tools to generate graphs and charts based on the summarized data to visually display the data, using Python's matplotlib library to visualize the distribution of attack techniques as pie charts and frequency of occurrence as histograms.

[1299] 5. Data Comparison

[1300] The server compares the newly acquired data with the existing data, analyzing similarities and differences. The program then presents the results of these comparisons in a report. Statistical analysis libraries such as Scipy are used to collate the old and new data.

[1301] 6. Regulatory Updates

[1302] The server periodically retrieves the latest regulatory information from the websites of security regulatory agencies in each country and updates the database using web scraping technology and APIs.

[1303] 7. User Interface

[1304] Users use a smartphone or a head-mounted display (HMD) to search for the required data and view summaries and reports. The user interface is built using a web browser-based front-end framework (e.g., React).

[1305] Program examples and prompts

[1306] A specific example would be a scenario where a user uses a smartphone or HMD to search for data on the latest phishing attack trends, view the resulting summaries, and refine their company's countermeasures.

[1307] Example prompt sentence:

[1308] "Build a system that collects information on the latest phishing attack trends and provides a formatted summary of the data. Use Python to collect, format, summarize, visualize, compare, and update the latest trend information. Specifically, demonstrate a process for collecting information on phishing attacks from security blogs and forums and generating a summary and comparison report."

[1309] In this way, the present invention functions as a system that efficiently processes security incident data and enables a rapid response.

[1310] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1311] Step 1:

[1312] The server automatically collects data from public security incident databases, specialized blogs, and security forums based on specific security keywords (e.g., "phishing attack" or "ransomware"), using Python and the Requests library to send web requests and BeautifulSoup to parse HTML pages. The input data is the URL of the collected web page, and the output data is the parsed security-related information.

[1313] Step 2:

[1314] The server formats the collected data and converts it into a unified format. Specifically, it removes duplicate entries from the collected data and fills in missing values ​​with reliable default values. MongoDB is used as the database to maintain data integrity. The input data is the collected raw data, and the output data is formatted data in a unified format.

[1315] Step 3:

[1316] The server uses natural language processing (NLP) technology to extract important information from the formatted data and generate a summary. It uses the NLTK library to extract key keywords from the incident report and generate a summary. The input data is the formatted data, and the output data is the summarized information.

[1317] Step 4:

[1318] The server generates graphs and charts based on the summarized data. It uses Python's matplotlib library to visually display the data. Specifically, it displays the distribution of attack techniques as a pie chart and visualizes their occurrence frequency as a histogram. The input data is the summarized data, and the output data is the generated graphs and charts.

[1319] Step 5:

[1320] The server compares the newly acquired data with the existing data and analyzes the similarities and differences. This process uses statistical analysis libraries such as Scipy to collate and analyze the data, and provides the results in the form of a report. The input data is the old and new data, and the output data is a report of the comparison results.

[1321] Step 6:

[1322] The server periodically retrieves the latest regulatory information from the websites of security regulatory agencies in each country and updates the database. Using APIs and web scraping technology, the server collects the latest information and stores it in MongoDB. The input data is regulatory agency website information, and the output data is the updated database information.

[1323] Step 7:

[1324] Users use a smartphone or head-mounted display (HMD) to search for the data they need and view summaries and reports. The user interface is built using a front-end framework such as React to allow users to easily manipulate the data. The input data are search keywords from the user, and the output data are search results and summary reports.

[1325] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1326] The present invention provides a system that streamlines data collection, formatting, summarization, visualization, comparison, and updating of regulatory information in the pharmaceutical application process, and also includes an emotion engine that recognizes user emotions and adjusts interface display and data search results based on the emotions. The system has the following components and functions:

[1327] System Overview

[1328] The system consists of a server, a terminal, and a user. The server is mainly responsible for data collection, automatic formatting, summary generation, data visualization, data comparison, regulatory information updates, and user emotion recognition. The terminal functions as the interface with the user, allowing the user to search and view data through the system and create documents required for drug applications.

[1329] Program processing

[1330] 1. Data Collection

[1331] Subject: Server

[1332] The server uses pre-configured API keys and web scraping techniques to collect data from scientific papers, clinical trial databases, and drug regulatory agency websites. Specifically, the server uses the PubMed API to retrieve paper titles and abstracts related to "cancer treatment" and stores them in the database in JSON format. It also uses web scraping technology to crawl the FDA's latest guidelines page, retrieves and stores the content in text format.

[1333] 2. Data Formatting

[1334] Subject: Server

[1335] The server analyzes the collected data and unifies different data formats. Specifically, it parses JSON-formatted data, removes duplicate data, and fills in missing values ​​with appropriate default values. For example, in the case of paper data, if multiple papers with the same title are obtained, they are merged into one, and if information such as author names or publication dates is missing, it is filled in from reliable external information.

[1336] 3. Data Summary

[1337] Subject: Server

[1338] The server uses natural language processing (NLP) technology to extract important information from the collected data and generate summaries. Specifically, it uses an NLP library (e.g., NLTK or SpaCy) to analyze the main keywords in the paper, as well as descriptions of therapeutic effects and side effects, and generates summaries based on this. The generated summaries are then stored in a database.

[1339] 4. Data Visualization

[1340] Subject: Server

[1341] The server uses tools to generate graphs and charts based on the summarized data for visualization. For example, Matplotlib is used to generate line graphs showing the effectiveness of each treatment and bar graphs showing the frequency of side effects. The generated graphs and charts are converted into a format that can be viewed by the user interface and displayed on a dashboard in real time.

[1342] 5. Data Comparison

[1343] Subject: Server

[1344] The server compares the newly acquired data with existing data and analyzes differences and commonalities. Specifically, it cross-references data on existing cancer treatments with data on new treatments, comparing them in terms of treatment effectiveness, side effects, etc. The comparison results are generated in report format, making them easy for users to understand.

[1345] 6. Regulatory Updates

[1346] Subject: Server

[1347] The server periodically retrieves the latest information from the websites of drug regulatory agencies in each country and updates the database. Specifically, it uses a web scraping tool to retrieve the latest guidelines and regulatory changes from the FDA and EMA, stores them in the database, and sends notifications to users when changes are confirmed.

[1348] 7. Data Search and Browsing

[1349] Subject: User

[1350] Users use the device's web interface to search for and retrieve the data they need. Specifically, they enter keywords such as "new treatments for lung cancer" into the search screen and click the search button to send a request to the server. The server then retrieves relevant summaries and comparison reports from the database and displays them on the user's device.

[1351] 8. Emotion Recognition and Interface Adjustment

[1352] Subject: Server

[1353] The server uses an emotion engine to recognize the user's emotions, for example, by analyzing the user's facial expressions and tone of voice, and then adjusts the priority of data search results and optimizes the interface display.

[1354] Examples:

[1355] If the server detects that a user is feeling stressed while completing a regulatory application, it will simplify the interface and adjust the display to make important information easier to access. If there are too many search results, it will adjust the ranking to prioritize the most relevant data.

[1356] 9. Preparation of pharmaceutical application documents

[1357] Subject: User

[1358] Users create regulatory application documents based on the displayed summaries and comparison reports. Specifically, they download templates provided by the system and create documents by entering the displayed information. They can also submit the created documents for review by experts, who can make revisions based on the feedback.

[1359] In this way, the present invention, equipped with an emotion engine, is a system that not only efficiently collects, formats, summarizes, visualizes, and compares the various data required for the pharmaceutical application process, and updates the latest regulatory information, but also further supports the application process and improves its accuracy and speed by optimizing the interface in consideration of the user's emotions.

[1360] The processing flow will be explained below.

[1361] Step 1: Data collection

[1362] Subject: Server

[1363] The server uses pre-configured API keys and web scraping techniques to collect data from scientific papers, clinical trial databases, and drug regulatory agency websites. Specifically, the server uses the PubMed API to retrieve paper titles and abstracts related to "cancer treatment." The retrieved data is stored in a database in JSON format. It also uses web scraping technology to crawl the FDA's latest guidelines page, retrieve their contents in text format, and store them.

[1364] Step 2: Data Shaping

[1365] Subject: Server

[1366] The server analyzes the collected data and unifies different data formats. Specifically, it parses JSON-formatted data, removes duplicate data, and fills in missing values ​​with appropriate default values. For example, if multiple papers with the same title are retrieved, they are merged into one, and if missing information such as author names or publication dates is provided, it is filled in using reliable external information.

[1367] Step 3: Data Summarization

[1368] Subject: Server

[1369] The server uses natural language processing (NLP) technology to extract important information from the collected data and generate summaries. Specifically, it uses NLP libraries (e.g., NLTK and SpaCy) to analyze each paper's main keywords, therapeutic effects, and side effects. Based on the extracted information, it automatically generates summaries and stores them in a database.

[1370] Step 4: Data visualization

[1371] Subject: Server

[1372] The server uses tools to generate graphs and charts based on the summarized data for visualization. Specifically, Matplotlib is used to generate line graphs showing the effectiveness of each treatment and bar graphs showing the frequency of side effects. The generated graphs and charts are converted into a format that can be viewed by the user interface and displayed on a dashboard in real time.

