System

Abstract

A system is provided.SOLUTION: A system comprising: means for acquiring brainwave information from a brainwave measuring device; means for preprocessing the acquired brainwave information to convert the acquired brainwave information into an analyzable format; means for inputting the preprocessed brainwave information to a deep learning model to analyze emotions and thoughts; means for converting an analysis result into natural languages using a generative AI model; and means for displaying the converted natural languages.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] In medical and nursing care settings, communication with care recipients and medical professionals who have difficulty expressing their intentions and feelings appropriately through words is a major challenge. This lack of communication often hinders the provision of appropriate care and treatment. Furthermore, conventional methods have limited means for grasping the real-time emotions and thoughts of care recipients and medical professionals, which can lead to delayed responses by caregivers and medical professionals. This invention aims to solve these problems. [Means for solving the problem]

[0005] To solve the above-mentioned problems, the present invention provides the following means. First, it includes a means for acquiring electroencephalogram data from a care recipient or medical recipient using an electroencephalogram (EEG) measuring device. Next, it incorporates a means for preprocessing the acquired EEG data and converting it into an analyzable data format. This preprocessing includes a means for applying a noise reduction filter. Furthermore, it provides a means for inputting the preprocessed EEG data into a deep learning model to analyze emotions and thoughts. The analysis results are converted into natural language using a generative AI model. These results are visually displayed using a display means, allowing caregivers and medical professionals to grasp the emotions and thoughts of the care recipient or medical recipient in real time. Furthermore, it also includes a means for transmitting the collected EEG data using a secure communication protocol. This allows for the construction of a system for safely and efficiently analyzing and displaying data, making it possible to support communication in medical and nursing care settings.

[0006] An "EEG measuring device" is a device that detects the brain waves of care recipients or medical recipients and collects that data.

[0007] "Electroencephalogram data" refers to data that indicates electrical activity of the brain, obtained by an electroencephalogram measuring device.

[0008] "Preprocessing" refers to performing processes such as noise removal and normalization to convert raw EEG data into an analyzable data format.

[0009] A "deep learning model" is a machine learning technology that learns patterns from large amounts of data and analyzes emotions and thoughts.

[0010] A "generative AI model" is an artificial intelligence model that generates natural language based on analysis results.

[0011] "Natural language" refers to the words and sentences that people use on a daily basis.

[0012] The "display means" is a device that visually displays the analysis results and the generated natural language.

[0013] A "secure communication protocol" is a secure communication method that ensures confidentiality and integrity when transmitting and receiving data. [Brief explanation of the drawings]

[0014] [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

[0015] 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.

[0016] First, the terms used in the following description will be explained.

[0017] 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).

[0018] 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.

[0019] 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.

[0020] 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.

[0021] 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."

[0022] [First embodiment]

[0023] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0024] 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.

[0025] 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).

[0026] 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.

[0027] 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.

[0028] 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.

[0029] 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.

[0030] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0031] 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.

[0032] 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.

[0033] 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.

[0034] 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."

[0035] This invention is a system designed to support communication in medical and care settings, and analyzes the electroencephalogram data of care recipients and medical care recipients, converting their emotions and thoughts into natural language and displaying them. Below, we will explain in detail the role that the program in this system plays and the processing it performs.

[0036] First, the terminal collects EEG data in real time from an EEG measurement device ("EEG measuring device") worn on the head of the care recipient or medical care recipient. The collected data is temporarily stored in a buffer and then transmitted to a server using a secure communication protocol. The secure communication protocol guarantees the confidentiality and integrity of the data.

[0037] On the server side, the received EEG data is first preprocessed. Noise reduction filters are applied to clean up the data and convert it into a format suitable for model input. The preprocessed data is then input into a deep learning model. This model learns from large amounts of data and analyzes the emotions and thoughts of care recipients or medical care recipients with high accuracy.

[0038] The analysis results obtained by the deep learning model are input into a generative AI model, which converts emotions and thoughts into natural language. For example, if the emotion "anxiety" is analyzed, the generative AI model converts this into a natural language sentence such as "I am anxious. I am worried about something." The converted natural language message is then sent back to the device using a secure communication protocol.

[0039] Finally, the device displays the received message on a monitor, allowing caregivers and medical professionals to visually understand the emotions and thoughts of the person receiving care or medical care, and respond appropriately.

[0040] Specific examples

[0041] For example, suppose a care recipient is having difficulty communicating. The device collects the care recipient's brain wave data in real time and sends it to a server. The server performs preprocessing and uses a deep learning model to analyze the emotion "anxiety." The generative AI model converts this analysis result into natural language, saying "I'm anxious, I'm worried about something," and sends this message to the device. The device's monitor displays "I'm anxious, I'm worried about something," and the caregiver immediately understands that the care recipient is feeling anxious.

[0042] This system will significantly improve communication between care recipients and medical professionals in medical and nursing care settings, allowing for the provision of appropriate care and treatment.

[0043] The processing flow will be explained below.

[0044] Step 1:

[0045] The terminal collects real-time brain wave data from an EEG measurement device worn on the head of the care recipient or medical care recipient, samples the electrical signals from the device at regular intervals, and temporarily stores the data in a buffer.

[0046] Step 2:

[0047] The device divides the collected EEG data into packets and sends them to the server using a secure communication protocol (e.g., HTTPS). When a certain amount of data has accumulated in the buffer, it is sent sequentially.

[0048] Step 3:

[0049] The server reconstructs the received packets to obtain the raw EEG data, which is then stored in memory and sent to the next processing step.

[0050] Step 4:

[0051] The server applies noise reduction filters to clean up the EEG data, using band-pass filters and artifact removal algorithms to remove unwanted noise and outliers.

[0052] Step 5:

[0053] The server standardizes the clean EEG data and converts it into a format suitable for the analysis model, for example, by performing data normalization and reshaping operations to create a dataset for input to the deep learning model.

[0054] Step 6:

[0055] The server inputs the preprocessed data into a deep learning model to analyze emotions and thoughts. It uses techniques such as convolutional neural networks (CNNs) and recurrent neural networks (RNNs) to identify emotion categories. The analysis results are output along with a probability score.

[0056] Step 7:

[0057] The server creates a data structure to be input into a generative AI model based on the output of the deep learning model. For example, if the emotion category "anxiety" is identified, that information is used as input to a generative AI model (e.g., GPT-4 (registered trademark)).

[0058] Step 8:

[0059] The server uses a generative AI model to convert emotion categories into natural language, and the generated text is output as a natural language sentence such as "I'm anxious, I'm worried about something."

[0060] Step 9:

[0061] The server then divides the generated text message into packets and sends them to the device using a secure communication protocol, again using a protocol such as HTTPS for secure communication.

[0062] Step 10:

[0063] The terminal reconstructs the received message packets to obtain the generated natural language text, which is stored in memory.

[0064] Step 11:

[0065] The device displays the received text messages on a monitor, allowing caregivers and medical professionals to visually check the messages and understand the emotions and thoughts of the person receiving care or medical care in real time.

[0066] This series of processing steps realizes a system that analyzes the brain wave data of care recipients and medical care recipients and displays their emotions and thoughts in natural language.

[0067] Example 1

[0068] 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."

[0069] Conventional EEG data analysis systems require a lot of manual operations and time for collecting, preprocessing, analyzing, and presenting the results, making it difficult to provide fast and accurate communication support. Furthermore, they often lack important processing functions such as secure communication, noise reduction, and natural language generation, resulting in problems with data integrity and reliability. This limits their practical use in medical and nursing care settings.

[0070] 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.

[0071] In this invention, the server includes means for collecting EEG data acquired by an EEG measuring device, means for transmitting the collected EEG data to the server using a secure communication protocol, means for preprocessing the transmitted EEG data and converting it into an analyzable data format, means for applying a noise reduction filter in the preprocessing stage, means for inputting the preprocessed EEG data into a deep learning model to analyze emotions and thoughts, means for converting the analysis results into natural language using a generative AI model, means for transmitting the converted natural language message to a terminal using a secure communication protocol, and means for displaying the converted natural language. This makes it possible to analyze EEG data quickly and accurately and visually present the results in natural language.

[0072] An "EEG measuring device" is a device that measures and collects the brain wave data of care recipients and medical care recipients in real time.

[0073] A "secure communications protocol" is an encrypted communications method that ensures confidentiality and integrity during data transmission.

[0074] "Preprocessing" refers to the process of removing noise from raw EEG data and converting the data format to make it analyzable.

[0075] A "noise removal filter" is an algorithm or processing means for removing unnecessary noise contained in electroencephalogram data.

[0076] A "deep learning model" is an artificial intelligence model that learns using large amounts of data and analyzes emotions and thoughts from input data with high accuracy.