[1373] Step 5: Data comparison

[1374] Subject: Server

[1375] The server compares the newly acquired data with existing data and analyzes differences and commonalities. Specifically, it cross-references data on existing cancer treatments with data on new treatments to evaluate differences in treatment effectiveness and side effects. The comparison results are generated in a detailed report format that users can easily understand.

[1376] Step 6: Update regulatory information

[1377] Subject: Server

[1378] The server periodically retrieves the latest information from the websites of each country's drug regulatory agency and updates the database. Specifically, it uses a web scraping tool to retrieve the latest guidelines and regulatory changes from the FDA and EMA, and stores them in the database. If any changes are confirmed, the server notifies the user.

[1379] Step 7: Emotion Recognition

[1380] Subject: Server

[1381] The server uses an emotion engine to recognize the user's emotions. Specifically, it analyzes the user's facial expressions, voice tone, input speed, etc. to evaluate the user's emotions (e.g., stress, satisfaction). The results of emotion recognition are used to adjust the interface and prioritize search results.

[1382] Step 8: Interface Adjustment

[1383] Subject: Server

[1384] The server adjusts the interface display content based on the user's emotions recognized by the emotion engine. Specifically, if the user is feeling stressed, the interface will be simplified and important information will be highlighted. The server will also prioritize the display of content that matches the user's interests.

[1385] Step 9: Search and view data

[1386] Subject: User

[1387] Users use the device's web interface to search for and retrieve the data they need. Specifically, they enter keywords such as "new treatments for lung cancer" into the search screen and click the search button to send a request to the server. The server then retrieves relevant summaries and comparison reports from the database and displays them on the user's device.

[1388] Step 10: Prepare regulatory application documents

[1389] Subject: User

[1390] Users create regulatory application documents based on the displayed summaries and comparison reports. Specifically, they download templates provided by the system and create documents by entering the displayed information. They then submit the created documents for review by experts, who provide feedback and make revisions.

[1391] In this way, the present invention, equipped with an emotion engine, is a system that not only efficiently collects, formats, summarizes, visualizes, and compares the various data required for the pharmaceutical application process, and updates the latest regulatory information, but also further supports the application process and improves its accuracy and speed by optimizing the interface in consideration of the user's emotions.

[1392] Example 2

[1393] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1394] The pharmaceutical application process requires a huge amount of data collection, complex data formatting, detailed summarization, effective data visualization, rigorous data comparison, and the latest regulatory information updates, making it extremely time-consuming and labor-intensive. Furthermore, the lack of user-friendly interface adjustments can lead to reduced work efficiency. Therefore, there is a need for a system that can efficiently complete pharmaceutical application procedures without causing stress to users.

[1395] The specific processing by the specific processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for automatically collecting clinical data and non-clinical data, means for formatting the collected data and converting it into a unified format, means for extracting important information from the formatted data and generating a summary, means for visualizing the summary data in the form of graphs or charts, means for comparing new data with existing data and providing the results in the form of a report, means for periodically obtaining the latest pharmaceutical regulatory information and updating the database, means for a user to use an interface to search for necessary data and view summaries and reports, means for analyzing user emotions using an emotion engine and adjusting the interface display and search results, and means for generating prompt sentences to input to the generative AI model. This enables efficient data collection, data formatting, summarization, visualization, comparison, and updating of the latest regulatory information in the pharmaceutical application process, and further enables significant improvement in work efficiency and user experience through interface adjustment based on user emotions.

[1396] "Data Collection Instruments" refers to processes or mechanisms for the automated collection of clinical and non-clinical data.

[1397] "Data formatting means" refers to the process or method of analyzing collected data and converting it into a unified format.

[1398] "Data summarization means" refers to a mechanism or process that extracts important information from formatted data and generates a summary.

[1399] "Data visualization methods" refers to the process of visualizing summarized data in the form of graphs or charts.

[1400] "Data comparison means" refers to a method or technology that compares new data with existing data and provides the results of the comparison in a report format.

[1401] "Regulatory information update means" refers to the process or method of periodically obtaining the latest pharmaceutical regulatory information and updating the database.

[1402] "Data retrieval means" refers to the ability for users to use the interface to retrieve required data and view summaries and reports.

[1403] "Emotion analysis means" refers to a mechanism that uses an emotion engine to analyze user emotions and adjusts interface display and search results based on the results.

[1404] "Prompt generation means" refers to the process or method for generating prompt sentences to be input into a generative AI model.

[1405] "Automatic means" refers to processes or techniques that do not require manual intervention and are operated automatically by the system.

[1406] "User interface" refers to the screens and methods by which a user interacts with a system and searches and views data.

[1407] The present invention is a system that streamlines data collection, formatting, summarization, visualization, comparison, and updating of regulatory information in the pharmaceutical application process. Furthermore, it aims to improve operability by incorporating an emotion engine that recognizes user emotions and adjusts interface display and data search results based on the emotions. The system of the present invention has the following specific configuration:

[1408] System Components

[1409] 1. Server

[1410] The server is the central part of the system and performs the following functions:

[1411] Data collection: Using PubMed API and web scraping techniques, we collect data from scientific articles, clinical trial databases, and drug regulatory agency websites. For example, we use PubMed API to retrieve titles, authors, and abstracts of articles related to "cancer treatment" and store them in a database in JSON format.

[1412] Data Transformation: Analyze the collected data and convert it into a unified format, specifically removing duplicate entries in the JSON data and imputing missing values ​​from external reliable data.

[1413] Data summarization: Using natural language processing (NLP) techniques, we extract important information from the collected data and create summaries. For example, we use NLP libraries (e.g., NLTK and SpaCy) to analyze the main keywords in papers and descriptions of treatment effects and side effects.

[1414] Data visualization: Visualize the data using tools that generate graphs and charts (e.g., Matplotlib) based on the summarized data. The generated graphs and charts are displayed in real time on a dashboard.

[1415] Data comparison: Compare newly acquired data with existing data and analyze differences and commonalities. Generate a report of the comparison results so that users can easily understand them.

[1416] Regulatory information updates: Regularly retrieve the latest information from the websites of national drug regulatory agencies and update the database. For example, use web scraping tools to retrieve the latest guidelines from the FDA and EMA.

[1417] Emotion recognition: An emotion engine is used to analyze emotions from a user's facial expressions and tone of voice to adjust interface display and data search result priorities.

[1418] 2. Terminal

[1419] The terminal acts as an interface with the user:

[1420] Data search and browsing: Users use a web interface to search for and retrieve the data they need. For example, they enter keywords such as "new treatments for lung cancer" into the search screen and click the search button to send a request to the server. The server retrieves relevant summaries and comparison reports from the database and displays them on the terminal.

[1421] Interface Adaptation: If the emotion engine detects stress in the user, it will automatically simplify the interface and adjust to make important information easier to access.

[1422] 3. Users

[1423] Users use the system to:

[1424] Data retrieval: Users can search for the required data through a web interface and view summaries and reports.

[1425] Creating regulatory application documents: Users create regulatory application documents based on the displayed summary and comparison reports. They download templates provided by the system and enter the required information. They submit the completed documents for expert review and make revisions based on the feedback.

[1426] Specific examples

[1427] 1. The server uses the PubMed API to search for papers on "lung cancer treatments" and retrieves their titles, authors, and abstracts. Then, it uses a web crawler to parse the FDA's "New Treatment Guidelines" page and saves the content in text format.

[1428] 2. The server eliminates duplicates from the collected data and completes missing author names, publication dates, etc. from external databases.

[1429] 3. Using natural language processing technology, generate a summary from the collected papers: "This new drug improved the survival rate of lung cancer patients by 20%, but nausea was reported as a side effect."

[1430] 4. Use Matplotlib to generate a line graph showing the effectiveness of "Lung Cancer Treatment A" and a bar graph showing the frequency of side effects. Display these graphs on a dashboard.

[1431] 5. Compare the effectiveness and side effects of "existing treatment A" and "new treatment B" and generate a comparative report showing that new treatment B is slightly less effective but has fewer side effects.

[1432] 6. Use a web crawler to regularly check for the latest EMA guidelines and store new regulatory information in a database.

[1433] 7. If users are experiencing stress while completing regulatory submissions, simplify the interface, make important information more accessible, and adjust the ranking of search results to surface more relevant information.

[1434] Example prompts to input to the generative AI model

[1435] "Collect the latest research data on lung cancer treatments, summarize and compare them, and use them to draft regulatory applications."

[1436] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1437] Step 1:

[1438] Data collection

[1439] The server collects data using the PubMed API and a web scraping tool. Specifically, it retrieves the titles, authors, and abstracts of papers related to "lung cancer treatment" from the PubMed API. It also uses the web scraping tool to crawl the FDA's latest guidelines page and saves the content in text format.

[1440] Input: Keyword "lung cancer treatment", API key, website URL

[1441] Output: Paper data (JSON format), guideline text data

[1442] Specific operation: The server collects paper data on "lung cancer treatment" from the PubMed API, and at the same time uses a web crawler to retrieve and save the contents of the FDA's latest guidelines page.

[1443] Step 2:

[1444] Data Formatting

[1445] The server parses the collected JSON-formatted data and unifies different formats, specifically removing duplicate data and completing missing values ​​from external information.

[1446] Input: Collected paper data (JSON format), guideline text data

[1447] Output: Formatted data (Uniform JSON format)

[1448] What it does: The server reads a file such as "Lung Cancer Treatments.pdf," merges duplicate entries, and completes missing author names and publication dates.

[1449] Step 3:

[1450] Data Summary

[1451] The server uses natural language processing (NLP) technology to analyze the collected data and extract important information. It uses NLP libraries (NLTK and SpaCy) to analyze the main keywords of the papers, as well as descriptions of therapeutic effects and side effects, and generates summaries.

[1452] Input: Formatted data (Uniform JSON format)

[1453] Output: Summary data (text format)

[1454] Specific operation: The server analyzes the abstract of a paper on "lung cancer treatment" using SpaCy, generates a summary, and saves it.

[1455] Step 4:

[1456] Data Visualization

[1457] The server generates graphs and charts based on the summarized data, using Matplotlib to graphically display the effectiveness of each treatment and the frequency of side effects.

[1458] Input: Summary data (text format)

[1459] Output: Graphs and charts (image format)

[1460] Specific operation: Based on the summarized data, the server uses Matplotlib to create a line graph showing the effectiveness of "Lung Cancer Treatment A" and a bar graph showing the frequency of side effects.

[1461] Step 5:

[1462] Data comparison

[1463] The server cross-references the newly acquired data with the existing data, analyzes the differences and commonalities between them, and generates a report of the comparison results.

[1464] Input: Summary data (text format), existing data

[1465] Output: Comparison report (text format)

[1466] Specific operation: The server cross-references the effectiveness and side effects of "existing treatment A" and "new treatment B" and generates a report stating that treatment B is slightly less effective but has fewer side effects.

[1467] Step 6:

[1468] Regulatory updates

[1469] The server periodically scrapes the websites of each country's drug regulatory agency to collect the latest information and update the database.

[1470] Input: Regulatory agency website URL

[1471] Output: Latest regulatory information (text format)

[1472] What it does: The server uses a web crawler to crawl the latest FDA and EMA guideline pages and stores the necessary information in a database.

[1473] Step 7:

[1474] Data Search and Viewing

[1475] Users use the device's web interface to search for and retrieve the data they need. By entering search keywords and clicking the search button, the request is sent to the server, which retrieves the relevant data and displays it on the user's device.

[1476] Input: Search keyword

[1477] Output: Relevant data (summary and comparison reports)

[1478] Specific operation: When a user enters "new treatments for lung cancer" into the search box and performs a search, the server displays relevant summaries and reports on the device.

[1479] Step 8:

[1480] Emotion Recognition and Interface Adjustment

[1481] The server uses an emotion engine to analyze the user's emotions, including facial expressions and tone of voice, and adjusts the interface display and search result priorities based on the results.

[1482] Input: User's facial expression data, voice data

[1483] Output: Tailored interface, prioritized search results

[1484] What it does: The server collects data from the user's webcam and microphone and simplifies the interface if the user is stressed.

[1485] Step 9:

[1486] Preparation of pharmaceutical application documents

[1487] Users create regulatory application documents based on the displayed summary and comparison reports. They download templates provided by the system, enter the required information, and submit the completed documents for expert review, where they can make revisions based on the feedback.

[1488] Inputs: Summary data, comparison reports, templates

[1489] Output: Pharmaceutical application documents

[1490] What it does: A user downloads a template from the system, copies information from the summary and comparison reports to complete the application, and then submits it to an expert for feedback and makes any necessary revisions.

[1491] (Application example 2)

[1492] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1493] Conventional systems do not efficiently collect, organize, summarize, visualize, and compare clinical and non-clinical data, or update and search regulatory information. Furthermore, they do not allow for interface adjustments that take into account users' emotions and stress levels, or optimization of productivity and work environments. This makes it difficult for managers to grasp the status of workers in real time and provide an optimal work environment.

[1494] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1495] In this invention, the server includes a means for automatically collecting clinical and non-clinical data, a means for formatting the data and converting it into a unified format, and a means for extracting important information and generating summaries. This allows for efficient processing of various types of data. It also includes a means for collecting biometric information from workers, analyzing their emotions and stress levels, a means for adjusting the interface display content based on their emotions and stress levels, and a means for visualizing production data and optimizing production efficiency and the work environment. This allows managers to understand the status of workers in real time and provide a comfortable and efficient work environment.

[1496] "Clinical Data" refers to patient medical information collected in a medical setting, including data related to diagnosis, treatment, and follow-up.

[1497] "Non-clinical data" refers to research data obtained outside of clinical trials, including data obtained from animal experiments, cell experiments, etc.

[1498] "Data transformation" refers to the process of analyzing collected data and unifying different formats and forms.

[1499] "Unified format" refers to a format that converts data obtained from different sources into a consistent format.

[1500] "Important information extraction" refers to the process of selecting the necessary data from a large amount of data according to a specific purpose or requirement.

[1501] A "summary" is a document or text that succinctly summarizes the main points or essence of the collected data.

[1502] "Worker biometric information" refers to data about a worker's physical condition, such as heart rate, body temperature, and stress level.

[1503] Analyzing "emotions and stress levels" refers to the process of assessing a worker's emotional state and stress level from collected biometric information.

[1504] "Adjusting the interface display content" refers to the process of changing the system display content and operation method according to the user's condition to provide the optimal user experience.

[1505] "Production data" refers to data related to the manufacturing process of products in factories and production lines, including operating hours, failure rates, production efficiency, etc.

[1506] "Production efficiency" refers to an indicator that shows how much product can be produced using a certain amount of resources (time, manpower, materials).

[1507] "Working environment" refers to the conditions in the place where workers perform their work, including temperature, humidity, noise level, etc.

[1508] The present invention provides a system for achieving efficient data management and optimizing the working environment at a production site. The system includes the following hardware and software.

[1509] Hardware

[1510] IoT devices and sensors: installed to collect temperature, humidity, and vibration data.

[1511] Wearable devices: Worn to collect workers' biometric information (heart rate, stress levels, etc.).

[1512] Smart glasses: worn by workers, they display the information they need in real time while they work.

[1513] software

[1514] Server: Responsible for data collection, formatting, analysis, visualization, comparison, and regulatory information updates.

[1515] Data Collection API: Collects data from sensors and wearable devices using the MQTT protocol or Bluetooth communication.

[1516] Parsing libraries: Use natural language processing libraries such as NLTK or SpaCy to generate summaries of your data.

[1517] Visualization tools: Use Matplotlib and Plotly to visualize data in the form of graphs and charts.

[1518] Machine learning model: Using TensorFlow to analyze worker emotions and stress levels.

[1519] System Details

[1520] 1. Data Collection:

[1521] The server collects data in real time from IoT devices and sensors, capturing temperature, humidity, and vibration data using the MQTT protocol and storing them in a database. It also collects biometric data from workers' smart glasses and wearable devices via Bluetooth and transmits it to the server.

[1522] 2. Data Formatting:

[1523] The server parses the collected data and converts it into a standard format, cleaning it using Python's Pandas library, removing duplicates, and imputing missing values.

[1524] 3. Data Summary:

[1525] Use natural language processing libraries (e.g., NLTK, SpaCy) to extract important information from the collected data and generate summaries. For example, analyze temperature fluctuations obtained from multiple sensor data and generate a summary.

[1526] 4. Data visualization:

[1527] The server generates graphs and charts in real time based on the collected data, using Matplotlib and Plotly to visually display the operation status of the production line and the stress levels of workers.

[1528] 5. Data comparison:

[1529] The server compares and analyzes newly acquired data with existing data to identify anomalies and patterns, particularly by comparing machine operating hours and failure rates with past data.

[1530] 6. Regulatory Updates:

[1531] Regularly obtain information on each country's occupational safety regulations and environmental standards using a web scraping tool (e.g., BeautifulSoup) and update the database. If there are any changes in regulations, notify the administrator.

[1532] 7. Emotion Recognition and Interface Adjustment:

[1533] The server analyzes the worker's emotions and stress levels from facial expressions and biometric data. It uses TensorFlow to classify emotions and adjusts the interface display if it recognizes a high level of stress. For example, if a worker's stress level is high, it changes the work priority and displays an alert suggesting a break.

[1534] 8. Optimizing production efficiency and working environment:

[1535] The server analyzes the collected production data and optimizes production efficiency and the working environment, thereby improving productivity in the factory and providing a comfortable working environment for workers.

[1536] Specific examples

[1537] For example, when Worker A is wearing smart glasses while working, the glasses measure his heart rate and stress level using a data collection API sent from the server. Using TensorFlow, if the data determines that Worker A's stress level is high, the server changes the work priority and immediately displays an alert recommending a break. Also, if the factory temperature or humidity data exceeds the standard, the server simultaneously sends an alert and issues instructions for improving the environment.