[0077] A "generative AI model" is an artificial intelligence model that converts analyzed emotions and thoughts into natural language.

[0078] "Natural language" refers to the language used by humans on a daily basis, and refers to the textual output generated by generative AI models.

[0079] "Terminal" means an electronic device that collects data from the EEG measurement device and receives and displays messages from the server.

[0080] "Display" refers to presenting the generated natural language message on a monitor or screen.

[0081] This invention is a system designed to support communication in medical and care settings, analyzing the electroencephalogram (EEG) data of care recipients and converting their emotions and thoughts into natural language for display. To implement this invention, an EEG measurement device, a terminal, a server, a deep learning model, and a generative AI model are required.

[0082] First, the terminal collects brain wave data in real time via an electroencephalogram (EEG) device worn on the head of the care recipient or medical care recipient. An example of an EEG device is an EEG device. The EEG device measures the brain waves of the care recipient or medical care recipient through electrodes and transmits them to the terminal as a digital signal.

[0083] The collected EEG data is then temporarily stored in a buffer on the device and then transmitted to the server using a secure communication protocol (e.g., HTTPS or SSL / TLS), which ensures data confidentiality and integrity.

[0084] On the server side, the received EEG data is first preprocessed. This preprocessing involves applying a noise reduction filter, specifically an algorithm such as a Butterworth filter. The noise reduction filter removes unnecessary noise from the EEG data and converts it into a clean data format.

[0085] The preprocessed data is then input into a deep learning model. The deep learning model is built using machine learning libraries such as TENSORFLOW (registered trademark) and PyTorch, and analyzes the emotions and thoughts of care recipients and medical care recipients with high accuracy from large amounts of training data. The analysis results in labels for emotions and thoughts such as "anxiety," "joy," and "sadness."

[0086] The analysis results are then fed into a generative AI model (e.g., GPT-3®), which converts emotions and thoughts into natural language. Specifically, the generative AI model is fed the following prompt:

[0087] Analyze the care recipient's EEG data. Generate a message to display if "anxiety" is detected.

[0088] Based on this prompt, the generative AI model generates a natural language message such as "I'm anxious, something is bothering me."

[0089] The generated natural language message is then sent to the device using a secure communication protocol. The device then displays the received message on a monitor, allowing caregivers and medical professionals to visually understand the emotions and thoughts of the person receiving care or medical care, enabling them to respond quickly and appropriately.

[0090] For example, suppose a care recipient is having difficulty communicating. In this case, the device collects the care recipient's brain wave data in real time and sends it to the server. The server performs preprocessing and uses a deep learning model to analyze the emotion "anxiety." The generative AI model converts this analysis result into natural language, saying "I'm anxious, I'm worried about something," and sends this message to the device. The device's monitor displays "I'm anxious, I'm worried about something," allowing the caregiver to immediately understand that the care recipient is feeling anxious.

[0091] This system will significantly improve communication between care recipients and medical professionals in medical and nursing care settings, allowing for the provision of appropriate care and treatment.

[0092] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0093] Step 1:

[0094] The terminal collects real-time brain wave data from the EEG measurement device.

[0095] Input: EEG data of care recipients and medical care recipients

[0096] How it works: The device connects to the EEG device via wireless communication such as Bluetooth or Wi-Fi, and periodically acquires EEG data. The EEG device measures the brainwaves through electrodes and transmits them to the device as digital signals.

[0097] Output: Collected EEG data

[0098] Step 2:

[0099] The terminal temporarily stores the collected brain wave data in a buffer.

[0100] Input: Collected EEG data

[0101] What it does: Stores data in temporary storage on the device and prepares it for transmission.

[0102] Output: Buffered EEG data

[0103] Step 3:

[0104] The device transmits the brainwave data to the server using a secure communication protocol.

[0105] Input: Buffered EEG data

[0106] How it works: The device sends data to the server using an encrypted channel (e.g. HTTPS or SSL / TLS) and waits for an acknowledgment to confirm the communication was successful.

[0107] Output: EEG data sent to the server

[0108] Step 4:

[0109] The server performs preprocessing on the received EEG data.

[0110] Input: Transmitted EEG data

[0111] Behavior: The server first checks the integrity of the received data and makes sure there is no missing data.

[0112] Output: Data ready for preprocessing

[0113] Step 5:

[0114] The server applies a noise reduction filter as part of the preprocessing.

[0115] Input: Data ready for preprocessing

[0116] How it works: The server uses algorithms such as Butterworth filters to remove noise.

[0117] Output: Clean EEG data

[0118] Step 6:

[0119] The server inputs the pre-processed, clean EEG data into a deep learning model to analyze emotions and thoughts.

[0120] Input: Clean EEG data

[0121] How it works: The server runs deep learning models using libraries such as TensorFlow and PyTorch. The models use pre-trained parameters to analyze input data and output emotion and thought labels.

[0122] Output: Emotion or thought analysis (e.g., "anxiety")

[0123] Step 7:

[0124] The server inputs the analysis results into a generative AI model, converting emotions and thoughts into natural language.

[0125] Input: Emotion and thought analysis results

[0126] How it works: A generative AI model (e.g., GPT-3) receives the analysis results as a prompt and generates a natural language message accordingly. A specific prompt might be something like, "The care recipient's emotion has been analyzed as 'anxiety.' Please generate a sentence that describes this emotion."

[0127] Output: Generated natural language message (e.g., "I'm anxious, something is bothering me")

[0128] Step 8:

[0129] The server transmits the generated natural language message to the terminal using a secure communication protocol.

[0130] Input: Generated natural language message

[0131] How it works: The server sends the generated message back to the device over an encrypted channel.

[0132] Output: A natural language message sent to the terminal

[0133] Step 9:

[0134] The terminal displays the received message on the monitor.

[0135] Input: A natural language message sent to the terminal

[0136] BEHAVIOR: The device receives the message, formats it appropriately as text, and displays it on the screen.

[0137] Output: Natural language message displayed on the monitor

[0138] Through this series of steps, the emotions and thoughts of care recipients and medical personnel in medical and nursing care settings can be analyzed quickly and accurately, and presented visually as natural language messages, enabling appropriate and prompt responses.

[0139] (Application example 1)

[0140] 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."

[0141] With conventional methods, it is difficult to accurately grasp the emotions and thoughts of care recipients in nursing and medical settings, which often results in delays in appropriate responses and care. Furthermore, it is difficult to understand customer emotions in real time in physical stores, making it difficult to provide personalized services to improve customer satisfaction. To solve this problem, it is necessary to analyze the brainwave data of care recipients, medical recipients, or customers, understand their emotions and thoughts in real time, and provide appropriate responses.

[0142] 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.

[0143] In this invention, the server includes means for collecting EEG data acquired by an EEG measuring device, means for preprocessing the collected EEG data and converting it into an analyzable data format, means for inputting the preprocessed EEG data into a deep learning model to analyze emotions and thoughts, means for converting the analysis results into natural language using a generative AI model, means for displaying the converted natural language, means for displaying on smart glasses, means for collecting and analyzing EEG data of customers in a physical store, and means for providing the analysis results to store staff in the physical store in real time. This not only enables the emotions and thoughts of care recipients and medical care recipients to be grasped in real time and appropriate care and responses to be provided, but also enables understanding of customer emotions in physical stores and personalized customer service.

[0144] An "EEG measuring device" is a device worn on the head to collect electroencephalogram data.

[0145] "Collecting" is the process of gathering and storing data obtained from an EEG measuring device.

[0146] "Preprocessing" is the process of converting collected raw data into an analyzable format and removing noise and unnecessary information.

[0147] A "deep learning model" is a type of artificial intelligence that learns from large amounts of data and extracts specific patterns and features from the input data.

[0148] "Emotion and thought analysis" is the process of using deep learning models to infer a subject's emotional and thought state from EEG data.

[0149] A "generative AI model" is an artificial intelligence model that converts analyzed emotional and thought data into natural language.

[0150] "Conversion to natural language" is the process of converting numerical data and analytical results into textual language that is easy for humans to understand.

[0151] "Display" refers to the act of showing the converted natural language message on an output device such as a display.

[0152] "Smart glasses" are glasses-type devices that have a display function and can display information in real time.

[0153] A "physical store" is a store that sells or provides services at a physical location.

[0154] "Providing in real time" means instantly obtaining and displaying the latest data and information on the subject and providing analysis results.

[0155] The system for implementing this invention is mainly composed of smart glasses, a server, and a secure communication infrastructure. Specific examples of the main hardware and software of this system include an EEG measurement device, deep learning libraries (TensorFlow, PyTorch), secure communication protocols (HTTPS, MQTT), a generative AI model (OpenAI® GPT), and data preprocessing libraries (NumPy, SciPy).