[1538] Example prompts for generative AI models

[1539] Based on Worker A's latest biometric data, generate a response plan for when stress levels are high.

[1540] Working environment data: Temperature: 28°C, Humidity: 60%, Vibration: Low

[1541] Vital Data: Heart rate: 80 bpm, Stress level: High

[1542] In this way, the system can maintain production efficiency while ensuring worker health and comfort.

[1543] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1544] Step 1:

[1545] The server collects environmental data (temperature, humidity, vibration) from IoT devices and sensors.

[1546] Input: Data sent from each sensor via MQTT protocol.

[1547] Processing: The server parses the received data and stores it in the database.

[1548] Output: Formatted environmental data (Temperature: 25°C, Humidity: 60%, Vibration: Low).

[1549] Step 2:

[1550] The server collects biometric information (heart rate, stress level) from smart glasses or wearable devices worn by workers.

[1551] Input: Real-time biometric data transmitted via Bluetooth.

[1552] Processing: The server receives the biometric data and stores it in a database.

[1553] Output: Formatted biometric data (heart rate: 80 bpm, stress level: high).

[1554] Step 3:

[1555] The server analyzes the collected environmental and biological data and converts it into a unified format.

[1556] Input: Environmental and biological data stored in a database.

[1557] Processing: Using Python's Pandas library, the data is cleaned, missing values ​​are imputed, and it is converted into a standard format.

[1558] Output: A uniformly formatted dataset.

[1559] Step 4:

[1560] The server extracts important information from the dataset and generates a summary.

[1561] Input: A uniformly formatted dataset.

[1562] Processing: Use NLTK or SpaCy to analyze the data and generate summaries.

[1563] Output: Summarized data.

[1564] Step 5:

[1565] The server visualizes the summarized data and displays it on the factory dashboard.

[1566] Input: Summarized data.

[1567] Processing: Visualize the data in the form of graphs and charts using Matplotlib and Plotly.

[1568] Output: Graphs and charts displayed on a dashboard.

[1569] Step 6:

[1570] The server compares the newly acquired data with past data to detect any anomalies.

[1571] Input: Newly acquired data and existing datasets.

[1572] Processing: Use Pandas and SciPy to compare data and identify anomalies and patterns.

[1573] Output: Anomaly detection results report.

[1574] Step 7:

[1575] The server regularly obtains each country's occupational safety regulations and environmental standards using a web scraping tool and updates the database.

[1576] Input: Websites that provide regulatory information.

[1577] Processing: Use BeautifulSoup to perform web scraping and store new regulatory information in a database.

[1578] Output: Updated regulatory information.

[1579] Step 8:

[1580] The server analyzes the worker's biometric data and adjusts the interface display based on their emotions and stress levels.

[1581] Input: Worker's biometric data.

[1582] Processing: TensorFlow is used to analyze biometric data and classify emotions and stress levels, while simultaneously adjusting the interface display.

[1583] Output: The adjusted interface display.

[1584] Step 9:

[1585] The server analyzes production data and optimizes production efficiency and the working environment.

[1586] Input: Production data and work environment data.

[1587] Processing: Using data analysis tools, optimization algorithms are applied to evaluate production efficiency.

[1588] Output: Improvement suggestions and alert messages.

[1589] For example, if Worker A's biometric data is analyzed and it is determined that his stress level is high, an alert will be displayed immediately recommending that he take a break. Also, if the temperature or humidity in the factory exceeds the standard, instructions to improve the environment will be automatically sent.

[1590] Example prompts for generative AI models

[1591] Based on Worker A's latest biometric data, generate a response plan for when stress levels are high.

[1592] Working environment data: Temperature: 28°C, Humidity: 60%, Vibration: Low

[1593] Vital Data: Heart rate: 80 bpm, Stress level: High

[1594] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1595] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs 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.

[1596] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1597] [Fourth embodiment]

[1598] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1599] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1600] 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).

[1601] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. 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 microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1602] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1603] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1604] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1605] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1606] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1607] The specific processing program 56 is an example of a "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 processing 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.

[1608] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1609] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1610] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1611] The present invention is a system for streamlining data collection, formatting, summarization, visualization, comparison, and updating of regulatory information in the pharmaceutical application process. This system has the following components and functions.

[1612] System Overview

[1613] The system consists of a server, terminals, and users. The server is primarily responsible for data collection, automatic formatting, summary generation, data visualization, data comparison, and regulatory information updates, while the terminals function as an interface with users. Users use the system to search for and view the necessary data and create the documents required for pharmaceutical applications.

[1614] Program processing

[1615] 1. Data Collection

[1616] Subject: Server

[1617] The server automatically collects data based on pre-defined keywords from publicly available scientific papers, clinical trial databases, drug regulatory agency websites, etc. Specifically, it uses various APIs and web scraping technologies to retrieve the required data and store it in a database.

[1618] Examples:

[1619] The server retrieves the titles, authors, abstracts, and full texts of relevant articles from PubMed using the keyword "cancer treatment" and stores them in a database.

[1620] 2. Data Formatting

[1621] Subject: Server

[1622] The server reviews the collected data and converts it into a unified format, removing duplicates and imputing missing values ​​to ensure data integrity.

[1623] Examples:

[1624] The server analyzes multiple article data retrieved from PubMed, removes duplicate entries, and adds reliable default values ​​for missing entries.

[1625] 3. Data Summary

[1626] Subject: Server

[1627] The server uses natural language processing technology to extract important information from the data and generate a summary of key points (e.g., treatment effects, side effects, etc.).

[1628] Examples:

[1629] The server extracts information about each paper's key keywords, therapeutic effects, and side effects, and provides it as a summary.

[1630] 4. Data Visualization

[1631] Subject: Server

[1632] The server uses tools to generate graphs and charts based on the summarized data to visually display the data.

[1633] Examples:

[1634] The server visualizes the treatment efficacy data as a line graph and displays the frequency of side effects as a bar graph.

[1635] 5. Data Comparison

[1636] Subject: Server

[1637] The server compares the newly acquired data with the existing data, analyzes similarities and differences, and provides the comparison results in the form of a report.

[1638] Examples:

[1639] The server compares data on existing cancer treatments with data on newly acquired treatments and generates a detailed report on differences in treatment effectiveness and side effects.

[1640] 6. Regulatory Updates

[1641] Subject: Server

[1642] The server periodically retrieves the latest regulatory information from the websites of each country's drug regulatory agency and updates the database.

[1643] Examples:

[1644] The server uses web scraping technology to obtain the latest US FDA guidelines and imports them into the database for updating.

[1645] 7. User Interface

[1646] Subject: User

[1647] Users use a web interface to search for the data they need, view summaries and reports, and generate regulatory submissions based on the information provided.

[1648] Examples:

[1649] The user searches for data on "new treatments for lung cancer" in the interface, checks the obtained summary and comparison reports, and downloads a template for pharmaceutical application documents based on the information and fills in the required fields to complete the application.

[1650] In this way, the present invention is a system that automates application procedures and improves accuracy and speed by efficiently collecting, formatting, summarizing, visualizing, and comparing the various data required for the pharmaceutical application process, as well as updating the latest regulatory information.

[1651] The processing flow will be explained below.

[1652] Step 1: Data collection

[1653] Subject: Server

[1654] The server uses pre-configured API keys and web scraping techniques to collect data from scientific papers, clinical trial databases, and drug regulatory agency websites. Specifically, the server uses the PubMed API to retrieve paper titles and abstracts related to "cancer treatment" and stores them in a database in JSON format. It also uses web scraping technology to crawl the FDA's latest guidelines page, retrieves their contents in text format, and stores them.

[1655] Step 2: Data Shaping

[1656] Subject: Server

[1657] The server analyzes the collected data and unifies different data formats. Specifically, it parses JSON-formatted data, removes duplicate data, and fills in missing values ​​with appropriate default values. For example, in the case of paper data, if multiple papers with the same title are obtained, they are merged into one, and if information such as author names or publication dates is missing, it is filled in from reliable external information.

[1658] Step 3: Data Summarization

[1659] Subject: Server

[1660] The server uses natural language processing (NLP) technology to extract important information from the collected data and generate summaries. Specifically, it uses an NLP library (e.g., NLTK or SpaCy) to analyze the main keywords in the paper, as well as descriptions of therapeutic effects and side effects, and generates summaries based on this. The generated summaries are then stored in a database.

[1661] Step 4: Data visualization

[1662] Subject: Server

[1663] The server uses tools to generate graphs and charts based on the summarized data for visualization. For example, Matplotlib is used to generate line graphs showing the effectiveness of each treatment and bar graphs showing the frequency of side effects. The generated graphs and charts are converted into a format that can be viewed by the user interface and displayed on a dashboard in real time.

[1664] Step 5: Data comparison

[1665] Subject: Server

[1666] The server compares the newly acquired data with existing data and analyzes differences and commonalities. Specifically, it cross-references data on existing cancer treatments with data on new treatments, comparing them in terms of treatment effectiveness, side effects, etc. The comparison results are generated in report format, making them easy for users to understand.