[0156] Explanation of program processing

[0157] 1. Acquisition of EEG data

[0158] The EEG measurement device attached to the smart glasses collects the customer's brainwave data in real time, which is first temporarily stored in a buffer.

[0159] 2. Data transmission

[0160] The buffered EEG data is then sent to a server using a secure communication protocol (HTTPS), ensuring confidentiality and integrity of the data.

[0161] 3. Data Preprocessing

[0162] On the server side, a noise reduction filter (SciPy) is applied to the received EEG data to clean it up, and then the data is converted into a format suitable for the deep learning model (TensorFlow).

[0163] 4. Emotion or thought analysis

[0164] The cleaned data is then fed into a deep learning model (TensorFlow) to analyze customer emotions and thoughts. This model uses previous learnings from large amounts of data to predict emotions and thoughts with high accuracy.

[0165] 5. Natural Language Translation

[0166] The analyzed data is input into a generative AI model (OpenAI GPT) and converted into natural language. For example, if the emotion "excited" is analyzed, the generative AI model generates a text message such as "The customer is excited. They seem to have found something special."

[0167] 6. Provision and display of results

[0168] The generated text message is then sent to the smart glasses using a secure communication protocol (HTTPS), which then displays the message on the smart glasses' display, allowing store staff to understand the customer's emotions and thoughts in real time and take appropriate action.

[0169] Specific examples

[0170] Consider a scenario in which a customer enters a brick-and-mortar store and puts on smart glasses. While the customer is browsing products, an EEG measurement device captures brainwave data and sends it to a server in real time. The data is preprocessed on the server, and a deep learning model is used to analyze the data and determine that the customer is "excited." A generative AI model then generates a natural language message, such as "The customer is excited and seems to have found something special," which is displayed on the smart glasses' display. Based on this information, a store clerk can ask the customer, "Did you find anything interesting?", providing more personalized customer service.

[0171] Example prompt sentence:

[0172] Customer sentiment analysis results: Excitement

[0173] Generate natural language message: Your customer is excited, they've found something special.

[0174] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0175] Step 1:

[0176] The EEG measurement device attached to the smart glasses collects the customer's EEG data in real time. The collected EEG data is temporarily stored in a buffer inside the smart glasses. Specifically, the sensor inside the smart glasses measures bioelectric potentials, converts the data into digital format, and stores it.

[0177] Step 2:

[0178] The smart glasses transmit the buffered EEG data to the server using a secure communication protocol (HTTPS), in an encrypted format to ensure confidentiality and integrity of the data. The input EEG data is transmitted and received by the server, and the data is then transferred.

[0179] Step 3:

[0180] The server performs data preprocessing on the received EEG data. Specifically, it applies a noise reduction filter (SciPy) to remove unnecessary noise and outliers. At this stage, the input is encrypted EEG data, and the output is cleaned EEG data.

[0181] Step 4:

[0182] The pre-processed clean EEG data is input into a deep learning model (TensorFlow). The deep learning model analyzes the customer's emotions and thoughts from the EEG data based on pre-learned patterns. The input is the clean EEG data, and the output is the analysis result of emotions or thoughts. Specifically, pattern recognition based on numerical data is performed.

[0183] Step 5:

[0184] The analysis results are input into a generative AI model (OpenAI GPT) and converted into natural language. For example, if the emotion "excited" is analyzed, the generative AI model generates a natural language message such as "The customer is excited and seems to have found something special." The input is the analysis result of the emotion or thought, and the output is a natural language text message.

[0185] Step 6:

[0186] The generated natural language message is then sent to the smart glasses using a secure communication protocol (HTTPS). This message is then displayed on the smart glasses' display, allowing the store clerk to understand the customer's emotions and thoughts in real time and take appropriate action. Specifically, the input is a natural language text message, and the output is a text message displayed on the smart glasses' display.

[0187] 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.

[0188] This invention is a system designed to support communication in medical and care settings, analyzing the electroencephalogram data of care recipients and converting their emotions and thoughts into natural language for display. This system also combines an emotion engine to improve the accuracy of emotion recognition.

[0189] First, the terminal collects real-time EEG data from an EEG measuring device ("EEG measuring device") worn on the head of the care recipient or medical care recipient. The collected data is temporarily stored in a buffer and then transmitted to a server using a secure communication protocol. The secure communication protocol guarantees the confidentiality and integrity of the data.

[0190] On the server side, the received EEG data is first preprocessed. Noise reduction filters are applied to clean up the data and convert it into a format suitable for model input. The preprocessed data is then input into a deep learning model. This model learns from large amounts of data and analyzes the emotions and thoughts of care recipients or medical care recipients with high accuracy.

[0191] Meanwhile, the emotion engine has the ability to recognize user emotions in real time. The emotion data obtained by this emotion engine is also used as input data for deep learning models, improving the accuracy of emotion analysis. In addition, the data output by the emotion engine is also input into generative AI models, allowing for more appropriate adjustment of the content of generated natural language messages.

[0192] The analysis results of the deep learning model are input into a generative AI model, which converts emotions and thoughts into natural language. For example, if the emotion "anxiety" is analyzed, the generative AI model converts this into a natural language sentence such as "I am anxious. I am worried about something." The converted natural language message is then sent back to the device using a secure communication protocol.

[0193] Finally, the device displays the received message on a monitor, allowing caregivers and medical professionals to visually understand the emotions and thoughts of the person receiving care or medical care, and respond appropriately.

[0194] Specific examples

[0195] For example, suppose a care recipient is having difficulty communicating. The device collects the recipient's brainwave data in real time and sends it to the server. The server then performs preprocessing and analyzes the emotion of "anxiety" using a deep learning model. The analysis results are linked to the emotion engine, which then provides more detailed emotional data.

[0196] Based on the analysis results and the emotion engine data, the generative AI model converts the message into natural language: "I'm anxious, I'm worried about something." This message is sent to the device, and the message is displayed on the device's monitor. The caregiver can immediately understand that the person being cared for is feeling anxious and provide appropriate care.

[0197] This system will significantly improve communication between care recipients and those receiving medical care in medical and nursing care settings, enabling appropriate care and treatment to be provided. By combining it with an emotion engine, it will be possible to respond more accurately and appropriately.

[0198] The processing flow will be explained below.

[0199] Step 1:

[0200] The terminal collects real-time brain wave data from an EEG measurement device worn on the head of the care recipient or medical care recipient, samples the electrical signals from the device at regular intervals, and temporarily stores the data in a buffer.

[0201] Step 2:

[0202] The device divides the collected EEG data into packets and sends them to the server using a secure communication protocol (e.g., HTTPS). When a certain amount of data has accumulated in the buffer, it is sent sequentially.

[0203] Step 3:

[0204] The server reconstructs the received packets to obtain the raw EEG data, which is then stored in memory and sent to the next processing step.

[0205] Step 4:

[0206] The server applies noise reduction filters to clean up the EEG data, using band-pass filters and artifact removal algorithms to remove unwanted noise and outliers.

[0207] Step 5:

[0208] The server standardizes the clean EEG data and converts it into a format suitable for the analysis model, for example, by performing data normalization and reshaping operations to create a dataset for input to the deep learning model.

[0209] Step 6:

[0210] The server inputs the preprocessed data into a deep learning model to analyze emotions and thoughts. It uses techniques such as convolutional neural networks (CNNs) and recurrent neural networks (RNNs) to identify emotion categories. The analysis results are output along with a probability score.

[0211] Step 7:

[0212] The server uses an emotion engine to recognize the user's emotions in real time, analyzing facial expressions, voice, and body movements to generate emotion data.

[0213] Step 8:

[0214] The server uses the emotion data obtained by the emotion engine in conjunction with a deep learning model, which improves the accuracy of emotion analysis.

[0215] Step 9:

[0216] The server creates a data structure to be input into the generative AI model based on the analysis results of the deep learning model and the emotion data obtained from the emotion engine. For example, if the emotion category "anxiety" and detailed emotion data are identified, that information is used as input to the generative AI model.

[0217] Step 10:

[0218] The server uses a generative AI model to convert emotion categories and detailed emotion data into natural language, and the generated text is output as a natural language sentence such as "I'm anxious, I'm worried about something."

[0219] Step 11:

[0220] The server then divides the generated text message into packets and sends them to the device using a secure communication protocol, again using a protocol such as HTTPS for secure communication.

[0221] Step 12:

[0222] The terminal reconstructs the received message packets to obtain the generated natural language text, which is stored in memory.

[0223] Step 13:

[0224] The device displays the received text messages on a monitor, allowing caregivers and medical professionals to visually check the messages and understand the emotions and thoughts of the person receiving care or medical care in real time.