[1667] Step 6: Update regulatory information

[1668] Subject: Server

[1669] The server periodically retrieves the latest information from the websites of each country's drug regulatory agency and updates the database. Specifically, it uses a web scraping tool to retrieve the latest guidelines and regulatory changes from the FAA and EMA, stores them in the database, and sends notifications to users when changes are confirmed.

[1670] Step 7: Search and view data

[1671] Subject: User

[1672] Users use the device's web interface to search for and retrieve the data they need. Specifically, they enter keywords such as "new treatments for lung cancer" into the search screen and click the search button to send a request to the server. The server then retrieves relevant summaries and comparison reports from the database and displays them on the user's device.

[1673] Step 8: Prepare regulatory submission documents

[1674] Subject: User

[1675] Users create regulatory application documents based on the displayed summaries and comparison reports. Specifically, they download templates provided by the system and create documents by entering the displayed information. They can also submit the created documents for review by experts, who can make revisions based on the feedback.

[1676] Example 1

[1677] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1678] The current pharmaceutical application process involves a wide range of tasks, including collecting, organizing, summarizing, visualizing, comparing, and updating regulatory information, and is extremely time-consuming and labor-intensive. Furthermore, the quality and consistency of the collected data varies, and the accuracy and efficiency of application document preparation is not fully guaranteed. This leads to delays in the application process and the submission of incorrect information.

[1679] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1680] In this invention, the server includes, in data collection, means for automatically obtaining data from publicly available information sources based on preset keywords, means for reviewing the collected data and checking for format inconsistencies and missing values, means for formatting the collected data, converting it into a unified format, deleting duplicate data, and imputing missing values, means for extracting important information from the collected data using natural language processing technology and generating a summary, means for visualizing the summary data in the form of a graph or chart, means for comparing newly obtained data with existing data, analyzing similarities and differences, and providing the results in the form of a report, means for periodically obtaining the latest regulatory information and updating the database, means for a user to use an interface to search for necessary data and view summaries and reports, and means for providing application document templates based on the search results and for the user to enter the necessary information to create them. This makes it possible to improve the efficiency and accuracy of the pharmaceutical application process.

[1681] "Data harvesting" refers to the automatic retrieval of data from publicly available sources based on pre-defined keywords.

[1682] "Format inconsistency" refers to a situation where collected data exists in different structures or formats.

[1683] "Missing values" refers to a state in which some data is missing or necessary information is not entered.

[1684] "Duplicate data" refers to a situation in which multiple pieces of data with the same content exist.

[1685] "Missing value imputation" refers to filling in missing data based on predictions or reliable sources.

[1686] "Natural language processing" refers to the technology for understanding, analyzing, and generating human language.

[1687] "Extracting important information" refers to extracting only specific, necessary information from collected data.

[1688] A "summary" is a short summary of the main points from a long piece of text or a large amount of data.

[1689] "Visualization" refers to the display of data in a visual format such as a graph or chart.

[1690] "Report format" refers to a format in which the results of analysis and comparison are systematically summarized in documents and charts.

[1691] "Regulatory information" refers to information on norms and standards provided by public institutions, such as laws, regulations, and guidelines.

[1692] "Database updating" refers to adding new information to existing data and keeping it up to date.

[1693] "User interface" refers to the screens and controls that allow a user to interact with a computer system or application.

[1694] "Application template" refers to a template for application documents in a predefined format.

[1695] This invention is a system that streamlines data collection, formatting, summarization, visualization, comparison, and updating of regulatory information in the pharmaceutical application process. This system is mainly composed of a server, terminals, and users. The specific system structure and operation are described in detail below.

[1696] System Configuration

[1697] server:

[1698] The server is responsible for data collection, automatic formatting, summary generation, data visualization, data comparison, and regulatory information updates. The server uses the following hardware and software:

[1699] Hardware: High-performance server (e.g., Xeon-based server)

[1700] Software: Python, natural language processing libraries (e.g., NLTK, SpaCy), data visualization tools (e.g., Matplotlib, D3.js)

[1701] Device:

[1702] The terminal provides the interface through which the user interacts with the system. The terminal uses the following hardware and software:

[1703] Hardware: Personal computers, tablets

[1704] Software: Web browser (e.g. Google Chrome, Firefox)

[1705] User:

[1706] Users are drug developers and regulatory officials involved in the regulatory application process, who use the system to search for the data they need, view summaries and reports, and prepare application documents.

[1707] How it works

[1708] 1. Data Collection:

[1709] The server collects data based on pre-defined keywords from published scientific papers, clinical trial databases, drug regulatory agency websites, etc., using various API calls and web scraping techniques.

[1710] Examples:

[1711] The server retrieves the titles, authors, abstracts, and full texts of relevant articles from PubMed using the keyword "cancer treatment" and stores them in a database.

[1712] 2. Data Formatting:

[1713] The server reviews the collected data and converts it into a unified format, removing duplicates and imputing missing values ​​to ensure data integrity.

[1714] Examples:

[1715] The server analyzes multiple article data retrieved from PubMed, removes duplicate entries, and adds reliable default values ​​for missing entries.

[1716] 3. Data Summary:

[1717] The server uses natural language processing techniques to extract key information from the data and generate a summary.

[1718] Examples:

[1719] The server extracts information about each paper's key keywords, therapeutic effects, and side effects, and provides it as a summary.

[1720] 4. Data visualization:

[1721] The server uses tools to generate graphs and charts based on the summarized data to visually display the data.

[1722] Examples:

[1723] The server visualizes the treatment efficacy data as a line graph and displays the frequency of side effects as a bar graph.

[1724] 5. Data comparison:

[1725] The server compares the newly acquired data with the existing data, analyzes similarities and differences, and provides the results in a report.

[1726] Examples:

[1727] The server compares data on existing cancer treatments with data on newly acquired treatments and generates a detailed report on differences in treatment effectiveness and side effects.

[1728] 6. Regulatory Updates:

[1729] The server periodically retrieves the latest regulatory information from the websites of each country's drug regulatory agency and updates the database.

[1730] Examples:

[1731] The server uses web scraping technology to obtain the latest US FDA guidelines and imports them into the database for updating.

[1732] 7. User Interface:

[1733] Users use a web interface to search for the data they need, view summaries and reports, and download application templates based on their search results, then fill in the required fields to complete the application.

[1734] Examples:

[1735] Users can search for data on "new treatments for lung cancer" in the interface, check the resulting summary and comparison reports, and download a template for a pharmaceutical application based on the information, then fill in the required fields to create it.

[1736] Prompt Sentence Examples

[1737] "Collect the latest articles on 'cancer treatment' from PubMed and create summaries that extract key keywords, therapeutic effects, and side effects."

[1738] As described above, this system comprehensively covers the entire process from data collection to preparation of application documents, dramatically improving the efficiency and accuracy of pharmaceutical applications.

[1739] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1740] Step 1: Data collection

[1741] Input: Predefined keywords (e.g., "cancer treatment").

[1742] Specific behavior:

[1743] The server automatically collects data from various data sources (e.g., scientific paper databases, clinical trial databases, regulatory agency websites, etc.) based on configured keywords, using API calls and web scraping techniques.

[1744] Data processing:

[1745] The acquired data is stored in a temporary database.

[1746] Output: The raw data collected.

[1747] Step 2: Data review

[1748] Input: The raw data collected.

[1749] Specific behavior:

[1750] The server reviews the collected data and performs basic analysis to check for formatting inconsistencies and missing values.

[1751] Data processing:

[1752] Standardize data formats and detect inconsistencies and missing values.

[1753] Output: The formatted data.

[1754] Step 3: Data Shaping

[1755] Input: Data with confirmed format.

[1756] Specific behavior:

[1757] The server converts the collected data into a unified format, removes duplicate data, and imputes missing values ​​using data cleansing algorithms.

[1758] Data processing:

[1759] Data is converted, duplicate entries are removed, and missing parts are filled in.

[1760] Output: Clean, formatted data.

[1761] Step 4: Data Summarization

[1762] Input: Clean, formatted data.

[1763] Specific behavior:

[1764] The server uses natural language processing technology to extract important information from the data and generate a summary. It identifies important items (e.g., treatment effects, side effects, etc.) and organizes them into a summary.

[1765] Data processing:

[1766] Natural language processing algorithms extract and summarize information.

[1767] Output: Summary statement.

[1768] Step 5: Data visualization

[1769] Input: Abstract text.

[1770] Specific behavior:

[1771] The server uses tools to generate graphs and charts based on the summarized data and visually display the data, using data visualization tools such as Matplotlib and D3.js.

[1772] Data processing:

[1773] Visualize data using graphs and charting tools.

[1774] Output: Graphs and charts.

[1775] Step 6: Data comparison

[1776] Input: Newly acquired and existing data.

[1777] Specific behavior:

[1778] The server compares the newly acquired data with existing data and analyzes the similarities and differences, using a comparison algorithm to identify differences and commonalities.

[1779] Data processing:

[1780] Compare data and generate analytical results.

[1781] Output: Comparison report.

[1782] Step 7: Update regulatory information

[1783] Input: Latest information obtained from national drug regulatory agency websites.

[1784] Specific behavior:

[1785] The server periodically retrieves the latest regulatory information from regulatory agencies' websites and updates the database using web scraping technology.

[1786] Data processing:

[1787] Integrate acquired regulatory information into existing databases.

[1788] Output: The updated database.

[1789] Step 8: User Interface

[1790] Input: The keyword the user wants to search for (e.g., "new treatment for lung cancer").

[1791] Specific behavior:

[1792] Users use a web interface to search for the data they need, view summaries and reports, and are provided with application templates based on the search results, which users can fill in and complete.

[1793] Data processing:

[1794] View search results and generate reports.

[1795] Output: Data search results, summary, comparison report and application template.

[1796] (Application example 1)

[1797] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1798] In today's world, the number and variety of security incidents are increasing, creating a need for rapid responses. However, the task of collecting, formatting, and summarizing the necessary information from vast amounts of data is extremely time-consuming, making it difficult to respond efficiently. It is also difficult to regularly obtain the latest information on security incidents and take appropriate measures based on that information. Therefore, an effective system that solves these issues is needed.

[1799] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1800] In this invention, the server includes means for automatically collecting clinical and non-clinical data, means for formatting the collected data and converting it into a unified format, means for extracting important information from the formatted data and generating a summary, means for visualizing the summary data in graph or chart format, means for comparing new data with existing data and providing the results in report format, means for periodically obtaining the latest pharmaceutical regulatory information and updating the database, means for a user to use an interface to search for necessary data and view summaries and reports, and means for automatically collecting, formatting, summarizing, visualizing, and comparing security incident data and periodically updating the latest security information. This enables efficient processing of massive amounts of security incident information and enables prompt and appropriate responses.

[1801] "Clinical data" refers to data collected in medical settings regarding patient diagnostic results and treatment.

[1802] "Non-clinical data" refers to data obtained outside of clinical practice, such as experimental results and observational data conducted in a laboratory or research lab.

[1803] "Collection methods" refers to the methods and techniques used to gather specific information, including the process of obtaining the required data from public databases and websites.

[1804] "Format" refers to the process of converting collected data into a unified format and, if necessary, removing duplicate data and completing missing values.

[1805] "Means for generating summaries" refers to the process of extracting important information from data and summarizing it concisely using natural language processing techniques, etc.

[1806] "Visualization means" refers to methods for displaying data in graphs or charts to make it easier to understand visually.

[1807] "Comparative means" refers to the process of matching newly acquired data with existing data and analyzing similarities and differences.

[1808] "Means of providing in report format" refers to a method of reporting comparison results and other important information in the form of documents, electronic files, etc.

[1809] "Means for regularly obtaining and updating regulatory information" refers to the process of regularly collecting the latest guidelines and regulatory information from regulatory agencies in each country and keeping the database up to date.

[1810] "User interface" refers to the screens and operating means that users use to search, view, and manipulate data using a system.

[1811] A "security incident" refers to an accident or incident related to information security, such as unauthorized access or data leakage.

[1812] "Security incident data" refers to data containing information related to security incidents, including specific examples such as attack methods, scope of impact, and countermeasures.

[1813] This invention is a system that streamlines the collection, formatting, summarization, visualization, and comparison of security incident data, as well as the updating of the latest security information. This system consists of a server, terminals, and users. The server is primarily responsible for data collection, automatic formatting, summary generation, data visualization, data comparison, and security information updates, while the terminals function as an interface with the users. Users use the system to search for and view the necessary data and obtain the information necessary for security countermeasures.

[1814] System Overview

[1815] 1. Data Collection

[1816] The server automatically collects data from public security incident databases, specialized blogs, security forums, etc. based on specific keywords (e.g., "phishing attack" or "ransomware"). Specifically, it uses various APIs and web scraping technologies to obtain the necessary data and store it in a database. The technologies used include Python, the Requests library, and BeautifulSoup.

[1817] 2. Data Formatting

[1818] The server reviews the collected data and converts it into a unified format. To ensure data integrity, it removes duplicate data and imputes missing values. MongoDB is used for database management.

[1819] 3. Data Summary

[1820] The server uses natural language processing techniques to extract important information from the data and generate summaries. For example, it uses the NLP library NLTK to extract important keywords from each incident report and provide a summary.

[1821] 4. Data Visualization

[1822] The server uses tools to generate graphs and charts based on the summarized data to visually display the data, using Python's matplotlib library to visualize the distribution of attack techniques as pie charts and frequency of occurrence as histograms.

[1823] 5. Data Comparison

[1824] The server compares the newly acquired data with the existing data, analyzing similarities and differences. The program then presents the results of these comparisons in a report. Statistical analysis libraries such as Scipy are used to collate the old and new data.

[1825] 6. Regulatory Updates

[1826] The server periodically retrieves the latest regulatory information from the websites of security regulatory agencies in each country and updates the database using web scraping technology and APIs.

[1827] 7. User Interface

[1828] Users use a smartphone or a head-mounted display (HMD) to search for the required data and view summaries and reports. The user interface is built using a web browser-based front-end framework (e.g., React).

[1829] Program examples and prompts

[1830] A specific example would be a scenario where a user uses a smartphone or HMD to search for data on the latest phishing attack trends, view the resulting summaries, and refine their company's countermeasures.

[1831] Example prompt sentence:

[1832] "Build a system that collects information on the latest phishing attack trends and provides a formatted summary of the data. Use Python to collect, format, summarize, visualize, compare, and update the latest trend information. Specifically, demonstrate a process for collecting information on phishing attacks from security blogs and forums and generating a summary and comparison report."

[1833] In this way, the present invention functions as a system that efficiently processes security incident data and enables a rapid response.

[1834] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1835] Step 1:

[1836] The server automatically collects data from public security incident databases, specialized blogs, and security forums based on specific security keywords (e.g., "phishing attack" or "ransomware"), using Python and the Requests library to send web requests and BeautifulSoup to parse HTML pages. The input data is the URL of the collected web page, and the output data is the parsed security-related information.

[1837] Step 2:

[1838] The server formats the collected data and converts it into a unified format. Specifically, it removes duplicate entries from the collected data and fills in missing values ​​with reliable default values. MongoDB is used as the database to maintain data integrity. The input data is the collected raw data, and the output data is formatted data in a unified format.

[1839] Step 3:

[1840] The server uses natural language processing (NLP) technology to extract important information from the formatted data and generate a summary. It uses the NLTK library to extract key keywords from the incident report and generate a summary. The input data is the formatted data, and the output data is the summarized information.

[1841] Step 4:

[1842] The server generates graphs and charts based on the summarized data. It uses Python's matplotlib library to visually display the data. Specifically, it displays the distribution of attack techniques as a pie chart and visualizes their occurrence frequency as a histogram. The input data is the summarized data, and the output data is the generated graphs and charts.

[1843] Step 5:

[1844] The server compares the newly acquired data with the existing data and analyzes the similarities and differences. This process uses statistical analysis libraries such as Scipy to collate and analyze the data, and provides the results in the form of a report. The input data is the old and new data, and the output data is a report of the comparison results.

[1845] Step 6:

[1846] The server periodically retrieves the latest regulatory information from the websites of security regulatory agencies in each country and updates the database. Using APIs and web scraping technology, the server collects the latest information and stores it in MongoDB. The input data is regulatory agency website information, and the output data is the updated database information.

[1847] Step 7:

[1848] Users use a smartphone or head-mounted display (HMD) to search for the data they need and view summaries and reports. The user interface is built using a front-end framework such as React to allow users to easily manipulate the data. The input data are search keywords from the user, and the output data are search results and summary reports.

[1849] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1850] The present invention provides a system that streamlines data collection, formatting, summarization, visualization, comparison, and updating of regulatory information in the pharmaceutical application process, and also includes an emotion engine that recognizes user emotions and adjusts interface display and data search results based on the emotions. The system has the following components and functions:

[1851] System Overview

[1852] The system consists of a server, a terminal, and a user. The server is mainly responsible for data collection, automatic formatting, summary generation, data visualization, data comparison, regulatory information updates, and user emotion recognition. The terminal functions as the interface with the user, allowing the user to search and view data through the system and create documents required for drug applications.

[1853] Program processing

[1854] 1. Data Collection

[1855] Subject: Server

[1856] The server uses pre-configured API keys and web scraping techniques to collect data from scientific papers, clinical trial databases, and drug regulatory agency websites. Specifically, the server uses the PubMed API to retrieve paper titles and abstracts related to "cancer treatment" and stores them in the database in JSON format. It also uses web scraping technology to crawl the FDA's latest guidelines page, retrieves and stores the content in text format.

[1857] 2. Data Formatting

[1858] Subject: Server

[1859] The server analyzes the collected data and unifies different data formats. Specifically, it parses JSON-formatted data, removes duplicate data, and fills in missing values ​​with appropriate default values. For example, in the case of paper data, if multiple papers with the same title are obtained, they are merged into one, and if information such as author names or publication dates is missing, it is filled in from reliable external information.