[0225] This series of processing steps will create a system that analyzes the brainwave data of care recipients and medical care recipients, and displays their emotions and thoughts in natural language. By combining it with an emotion engine, it will be possible to respond more accurately and appropriately.

[0226] Example 2

[0227] 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."

[0228] With conventional technology, it has been difficult to accurately grasp the emotions and thoughts of care recipients and medical care recipients. Especially in situations where communication is difficult, the lack of information necessary to provide appropriate care and treatment reduces the quality of communication, negatively impacting the patient's experience and treatment outcomes. Another problem is that security and data accuracy are not adequately guaranteed when collecting and analyzing EEG data.

[0229] The identification process by the identification 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 preprocessing transmitted electroencephalogram data and converting it into an analyzable data format, means for inputting the preprocessed electroencephalogram data into a deep learning model to analyze emotions and thoughts, and means for complementing the analysis results of the deep learning model using an emotion engine. This makes it possible to safely and accurately analyze the electroencephalogram data acquired by the electroencephalogram measurement device via a secure communication protocol, and ultimately convert the intended emotions and thoughts into natural language with high accuracy and display them.

[0230] An "EEG measuring device" is a device that is worn on the head of a care recipient or medical care recipient and measures the bioelectric potential signals generated by the brain in real time.

[0231] A "secure communication protocol" is a communication method used to maintain the confidentiality and integrity of communication data, and includes encryption technologies such as TLS and SSL.

[0232] "Preprocessing" is the process of preparing collected raw data for analysis by performing processes such as noise removal and data format conversion.

[0233] A "deep learning model" is a machine learning algorithm that uses large amounts of data to learn and analyze complex nonlinear relationships with high precision, and includes neural networks.

[0234] The "emotion engine" is a system that recognizes the user's emotions in real time and supplements them with analytical data.

[0235] A "generative AI model" is an artificial intelligence model that generates new information based on learned data, and has the ability to generate sentences in natural language processing tasks, for example.

[0236] "Natural language" refers to a language used by humans on a daily basis, specifically spoken or written language, as opposed to programming languages ​​or codes.

[0237] A "monitor" is a device connected to a computer or other electronic device for displaying information.

[0238] This invention is a system for supporting communication in medical and nursing care settings. This system analyzes the emotions and thoughts of care recipients and medical care recipients with high accuracy, converts them into natural language, and displays them.

[0239] First, a terminal collects EEG data in real time using an electroencephalogram (EEG) measuring device (for example, a generic name "EEG measuring device") attached to the head of the care recipient or medical care recipient. The EEG device acquires biopotential signals from the brain through electrodes and transmits the data to the terminal via Bluetooth. This collected data is temporarily stored in a buffer within the terminal.

[0240] The device then transmits the collected EEG data to a server using a secure communication protocol such as TLS, which ensures data confidentiality and integrity.

[0241] The EEG data received by the server is first preprocessed by applying a noise reduction filter. For example, a band-pass filter is used to remove noise and extract only the components in the desired frequency range (4 Hz to 40 Hz). This preprocessed data is then input into a deep learning model (e.g., LSTM or CNN).

[0242] A deep learning model analyzes emotions and thoughts from preprocessed EEG data. The model learns from large amounts of past data and can identify the emotions and thoughts of care recipients and medical care recipients with high accuracy. For example, the analysis can identify the emotion "anxiety."

[0243] The server inputs the analysis results into an emotion engine (e.g., a generic name for this is an "emotion recognition system") to obtain detailed emotion data in real time. This emotion engine further improves the accuracy of the analysis results.

[0244] Next, based on the analysis results and data obtained from the emotion engine, a generative AI model (such as GPT-4) is used to convert emotions and thoughts into natural language messages. For example, based on the emotion data "anxiety," a natural language sentence such as "I am anxious, I am worried about something" is generated.

[0245] The generated natural language message is then sent from the server to the device using a secure communication protocol. The device then displays the received message on a monitor. This allows caregivers and medical professionals to instantly understand the emotions and thoughts of the person receiving care or medical care, enabling them to respond appropriately.

[0246] A specific example is shown below. If a care recipient is experiencing difficulty communicating, the device collects the care recipient's brainwave data in real time and sends it to the server. The server preprocesses the data, removing noise and converting the data format. The deep learning model then analyzes the emotion "anxiety." Based on the results of this analysis, the emotion engine provides detailed emotion data, and finally, the generative AI model generates a natural language message such as "I'm anxious. I'm worried about something." This message is sent to the device and displayed on the monitor.

[0247] An example prompt is:

[0248] "Please analyze the following EEG data and translate the emotions and thoughts of the care recipient into natural language:

[0249] Input data: {EEG data}

[0250] Additional data from the emotion engine: {emotion data}"

[0251] This will significantly improve communication between care recipients and medical professionals in medical and nursing care settings, allowing for more appropriate care and treatment to be provided.

[0252] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0253] Step 1:

[0254] The terminal collects EEG data in real time using an EEG measuring device attached to the care recipient or medical care recipient. The input is the biopotential signal from the EEG measuring device, and the output is data stored in a buffer within the terminal. The data stored in this buffer is ready to be sent to the next stage.

[0255] Step 2:

[0256] The EEG data collected by the device is sent to the server using a secure communication protocol (e.g., TLS). The input is the EEG data stored in the buffer on the device, and the output is the data sent to the server via a secure channel. Data confidentiality and integrity are guaranteed.

[0257] Step 3:

[0258] The server performs preprocessing on the received EEG data. The input is raw EEG data received through a secure communication protocol, and the output is a clean data format with noise removed. Specifically, the processing involves applying a bandpass filter to remove noise and converting the signal into frequency components. The target frequency range is generally 4Hz to 40Hz.

[0259] Step 4:

[0260] The server inputs the preprocessed data into a deep learning model. The input is preprocessed EEG data, and the output is analyzed emotion and thought data. Deep learning models (e.g., LSTM or CNN) learn from large amounts of past training data and can identify emotions and thoughts with high accuracy. This analysis outputs emotions such as "anxiety" or "relief."

[0261] Step 5:

[0262] The server inputs the analysis results into an emotion engine to obtain detailed emotion data. The input is emotion data obtained from the deep learning model, and the output is detailed emotion data complemented by the emotion engine. The emotion engine (such as EmotionAI) performs effective emotion recognition in real time, improving the accuracy of the analysis results.

[0263] Step 6:

[0264] The server uses a generative AI model (e.g., GPT-4) to generate natural language messages based on data obtained from the deep learning model and emotion engine. The input is detailed emotion data and analysis results, and the output is a sentence converted into natural language. For example, when emotion data of "anxiety" is input, a message such as "I am anxious, I am worried about something" is generated.

[0265] Step 7:

[0266] The server transmits the generated natural language message to the terminal again using a secure communication protocol. The input is the generated natural language message, and the output is the message sent to the terminal through the secure channel.

[0267] Step 8:

[0268] The terminal displays the received natural language message on the monitor. The input is the natural language message sent from the server, and the output is the text displayed on the monitor. By looking at this display, caregivers and medical professionals can instantly understand the feelings and thoughts of the person receiving care or medical care, and take appropriate action.

[0269] (Application example 2)

[0270] 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."

[0271] In manufacturing sites that use industrial robots, there is a need to improve work efficiency and ensure safety by smoothing communication between workers and robots and quickly and accurately conveying the feelings and thoughts of workers to the robots. In particular, it is challenging to convey the stress and fatigue of workers to the robots in real time, so that appropriate measures can be taken, and to notify the workers of feedback from the robots in natural language.

[0272] 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.

[0273] In this invention, the server includes means for collecting electroencephalogram data acquired by an electroencephalogram measuring device, means for preprocessing the collected electroencephalogram data and converting it into an analyzable data format, means for inputting the preprocessed electroencephalogram data into a deep learning model to analyze emotions and thoughts, means for converting the analysis results into natural language using a generative AI model, means for displaying the converted natural language, means for notifying the industrial robot of an error based on the emotion analysis results, means for collecting feedback from the industrial robot, and means for converting the collected feedback into natural language and notifying the worker. This allows the worker to communicate their emotions and state to the robot in real time, and the robot can respond appropriately accordingly, thereby improving work efficiency and ensuring safety on site.

[0274] An "EEG measuring device" is a device for acquiring electroencephalogram data, which is worn on the head.

[0275] The "collection means" refers to a mechanism or processing method for capturing data from an electroencephalogram measuring device and accumulating or temporarily storing the data.

[0276] "Preprocessing" refers to the initial stage of data processing to convert collected data into an analyzable format, and includes noise removal and data shaping.