[1860] 3. Data Summary

[1861] Subject: Server

[1862] The server uses natural language processing (NLP) technology to extract important information from the collected data and generate summaries. Specifically, it uses an NLP library (e.g., NLTK or SpaCy) to analyze the main keywords in the paper, as well as descriptions of therapeutic effects and side effects, and generates summaries based on this. The generated summaries are then stored in a database.

[1863] 4. Data Visualization

[1864] Subject: Server

[1865] The server uses tools to generate graphs and charts based on the summarized data for visualization. For example, Matplotlib is used to generate line graphs showing the effectiveness of each treatment and bar graphs showing the frequency of side effects. The generated graphs and charts are converted into a format that can be viewed by the user interface and displayed on a dashboard in real time.

[1866] 5. Data Comparison

[1867] Subject: Server

[1868] The server compares the newly acquired data with existing data and analyzes differences and commonalities. Specifically, it cross-references data on existing cancer treatments with data on new treatments, comparing them in terms of treatment effectiveness, side effects, etc. The comparison results are generated in report format, making them easy for users to understand.

[1869] 6. Regulatory Updates

[1870] Subject: Server

[1871] The server periodically retrieves the latest information from the websites of drug regulatory agencies in each country and updates the database. Specifically, it uses a web scraping tool to retrieve the latest guidelines and regulatory changes from the FDA and EMA, stores them in the database, and sends notifications to users when changes are confirmed.

[1872] 7. Data Search and Browsing

[1873] Subject: User

[1874] Users use the device's web interface to search for and retrieve the data they need. Specifically, they enter keywords such as "new treatments for lung cancer" into the search screen and click the search button to send a request to the server. The server then retrieves relevant summaries and comparison reports from the database and displays them on the user's device.

[1875] 8. Emotion Recognition and Interface Adjustment

[1876] Subject: Server

[1877] The server uses an emotion engine to recognize the user's emotions, for example, by analyzing the user's facial expressions and tone of voice, and then adjusts the priority of data search results and optimizes the interface display.

[1878] Examples:

[1879] If the server detects that a user is feeling stressed while completing a regulatory application, it will simplify the interface and adjust the display to make important information easier to access. If there are too many search results, it will adjust the ranking to prioritize the most relevant data.

[1880] 9. Preparation of pharmaceutical application documents

[1881] Subject: User

[1882] Users create regulatory application documents based on the displayed summaries and comparison reports. Specifically, they download templates provided by the system and create documents by entering the displayed information. They can also submit the created documents for review by experts, who can make revisions based on the feedback.

[1883] In this way, the present invention, equipped with an emotion engine, is a system that not only efficiently collects, formats, summarizes, visualizes, and compares the various data required for the pharmaceutical application process, and updates the latest regulatory information, but also further supports the application process and improves its accuracy and speed by optimizing the interface in consideration of the user's emotions.

[1884] The processing flow will be explained below.

[1885] Step 1: Data collection

[1886] Subject: Server

[1887] The server uses pre-configured API keys and web scraping techniques to collect data from scientific papers, clinical trial databases, and drug regulatory agency websites. Specifically, the server uses the PubMed API to retrieve paper titles and abstracts related to "cancer treatment." The retrieved data is stored in a database in JSON format. It also uses web scraping technology to crawl the FDA's latest guidelines page, retrieve their contents in text format, and store them.

[1888] Step 2: Data Shaping

[1889] Subject: Server

[1890] The server analyzes the collected data and unifies different data formats. Specifically, it parses JSON-formatted data, removes duplicate data, and fills in missing values ​​with appropriate default values. For example, if multiple papers with the same title are retrieved, they are merged into one, and if missing information such as author names or publication dates is provided, it is filled in using reliable external information.

[1891] Step 3: Data Summarization

[1892] Subject: Server

[1893] The server uses natural language processing (NLP) technology to extract important information from the collected data and generate summaries. Specifically, it uses NLP libraries (e.g., NLTK and SpaCy) to analyze each paper's main keywords, therapeutic effects, and side effects. Based on the extracted information, it automatically generates summaries and stores them in a database.

[1894] Step 4: Data visualization

[1895] Subject: Server

[1896] The server uses tools to generate graphs and charts based on the summarized data for visualization. Specifically, Matplotlib is used to generate line graphs showing the effectiveness of each treatment and bar graphs showing the frequency of side effects. The generated graphs and charts are converted into a format that can be viewed by the user interface and displayed on a dashboard in real time.

[1897] Step 5: Data comparison

[1898] Subject: Server

[1899] The server compares the newly acquired data with existing data and analyzes differences and commonalities. Specifically, it cross-references data on existing cancer treatments with data on new treatments to evaluate differences in treatment effectiveness and side effects. The comparison results are generated in a detailed report format that users can easily understand.

[1900] Step 6: Update regulatory information

[1901] Subject: Server

[1902] The server periodically retrieves the latest information from the websites of each country's drug regulatory agency and updates the database. Specifically, it uses a web scraping tool to retrieve the latest guidelines and regulatory changes from the FDA and EMA, and stores them in the database. If any changes are confirmed, the server notifies the user.

[1903] Step 7: Emotion Recognition

[1904] Subject: Server

[1905] The server uses an emotion engine to recognize the user's emotions. Specifically, it analyzes the user's facial expressions, voice tone, input speed, etc. to evaluate the user's emotions (e.g., stress, satisfaction). The results of emotion recognition are used to adjust the interface and prioritize search results.

[1906] Step 8: Interface Adjustment

[1907] Subject: Server

[1908] The server adjusts the interface display content based on the user's emotions recognized by the emotion engine. Specifically, if the user is feeling stressed, the interface will be simplified and important information will be highlighted. The server will also prioritize the display of content that matches the user's interests.

[1909] Step 9: Search and view data

[1910] Subject: User

[1911] Users use the device's web interface to search for and retrieve the data they need. Specifically, they enter keywords such as "new treatments for lung cancer" into the search screen and click the search button to send a request to the server. The server then retrieves relevant summaries and comparison reports from the database and displays them on the user's device.

[1912] Step 10: Prepare regulatory application documents

[1913] Subject: User

[1914] Users create regulatory application documents based on the displayed summaries and comparison reports. Specifically, they download templates provided by the system and create documents by entering the displayed information. They then submit the created documents for review by experts, who provide feedback and make revisions.

[1915] In this way, the present invention, equipped with an emotion engine, is a system that not only efficiently collects, formats, summarizes, visualizes, and compares the various data required for the pharmaceutical application process, and updates the latest regulatory information, but also further supports the application process and improves its accuracy and speed by optimizing the interface in consideration of the user's emotions.

[1916] Example 2

[1917] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1918] The pharmaceutical application process requires a huge amount of data collection, complex data formatting, detailed summarization, effective data visualization, rigorous data comparison, and the latest regulatory information updates, making it extremely time-consuming and labor-intensive. Furthermore, the lack of user-friendly interface adjustments can lead to reduced work efficiency. Therefore, there is a need for a system that can efficiently complete pharmaceutical application procedures without causing stress to users.

[1919] The specific processing by the specific processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for automatically collecting clinical data and non-clinical data, means for formatting the collected data and converting it into a unified format, means for extracting important information from the formatted data and generating a summary, means for visualizing the summary data in the form of graphs or charts, means for comparing new data with existing data and providing the results in the form of a report, means for periodically obtaining the latest pharmaceutical regulatory information and updating the database, means for a user to use an interface to search for necessary data and view summaries and reports, means for analyzing user emotions using an emotion engine and adjusting the interface display and search results, and means for generating prompt sentences to input to the generative AI model. This enables efficient data collection, data formatting, summarization, visualization, comparison, and updating of the latest regulatory information in the pharmaceutical application process, and further enables significant improvement in work efficiency and user experience through interface adjustment based on user emotions.

[1920] "Data Collection Instruments" refers to processes or mechanisms for the automated collection of clinical and non-clinical data.

[1921] "Data formatting means" refers to the process or method of analyzing collected data and converting it into a unified format.

[1922] "Data summarization means" refers to a mechanism or process that extracts important information from formatted data and generates a summary.

[1923] "Data visualization methods" refers to the process of visualizing summarized data in the form of graphs or charts.

[1924] "Data comparison means" refers to a method or technology that compares new data with existing data and provides the results of the comparison in a report format.

[1925] "Regulatory information update means" refers to the process or method of periodically obtaining the latest pharmaceutical regulatory information and updating the database.

[1926] "Data retrieval means" refers to the ability for users to use the interface to retrieve required data and view summaries and reports.

[1927] "Emotion analysis means" refers to a mechanism that uses an emotion engine to analyze user emotions and adjusts interface display and search results based on the results.

[1928] "Prompt generation means" refers to the process or method for generating prompt sentences to be input into a generative AI model.

[1929] "Automatic means" refers to processes or techniques that do not require manual intervention and are operated automatically by the system.

[1930] "User interface" refers to the screens and methods by which a user interacts with a system and searches and views data.