[0277] A "deep learning model" is a machine learning model that automatically extracts features from large amounts of data and analyzes emotions, thoughts, and so on.

[0278] "Means for analyzing emotions and thoughts" refers to methods and technologies that use preprocessed data to analyze and identify the emotions, thoughts, and other information contained within it.

[0279] A "generative AI model" is an artificial intelligence model used to convert analysis results into natural language.

[0280] "Means for converting into natural language" refers to a mechanism or method for converting the analysis results of a deep learning model into language that humans can understand.

[0281] "Displaying means" refers to a device or interface for visually displaying the converted natural language.

[0282] A "secure communication protocol" is a set of communication procedures and rules that ensure data confidentiality and integrity.

[0283] A "noise reduction filter" is an algorithm or method for removing unwanted noise from collected data.

[0284] An "industrial robot" is an automated work machine used in production sites such as factories.

[0285] A "means for collecting feedback" is a method or mechanism for capturing and storing information from an industrial robot.

[0286] This invention is a system that analyzes brainwave data and converts emotions and thoughts into natural language, facilitating communication with industrial robots. The system includes a process that acquires a worker's brainwaves in real time, analyzes the data using a deep learning model, and converts emotions and thoughts into natural language. Furthermore, it can send feedback to the industrial robot based on the analysis results and communicate the robot's response to the worker.

[0287] Hardware used:

[0288] 1. Brainwave Sensor - worn on the worker's head to collect brainwave data in real time.

[0289] 2. Head-Mounted Display (HMD) - A device worn by the worker to provide visual feedback.

[0290] 3. RobotInterface - An interface device for communicating with and receiving feedback from an industrial robot.

[0291] Software used:

[0292] 1. Secure Communication Protocol - Sends EEG data to the server while maintaining data confidentiality.

[0293] 2. Deep learning model (AnalysisModel) - A model that analyzes emotions and thoughts based on preprocessed EEG data.

[0294] 3. Emotion Engine - An engine that analyzes analyzed emotional data with even greater precision.

[0295] 4. Natural Language Generation Model (NaturalLanguageGenerator) - A model for converting emotions and thoughts into natural language.

[0296] Specific processing details:

[0297] 1. EEG data collection and preprocessing:

[0298] The EEG data acquired from the EEG measurement device is transmitted via a secure communication protocol to a server, where it is preprocessed using a noise reduction filter and converted into a format suitable for analysis.

[0299] 2. Emotion and thought analysis:

[0300] The pre-processed data is then analyzed using a deep learning model, where the emotion engine analyzes the emotion data with high precision, enabling more accurate identification of emotions and thoughts.

[0301] 3. Natural Language Conversion:

[0302] Based on the analysis results, a generative AI model converts emotions and thoughts into natural language, yielding results such as "I'm feeling stressed" or "I'm tired."

[0303] 4. Notification and feedback for industrial robots:

[0304] The converted natural language messages are sent to the industrial robot through the robot interface. The robot takes appropriate action based on the information and sends feedback to the server if necessary. This feedback is also converted into natural language and communicated to the worker via a head-mounted display.

[0305] Example prompt sentence:

[0306] If a worker feels stressed: "I'm stressed. I need a break."

[0307] If the robot detects an abnormality: "An abnormality has been detected in the cooling system. Please perform maintenance."

[0308] These functions allow workers to communicate their emotions and state of mind to the robot in real time, allowing the robot to respond appropriately based on that information, improving work efficiency and ensuring safety on site.

[0309] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0310] Step 1:

[0311] EEG data collection

[0312] The terminal collects real-time brain wave data from an EEG measurement device attached to the head of the care recipient or worker. The collected data is temporarily stored in the terminal's memory.

[0313] Input: EEG data from an EEG measurement device

[0314] Output: Raw EEG data stored on the device

[0315] Specific operation: An EEG measuring device is attached to the head, and EEG data is continuously acquired.

[0316] Step 2:

[0317] Secure transmission of data

[0318] The device transmits the collected EEG data to a server using a secure communication protocol, and encryption techniques are used to ensure data confidentiality and integrity.

[0319] Input: Raw EEG data stored on the device

[0320] Output: Secure EEG data sent to a server

[0321] Specific operation: The collected EEG data is sent to a server via a secure communication protocol (encrypted communication channel).

[0322] Step 3:

[0323] Data Preprocessing

[0324] The server performs preprocessing on the EEG data it receives, applying a noise reduction filter and converting the data into an analyzable format. In preprocessing, unnecessary noise is removed and the data is shaped.

[0325] Input: Secure EEG data sent to the server

[0326] Output: Preprocessed and clean EEG data

[0327] What happens: A noise removal filter runs on the server, shaping and transforming the data.

[0328] Step 4:

[0329] Analysis using deep learning models

[0330] The pre-processed data is fed into a deep learning model to analyze emotions and thoughts. The model learns from a large amount of data and can identify emotions and thoughts with high accuracy.

[0331] Input: Preprocessed clean EEG data

[0332] Output: Analyzed emotion and thought data

[0333] Specific operation: The deep learning model works to extract features from input data and analyze emotions and thoughts with high accuracy.

[0334] Step 5:

[0335] Improved accuracy with emotion engine

[0336] The emotion engine is applied to the analysis results to improve the accuracy of the emotion data. The emotion engine has its own algorithm and provides more detailed emotion data.

[0337] Input: Analysis results from deep learning model

[0338] Output: Emotion data with improved accuracy

[0339] Specific operation: The emotion engine further analyzes the analysis results and improves the accuracy of the emotion data.

[0340] Step 6:

[0341] Natural language conversion using generative AI models

[0342] The improved emotion data is then fed into a generative AI model and converted into natural language, generating messages such as "I'm feeling stressed" or "I'm tired."

[0343] Input: Emotion data with improved accuracy

[0344] Output: Message converted into natural language

[0345] Specific operation: The generative AI model generates prompt sentences based on emotional data and outputs natural language messages.

[0346] Step 7:

[0347] Notification for industrial robots

[0348] The generated natural language message is sent to the industrial robot through the robot interface, and the robot then takes appropriate action based on the information.

[0349] Input: A message converted into natural language

[0350] Output: Feedback message sent to the industrial robot

[0351] Specific behavior: Natural language messages are sent to the robot interface, and the robot adjusts its behavior based on the information it receives.

[0352] Step 8:

[0353] Gathering feedback from industrial robots

[0354] Feedback from industrial robots is collected, converted into natural language, and communicated to workers. Feedback messages are displayed on a head-mounted display.

[0355] Input: Feedback from an industrial robot

[0356] Output: A natural language message sent to the worker

[0357] Specific operation: Feedback from the robot is sent to a server, converted into natural language through a generative AI model, and communicated to the worker via a head-mounted display.

[0358] 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.

[0359] 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.

[0360] 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.

[0361] [Second embodiment]

[0362] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0363] 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.

[0364] 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).

[0365] 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.

[0366] 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.

[0367] 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).

[0368] 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.

[0369] 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.

[0370] 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.

[0371] 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.

[0372] In the smart glasses 214, 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.

[0373] 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."

[0374] This invention is a system designed to support communication in medical and care settings, and analyzes the electroencephalogram data of care recipients and medical care recipients, converting their emotions and thoughts into natural language and displaying them. Below, we will explain in detail the role that the program in this system plays and the processing it performs.

[0375] First, the terminal collects EEG data in real time from an EEG measurement device ("EEG measuring device") worn on the head of the care recipient or medical care recipient. The collected data is temporarily stored in a buffer and then transmitted to a server using a secure communication protocol. The secure communication protocol guarantees the confidentiality and integrity of the data.

[0376] On the server side, the received EEG data is first preprocessed. Noise reduction filters are applied to clean up the data and convert it into a format suitable for model input. The preprocessed data is then input into a deep learning model. This model learns from large amounts of data and analyzes the emotions and thoughts of care recipients or medical care recipients with high accuracy.

[0377] The analysis results obtained by the deep learning model are input into a generative AI model, which converts emotions and thoughts into natural language. For example, if the emotion "anxiety" is analyzed, the generative AI model converts this into a natural language sentence such as "I am anxious. I am worried about something." The converted natural language message is then sent back to the device using a secure communication protocol.

[0378] Finally, the device displays the received message on a monitor, allowing caregivers and medical professionals to visually understand the emotions and thoughts of the person receiving care or medical care, and respond appropriately.

[0379] Specific examples

[0380] For example, suppose a care recipient is having difficulty communicating. The device collects the care recipient's brain wave data in real time and sends it to a server. The server performs preprocessing and uses a deep learning model to analyze the emotion "anxiety." The generative AI model converts this analysis result into natural language, saying "I'm anxious, I'm worried about something," and sends this message to the device. The device's monitor displays "I'm anxious, I'm worried about something," and the caregiver immediately understands that the care recipient is feeling anxious.

[0381] This system will significantly improve communication between care recipients and medical professionals in medical and nursing care settings, allowing for the provision of appropriate care and treatment.

[0382] The processing flow will be explained below.

[0383] Step 1:

[0384] The terminal collects real-time brain wave data from an EEG measurement device worn on the head of the care recipient or medical care recipient, samples the electrical signals from the device at regular intervals, and temporarily stores the data in a buffer.

[0385] Step 2:

[0386] The device divides the collected EEG data into packets and sends them to the server using a secure communication protocol (e.g., HTTPS). When a certain amount of data has accumulated in the buffer, it is sent sequentially.

[0387] Step 3:

[0388] The server reconstructs the received packets to obtain the raw EEG data, which is then stored in memory and sent to the next processing step.

[0389] Step 4:

[0390] The server applies noise reduction filters to clean up the EEG data, using band-pass filters and artifact removal algorithms to remove unwanted noise and outliers.

[0391] Step 5:

[0392] The server standardizes the clean EEG data and converts it into a format suitable for the analysis model, for example, by performing data normalization and reshaping operations to create a dataset for input to the deep learning model.

[0393] Step 6:

[0394] The server inputs the preprocessed data into a deep learning model to analyze emotions and thoughts. It uses techniques such as convolutional neural networks (CNNs) and recurrent neural networks (RNNs) to identify emotion categories. The analysis results are output along with a probability score.

[0395] Step 7:

[0396] The server creates a data structure to be input into a generative AI model based on the output of the deep learning model. For example, if the emotion category "anxiety" is identified, that information is used as input to a generative AI model (e.g., GPT-4).

[0397] Step 8:

[0398] The server uses a generative AI model to convert emotion categories into natural language, and the generated text is output as a natural language sentence such as "I'm anxious, I'm worried about something."

[0399] Step 9:

[0400] The server then divides the generated text message into packets and sends them to the device using a secure communication protocol, again using a protocol such as HTTPS for secure communication.

[0401] Step 10:

[0402] The terminal reconstructs the received message packets to obtain the generated natural language text, which is stored in memory.

[0403] Step 11:

[0404] The device displays the received text messages on a monitor, allowing caregivers and medical professionals to visually check the messages and understand the emotions and thoughts of the person receiving care or medical care in real time.

[0405] This series of processing steps realizes a system that analyzes the brain wave data of care recipients and medical care recipients and displays their emotions and thoughts in natural language.

[0406] Example 1

[0407] 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."

[0408] Conventional EEG data analysis systems require a lot of manual operations and time for collecting, preprocessing, analyzing, and presenting the results, making it difficult to provide fast and accurate communication support. Furthermore, they often lack important processing functions such as secure communication, noise reduction, and natural language generation, resulting in problems with data integrity and reliability. This limits their practical use in medical and nursing care settings.

[0409] 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.

[0410] In this invention, the server includes means for collecting EEG data acquired by an EEG measuring device, means for transmitting the collected EEG data to the server using a secure communication protocol, means for preprocessing the transmitted EEG data and converting it into an analyzable data format, means for applying a noise reduction filter in the preprocessing stage, means for inputting the preprocessed EEG data into a deep learning model to analyze emotions and thoughts, means for converting the analysis results into natural language using a generative AI model, means for transmitting the converted natural language message to a terminal using a secure communication protocol, and means for displaying the converted natural language. This makes it possible to analyze EEG data quickly and accurately and visually present the results in natural language.

[0411] An "EEG measuring device" is a device that measures and collects the brain wave data of care recipients and medical care recipients in real time.

[0412] A "secure communications protocol" is an encrypted communications method that ensures confidentiality and integrity during data transmission.

[0413] "Preprocessing" refers to the process of removing noise from raw EEG data and converting the data format to make it analyzable.

[0414] A "noise removal filter" is an algorithm or processing means for removing unnecessary noise contained in electroencephalogram data.

[0415] A "deep learning model" is an artificial intelligence model that learns using large amounts of data and analyzes emotions and thoughts from input data with high accuracy.

[0416] A "generative AI model" is an artificial intelligence model that converts analyzed emotions and thoughts into natural language.

[0417] "Natural language" refers to the language used by humans on a daily basis, and refers to the textual output generated by generative AI models.

[0418] "Terminal" means an electronic device that collects data from the EEG measurement device and receives and displays messages from the server.

[0419] "Display" refers to presenting the generated natural language message on a monitor or screen.

[0420] This invention is a system designed to support communication in medical and care settings, analyzing the electroencephalogram (EEG) data of care recipients and converting their emotions and thoughts into natural language for display. To implement this invention, an EEG measurement device, a terminal, a server, a deep learning model, and a generative AI model are required.

[0421] First, the terminal collects brain wave data in real time via an electroencephalogram (EEG) device worn on the head of the care recipient or medical care recipient. An example of an EEG device is an EEG device. The EEG device measures the brain waves of the care recipient or medical care recipient through electrodes and transmits them to the terminal as a digital signal.

[0422] The collected EEG data is then temporarily stored in a buffer on the device and then transmitted to the server using a secure communication protocol (e.g., HTTPS or SSL / TLS), which ensures data confidentiality and integrity.

[0423] On the server side, the received EEG data is first preprocessed. This preprocessing involves applying a noise reduction filter, specifically an algorithm such as a Butterworth filter. The noise reduction filter removes unnecessary noise from the EEG data and converts it into a clean data format.

[0424] The preprocessed data is then input into a deep learning model. The deep learning model is built using machine learning libraries such as TensorFlow and PyTorch, and analyzes the emotions and thoughts of care recipients and medical care recipients with high accuracy from large amounts of training data. The analysis results in labels for emotions and thoughts such as "anxiety," "joy," and "sadness."

[0425] The resulting analysis is then fed into a generative AI model (e.g., GPT-3), which converts emotions and thoughts into natural language. Specifically, the generative AI model is fed the following prompt:

[0426] Analyze the care recipient's EEG data. Generate a message to display if "anxiety" is detected.

[0427] Based on this prompt, the generative AI model generates a natural language message such as "I'm anxious, something is bothering me."

[0428] The generated natural language message is then sent to the device using a secure communication protocol. The device then displays the received message on a monitor, allowing caregivers and medical professionals to visually understand the emotions and thoughts of the person receiving care or medical care, enabling them to respond quickly and appropriately.

[0429] For example, suppose a care recipient is having difficulty communicating. In this case, the device collects the care recipient's brain wave data in real time and sends it to the server. The server performs preprocessing and uses a deep learning model to analyze the emotion "anxiety." The generative AI model converts this analysis result into natural language, saying "I'm anxious, I'm worried about something," and sends this message to the device. The device's monitor displays "I'm anxious, I'm worried about something," allowing the caregiver to immediately understand that the care recipient is feeling anxious.

[0430] This system will significantly improve communication between care recipients and medical professionals in medical and nursing care settings, allowing for the provision of appropriate care and treatment.

[0431] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0432] Step 1:

[0433] The terminal collects real-time brain wave data from the EEG measurement device.

[0434] Input: EEG data of care recipients and medical care recipients

[0435] How it works: The device connects to the EEG device via wireless communication such as Bluetooth or Wi-Fi, and periodically acquires EEG data. The EEG device measures the brainwaves through electrodes and transmits them to the device as digital signals.

[0436] Output: Collected EEG data

[0437] Step 2:

[0438] The terminal temporarily stores the collected brain wave data in a buffer.

[0439] Input: Collected EEG data

[0440] What it does: Stores data in temporary storage on the device and prepares it for transmission.

[0441] Output: Buffered EEG data

[0442] Step 3:

[0443] The device transmits the brainwave data to the server using a secure communication protocol.

[0444] Input: Buffered EEG data

[0445] How it works: The device sends data to the server using an encrypted channel (e.g. HTTPS or SSL / TLS) and waits for an acknowledgment to confirm the communication was successful.

[0446] Output: EEG data sent to the server

[0447] Step 4:

[0448] The server performs preprocessing on the received EEG data.

[0449] Input: Transmitted EEG data

[0450] Behavior: The server first checks the integrity of the received data and makes sure there is no missing data.

[0451] Output: Data ready for preprocessing

[0452] Step 5:

[0453] The server applies a noise reduction filter as part of the preprocessing.

[0454] Input: Data ready for preprocessing

[0455] How it works: The server uses algorithms such as Butterworth filters to remove noise.

[0456] Output: Clean EEG data

[0457] Step 6:

[0458] The server inputs the pre-processed, clean EEG data into a deep learning model to analyze emotions and thoughts.