[1931] The present invention is a system that streamlines data collection, formatting, summarization, visualization, comparison, and updating of regulatory information in the pharmaceutical application process. Furthermore, it aims to improve operability by incorporating an emotion engine that recognizes user emotions and adjusts interface display and data search results based on the emotions. The system of the present invention has the following specific configuration:

[1932] System Components

[1933] 1. Server

[1934] The server is the central part of the system and performs the following functions:

[1935] Data collection: Using PubMed API and web scraping techniques, we collect data from scientific articles, clinical trial databases, and drug regulatory agency websites. For example, we use PubMed API to retrieve titles, authors, and abstracts of articles related to "cancer treatment" and store them in a database in JSON format.

[1936] Data Transformation: Analyze the collected data and convert it into a unified format, specifically removing duplicate entries in the JSON data and imputing missing values ​​from external reliable data.

[1937] Data summarization: Using natural language processing (NLP) techniques, we extract important information from the collected data and create summaries. For example, we use NLP libraries (e.g., NLTK and SpaCy) to analyze the main keywords in papers and descriptions of treatment effects and side effects.

[1938] Data visualization: Visualize the data using tools that generate graphs and charts (e.g., Matplotlib) based on the summarized data. The generated graphs and charts are displayed in real time on a dashboard.

[1939] Data comparison: Compare newly acquired data with existing data and analyze differences and commonalities. Generate a report of the comparison results so that users can easily understand them.

[1940] Regulatory information updates: Regularly retrieve the latest information from the websites of national drug regulatory agencies and update the database. For example, use web scraping tools to retrieve the latest guidelines from the FDA and EMA.

[1941] Emotion recognition: An emotion engine is used to analyze emotions from a user's facial expressions and tone of voice to adjust interface display and data search result priorities.

[1942] 2. Terminal

[1943] The terminal acts as an interface with the user:

[1944] Data search and browsing: Users use a web interface to search for and retrieve the data they need. For example, they enter keywords such as "new treatments for lung cancer" into the search screen and click the search button to send a request to the server. The server retrieves relevant summaries and comparison reports from the database and displays them on the terminal.

[1945] Interface Adaptation: If the emotion engine detects stress in the user, it will automatically simplify the interface and adjust to make important information easier to access.

[1946] 3. Users

[1947] Users use the system to:

[1948] Data retrieval: Users can search for the required data through a web interface and view summaries and reports.

[1949] Creating regulatory application documents: Users create regulatory application documents based on the displayed summary and comparison reports. They download templates provided by the system and enter the required information. They submit the completed documents for expert review and make revisions based on the feedback.

[1950] Specific examples

[1951] 1. The server uses the PubMed API to search for papers on "lung cancer treatments" and retrieves their titles, authors, and abstracts. Then, it uses a web crawler to parse the FDA's "New Treatment Guidelines" page and saves the content in text format.

[1952] 2. The server eliminates duplicates from the collected data and completes missing author names, publication dates, etc. from external databases.

[1953] 3. Using natural language processing technology, generate a summary from the collected papers: "This new drug improved the survival rate of lung cancer patients by 20%, but nausea was reported as a side effect."

[1954] 4. Use Matplotlib to generate a line graph showing the effectiveness of "Lung Cancer Treatment A" and a bar graph showing the frequency of side effects. Display these graphs on a dashboard.

[1955] 5. Compare the effectiveness and side effects of "existing treatment A" and "new treatment B" and generate a comparative report showing that new treatment B is slightly less effective but has fewer side effects.

[1956] 6. Use a web crawler to regularly check for the latest EMA guidelines and store new regulatory information in a database.

[1957] 7. If users are experiencing stress while completing regulatory submissions, simplify the interface, make important information more accessible, and adjust the ranking of search results to surface more relevant information.

[1958] Example prompts to input to the generative AI model

[1959] "Collect the latest research data on lung cancer treatments, summarize and compare them, and use them to draft regulatory applications."

[1960] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1961] Step 1:

[1962] Data collection

[1963] The server collects data using the PubMed API and a web scraping tool. Specifically, it retrieves the titles, authors, and abstracts of papers related to "lung cancer treatment" from the PubMed API. It also uses the web scraping tool to crawl the FDA's latest guidelines page and saves the content in text format.

[1964] Input: Keyword "lung cancer treatment", API key, website URL

[1965] Output: Paper data (JSON format), guideline text data

[1966] Specific operation: The server collects paper data on "lung cancer treatment" from the PubMed API, and at the same time uses a web crawler to retrieve and save the contents of the FDA's latest guidelines page.

[1967] Step 2:

[1968] Data Formatting

[1969] The server parses the collected JSON-formatted data and unifies different formats, specifically removing duplicate data and completing missing values ​​from external information.

[1970] Input: Collected paper data (JSON format), guideline text data

[1971] Output: Formatted data (Uniform JSON format)

[1972] What it does: The server reads a file such as "Lung Cancer Treatments.pdf," merges duplicate entries, and completes missing author names and publication dates.

[1973] Step 3:

[1974] Data Summary

[1975] The server uses natural language processing (NLP) technology to analyze the collected data and extract important information. It uses NLP libraries (NLTK and SpaCy) to analyze the main keywords of the papers, as well as descriptions of therapeutic effects and side effects, and generates summaries.

[1976] Input: Formatted data (Uniform JSON format)

[1977] Output: Summary data (text format)

[1978] Specific operation: The server analyzes the abstract of a paper on "lung cancer treatment" using SpaCy, generates a summary, and saves it.

[1979] Step 4:

[1980] Data Visualization

[1981] The server generates graphs and charts based on the summarized data, using Matplotlib to graphically display the effectiveness of each treatment and the frequency of side effects.

[1982] Input: Summary data (text format)

[1983] Output: Graphs and charts (image format)

[1984] Specific operation: Based on the summarized data, the server uses Matplotlib to create a line graph showing the effectiveness of "Lung Cancer Treatment A" and a bar graph showing the frequency of side effects.

[1985] Step 5:

[1986] Data comparison

[1987] The server cross-references the newly acquired data with the existing data, analyzes the differences and commonalities between them, and generates a report of the comparison results.

[1988] Input: Summary data (text format), existing data

[1989] Output: Comparison report (text format)

[1990] Specific operation: The server cross-references the effectiveness and side effects of "existing treatment A" and "new treatment B" and generates a report stating that treatment B is slightly less effective but has fewer side effects.

[1991] Step 6:

[1992] Regulatory updates

[1993] The server periodically scrapes the websites of each country's drug regulatory agency to collect the latest information and update the database.

[1994] Input: Regulatory agency website URL

[1995] Output: Latest regulatory information (text format)

[1996] What it does: The server uses a web crawler to crawl the latest FDA and EMA guideline pages and stores the necessary information in a database.

[1997] Step 7:

[1998] Data Search and Viewing

[1999] Users use the device's web interface to search for and retrieve the data they need. By entering search keywords and clicking the search button, the request is sent to the server, which retrieves the relevant data and displays it on the user's device.

[2000] Input: Search keyword

[2001] Output: Relevant data (summary and comparison reports)

[2002] Specific operation: When a user enters "new treatments for lung cancer" into the search box and performs a search, the server displays relevant summaries and reports on the device.

[2003] Step 8:

[2004] Emotion Recognition and Interface Adjustment

[2005] The server uses an emotion engine to analyze the user's emotions, including facial expressions and tone of voice, and adjusts the interface display and search result priorities based on the results.

[2006] Input: User's facial expression data, voice data

[2007] Output: Tailored interface, prioritized search results

[2008] What it does: The server collects data from the user's webcam and microphone and simplifies the interface if the user is stressed.

[2009] Step 9:

[2010] Preparation of pharmaceutical application documents

[2011] Users create regulatory application documents based on the displayed summary and comparison reports. They download templates provided by the system, enter the required information, and submit the completed documents for expert review, where they can make revisions based on the feedback.

[2012] Inputs: Summary data, comparison reports, templates

[2013] Output: Pharmaceutical application documents

[2014] What it does: A user downloads a template from the system, copies information from the summary and comparison reports to complete the application, and then submits it to an expert for feedback and makes any necessary revisions.

[2015] (Application example 2)

[2016] Next, a description will be give...

Claims

1. means for automatically collecting clinical and non-clinical data; A means of formatting the collected data and converting it into a unified format; A means for extracting important information from the formatted data and generating a summary; A means of visualizing the summarized data in graphs and charts, a means for comparing the new data with existing data and providing the results in a report format; A means of regularly obtaining the latest regulatory information and updating the database; A means for users to use the interface to search for required data and view summaries and reports; A system including:

2. The system of claim 1 , further comprising means for using web scraping techniques in collecting the existing data.

3. The system according to claim 1, further comprising means for removing duplicate data and imputing missing values ​​in shaping the collected data.

4. 10. The system of claim 1, further comprising means for extracting meaningful information from the collected data using natural language processing techniques.

5. 10. The system according to claim 1, further comprising means for providing a template for creating a regulatory application document based on the displayed summary and comparison report.

Images