[0459] Input: Clean EEG data

[0460] How it works: The server runs deep learning models using libraries such as TensorFlow and PyTorch. The models use pre-trained parameters to analyze input data and output emotion and thought labels.

[0461] Output: Emotion or thought analysis (e.g., "anxiety")

[0462] Step 7:

[0463] The server inputs the analysis results into a generative AI model, converting emotions and thoughts into natural language.

[0464] Input: Emotion and thought analysis results

[0465] How it works: A generative AI model (e.g., GPT-3) receives the analysis results as a prompt and generates a natural language message accordingly. A specific prompt might be something like, "The care recipient's emotion has been analyzed as 'anxiety.' Please generate a sentence that describes this emotion."

[0466] Output: Generated natural language message (e.g., "I'm anxious, something is bothering me")

[0467] Step 8:

[0468] The server transmits the generated natural language message to the terminal using a secure communication protocol.

[0469] Input: Generated natural language message

[0470] How it works: The server sends the generated message back to the device over an encrypted channel.

[0471] Output: A natural language message sent to the terminal

[0472] Step 9:

[0473] The terminal displays the received message on the monitor.

[0474] Input: A natural language message sent to the terminal

[0475] BEHAVIOR: The device receives the message, formats it appropriately as text, and displays it on the screen.

[0476] Output: Natural language message displayed on the monitor

[0477] Through this series of steps, the emotions and thoughts of care recipients and medical personnel in medical and nursing care settings can be analyzed quickly and accurately, and presented visually as natural language messages, enabling appropriate and prompt responses.

[0478] (Application example 1)

[0479] 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."

[0480] With conventional methods, it is difficult to accurately grasp the emotions and thoughts of care recipients in nursing and medical settings, which often results in delays in appropriate responses and care. Furthermore, it is difficult to understand customer emotions in real time in physical stores, making it difficult to provide personalized services to improve customer satisfaction. To solve this problem, it is necessary to analyze the brainwave data of care recipients, medical recipients, or customers, understand their emotions and thoughts in real time, and provide appropriate responses.

[0481] 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.

[0482] In this invention, the server includes means for collecting EEG data acquired by an EEG measuring device, means for preprocessing the collected EEG data and converting it into an analyzable data format, means for inputting the preprocessed EEG data into a deep learning model to analyze emotions and thoughts, means for converting the analysis results into natural language using a generative AI model, means for displaying the converted natural language, means for displaying on smart glasses, means for collecting and analyzing EEG data of customers in a physical store, and means for providing the analysis results to store staff in the physical store in real time. This not only enables the emotions and thoughts of care recipients and medical care recipients to be grasped in real time and appropriate care and responses to be provided, but also enables understanding of customer emotions in physical stores and personalized customer service.

[0483] An "EEG measuring device" is a device worn on the head to collect electroencephalogram data.

[0484] "Collecting" is the process of gathering and storing data obtained from an EEG measuring device.

[0485] "Preprocessing" is the process of converting collected raw data into an analyzable format and removing noise and unnecessary information.

[0486] A "deep learning model" is a type of artificial intelligence that learns from large amounts of data and extracts specific patterns and features from the input data.

[0487] "Emotion and thought analysis" is the process of using deep learning models to infer a subject's emotional and thought state from EEG data.

[0488] A "generative AI model" is an artificial intelligence model that converts analyzed emotional and thought data into natural language.

[0489] "Conversion to natural language" is the process of converting numerical data and analytical results into textual language that is easy for humans to understand.

[0490] "Display" refers to the act of showing the converted natural language message on an output device such as a display.

[0491] "Smart glasses" are glasses-type devices that have a display function and can display information in real time.

[0492] A "physical store" is a store that sells or provides services at a physical location.

[0493] "Providing in real time" means instantly obtaining and displaying the latest data and information on the subject and providing analysis results.

[0494] The system for implementing this invention is mainly composed of smart glasses, a server, and a secure communication infrastructure. Specific examples of the main hardware and software of this system include an EEG measurement device, deep learning libraries (TensorFlow, PyTorch), secure communication protocols (HTTPS, MQTT), a generative AI model (OpenAI GPT), and data preprocessing libraries (NumPy, SciPy).

[0495] Explanation of program processing

[0496] 1. Acquisition of EEG data

[0497] The EEG measurement device attached to the smart glasses collects the customer's brainwave data in real time, which is first temporarily stored in a buffer.

[0498] 2. Data transmission

[0499] The buffered EEG data is then sent to a server using a secure communication protocol (HTTPS), ensuring confidentiality and integrity of the data.

[0500] 3. Data Preprocessing

[0501] On the server side, a noise reduction filter (SciPy) is applied to the received EEG data to clean it up, and then the data is converted into a format suitable for the deep learning model (TensorFlow).

[0502] 4. Emotion or thought analysis

[0503] The cleaned data is then fed into a deep learning model (TensorFlow) to analyze customer emotions and thoughts. This model uses previous learnings from large amounts of data to predict emotions and thoughts with high accuracy.

[0504] 5. Natural Language Translation

[0505] The analyzed data is input into a generative AI model (OpenAI GPT) and converted into natural language. For example, if the emotion "excited" is analyzed, the generative AI model generates a text message such as "The customer is excited. They seem to have found something special."

[0506] 6. Provision and display of results

[0507] The generated text message is then sent to the smart glasses using a secure communication protocol (HTTPS), which then displays the message on the smart glasses' display, allowing store staff to understand the customer's emotions and thoughts in real time and take appropriate action.

[0508] Specific examples

[0509] Consider a scenario in which a customer enters a brick-and-mortar store and puts on smart glasses. While the customer is browsing products, an EEG measurement device captures brainwave data and sends it to a server in real time. The data is preprocessed on the server, and a deep learning model is used to analyze the data and determine that the customer is "excited." A generative AI model then generates a natural language message, such as "The customer is excited and seems to have found something special," which is displayed on the smart glasses' display. Based on this information, a store clerk can ask the customer, "Did you find anything interesting?", providing more personalized customer service.

[0510] Example prompt sentence:

[0511] Customer sentiment analysis results: Excitement

[0512] Generate natural language message: Your customer is excited, they've found something special.

[0513] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0514] Step 1:

[0515] The EEG measurement device attached to the smart glasses collects the customer's EEG data in real time. The collected EEG data is temporarily stored in a buffer inside the smart glasses. Specifically, the sensor inside the smart glasses measures bioelectric potentials, converts the data into digital format, and stores it.

[0516] Step 2:

[0517] The smart glasses transmit the buffered EEG data to the server using a secure communication protocol (HTTPS), in an encrypted format to ensure confidentiality and integrity of the data. The input EEG data is transmitted and received by the server, and the data is then transferred.

[0518] Step 3:

[0519] The server performs data preprocessing on the received EEG data. Specifically, it applies a noise reduction filter (SciPy) to remove unnecessary noise and outliers. At this stage, the input is encrypted EEG data, and the output is cleaned EEG data.

[0520] Step 4:

[0521] The pre-processed clean EEG data is input into a deep learning model (TensorFlow). The deep learning model analyzes the customer's emotions and thoughts from the EEG data based on pre-learned patterns. The input is the clean EEG data, and the output is the analysis result of emotions or thoughts. Specifically, pattern recognition based on numerical data is performed.

[0522] Step 5:

[0523] The analysis results are input into a generative AI model (OpenAI GPT) and converted into natural language. For example, if the emotion "excited" is analyzed, the generative AI model generates a natural language message such as "The customer is excited and seems to have found something special." The input is the analysis result of the emotion or thought, and the output is a natural language text message.

[0524] Step 6:

[0525] The generated natural language message is then sent to the smart glasses using a secure communication protocol (HTTPS). This message is then displayed on the smart glasses' display, allowing the store clerk to understand the customer's emotions and thoughts in real time and take appropriate action. Specifically, the input is a natural language text message, and the output is a text message displayed on the smart glasses' display.

[0526] 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.

[0527] This invention is a system designed to support communication in medical and care settings, analyzing the electroencephalogram data of care recipients and converting their emotions and thoughts into natural language for display. This system also combines an emotion engine to improve the accuracy of emotion recognition.

[0528] First, the terminal collects real-time EEG data from an EEG measuring device ("EEG measuring device") worn on the head of the care recipient or medical care recipient. The collected data is temporarily stored in a buffer and then transmitted to a server using a secure communication protocol. The secure communication protocol guarantees the confidentiality and integrity of the data.

[0529] On the server side, the received EEG data is first preprocessed. Noise reduction filters are applied to clean up the data and convert it into a format suitable for model input. The preprocessed data is then input into a deep learning model. This model learns from large amounts of data and analyzes the emotions and thoughts of care recipients or medical care recipients with high accuracy.

[0530] Meanwhile, the emotion engine has the ability to recognize user emotions in real time. The emotion data obtained by this emotion engine is also used as input data for deep learning models, improving the accuracy of emotion analysis. In addition, the data output by the emotion engine is also input into generative AI models, allowing for more appropriate adjustment of the content of generated natural language messages.

[0531] The analysis results of the deep learning model are input into a generative AI model, which converts emotions and thoughts into natural language. For example, if the emotion "anxiety" is analyzed, the generative AI model converts this into a natural language sentence such as "I am anxious. I am worried about something." The converted natural language message is then sent back to the device using a secure communication protocol.

[0532] Finally, the device displays the received message on a monitor, allowing caregivers and medical professionals to visually understand the emotions and thoughts of the person receiving care or medical care, and respond appropriately.

[0533] Specific examples

[0534] For example, suppose a care recipient is having difficulty communicating. The device collects the recipient's brainwave data in real time and sends it to the server. The server then performs preprocessing and analyzes the emotion of "anxiety" using a deep learning model. The analysis results are linked to the emotion engine, which then provides more detailed emotional data.

[0535] Based on the analysis results and the emotion engine data, the generative AI model converts the message into natural language: "I'm anxious, I'm worried about something." This message is sent to the device, and the message is displayed on the device's monitor. The caregiver can immediately understand that the person being cared for is feeling anxious and provide appropriate care.

[0536] This system will significantly improve communication between care recipients and those receiving medical care in medical and nursing care settings, enabling appropriate care and treatment to be provided. By combining it with an emotion engine, it will be possible to respond more accurately and appropriately.

[0537] The processing flow will be explained below.

[0538] Step 1:

[0539] The terminal collects real-time brain wave data from an EEG measurement device worn on the head of the care recipient or medical care recipient, samples the electrical signals from the device at regular intervals, and temporarily stores the data in a buffer.

[0540] Step 2:

[0541] The device divides the collected EEG data into packets and sends them to the server using a secure communication protocol (e.g., HTTPS). When a certain amount of data has accumulated in the buffer, it is sent sequentially.

[0542] Step 3:

[0543] The server reconstructs the received packets to obtain the raw EEG data, which is then stored in memory and sent to the next processing step.

[0544] Step 4:

[0545] The server applies noise reduction filters to clean up the EEG data, using band-pass filters and artifact removal algorithms to remove unwanted noise and outliers.

[0546] Step 5:

[0547] The server standardizes the clean EEG data and converts it into a format suitable for the analysis model, for example, by performing data normalization and reshaping operations to create a dataset for input to the deep learning model.

[0548] Step 6:

[0549] The server inputs the preprocessed data into a deep learning model to analyze emotions and thoughts. It uses techniques such as convolutional neural networks (CNNs) and recurrent neural networks (RNNs) to identify emotion categories. The analysis results are output along with a probability score.

[0550] Step 7:

[0551] The server uses an emotion engine to recognize the user's emotions in real time, analyzing facial expressions, voice, and body movements to generate emotion data.

[0552] Step 8:

[0553] The server uses the emotion data obtained by the emotion engine in conjunction with a deep learning model, which improves the accuracy of emotion analysis.

[0554] Step 9:

[0555] The server creates a data structure to be input into the generative AI model based on the analysis results of the deep learning model and the emotion data obtained from the emotion engine. For example, if the emotion category "anxiety" and detailed emotion data are identified, that information is used as input to the generative AI model.

[0556] Step 10:

[0557] The server uses a generative AI model to convert emotion categories and detailed emotion data into natural language, and the generated text is output as a natural language sentence such as "I'm anxious, I'm worried about something."

[0558] Step 11:

[0559] The server then divides the generated text message into packets and sends them to the device using a secure communication protocol, again using a protocol such as HTTPS for secure communication.

[0560] Step 12:

[0561] The terminal reconstructs the received message packets to obtain the generated natural language text, which is stored in memory.

[0562] Step 13:

[0563] The device displays the received text messages on a monitor, allowing caregivers and medical professionals to visually check the messages and understand the emotions and thoughts of the person receiving care or medical care in real time.

[0564] This series of processing steps will create a system that analyzes the brainwave data of care recipients and medical care recipients, and displays their emotions and thoughts in natural language. By combining it with an emotion engine, it will be possible to respond more accurately and appropriately.

[0565] Example 2

[0566] 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."

[0567] With conventional technology, it has been difficult to accurately grasp the emotions and thoughts of care recipients and medical care recipients. Especially in situations where communication is difficult, the lack of information necessary to provide appropriate care and treatment reduces the quality of communication, negatively impacting the patient's experience and treatment outcomes. Another problem is that security and data accuracy are not adequately guaranteed when collecting and analyzing EEG data.

[0568] The identification process by the identification 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 preprocessing transmitted electroencephalogram data and converting it into an analyzable data format, means for inputting the preprocessed electroencephalogram data into a deep learning model to analyze emotions and thoughts, and means for complementing the analysis results of the deep learning model using an emotion engine. This makes it possible to safely and accurately analyze the electroencephalogram data acquired by the electroencephalogram measurement device via a secure communication protocol, and ultimately convert the intended emotions and thoughts into natural language with high accuracy and display them.

[0569] An "EEG measuring device" is a device that is worn on the head of a care recipient or medical care recipient and measures the bioelectric potential signals generated by the brain in real time.

[0570] A "secure communication protocol" is a communication method used to maintain the confidentiality and integrity of communication data, and includes encryption technologies such as TLS and SSL.

[0571] "Preprocessing" is the process of preparing collected raw data for analysis by performing processes such as noise removal and data format conversion.

[0572] A "deep learning model" is a machine learning algorithm that uses large amounts of data to learn and analyze complex nonlinear relationships with high precision, and includes neural networks.

[0573] The "emotion engine" is a system that recognizes the user's emotions in real time and supplements them with analytical data.

[0574] A "generative AI model" is an artificial intelligence model that generates new information based on learned data, and has the ability to generate sentences in natural language processing tasks, for example.

[0575] "Natural language" refers to a language used by humans on a daily basis, specifically spoken or written language, as opposed to programming languages ​​or codes.

[0576] A "monitor" is a device connected to a computer or other electronic device for displaying information.

[0577] This invention is a system for supporting communication in medical and nursing care settings. This system analyzes the emotions and thoughts of care recipients and medical care recipients with high accuracy, converts them into natural language, and displays them.

[0578] First, a terminal collects EEG data in real time using an electroencephalogram (EEG) measuring device (for example, a generic name "EEG measuring device") attached to the head of the care recipient or medical care recipient. The EEG device acquires biopotential signals from the brain through electrodes and transmits the data to the terminal via Bluetooth. This collected data is temporarily stored in a buffer within the terminal.

[0579] The device then transmits the collected EEG data to a server using a secure communication protocol such as TLS, which ensures data confidentiality and integrity.

[0580] The EEG data received by the server is first preprocessed by applying a noise reduction filter. For example, a band-pass filter is used to remove noise and extract only the components in the desired frequency range (4 Hz to 40 Hz). This preprocessed data is then input into a deep learning model (e.g., LSTM or CNN).

[0581] A deep learning model analyzes emotions and thoughts from preprocessed EEG data. The model learns from large amounts of past data and can identify the emotions and thoughts of care recipients and medical care recipients with high accuracy. For example, the analysis can identify the emotion "anxiety."

[0582] The server inputs the analysis results into an emotion engine (e.g., a generic name for this is an "emotion recognition system") to obtain detailed emotion data in real time. This emotion engine further improves the accuracy of the analysis results.

[0583] Next, based on the analysis results and data obtained from the emotion engine, a generative AI model (such as GPT-4) is used to convert emotions and thoughts into natural language messages. For example, based on the emotion data "anxiety," a natural language sentence such as "I am anxious, I am worried about something" is generated.

[0584] The generated natural language message is then sent from the server to the device using a secure communication protocol. The device then displays the received message on a monitor. This allows caregivers and medical professionals to instantly understand the emotions and thoughts of the person receiving care or medical care, enabling them to respond appropriately.

[0585] A specific example is shown below. If a care recipient is experiencing difficulty communicating, the device collects the care recipient's brainwave data in real time and sends it to the server. The server preprocesses the data, removing noise and converting the data format. The deep learning model then analyzes the emotion "anxiety." Based on the results of this analysis, the emotion engine provides detailed emotion data, and finally, the generative AI model generates a natural language message such as "I'm anxious. I'm worried about something." This message is sent to the device and displayed on the monitor.

[0586] An example prompt is:

[0587] "Please analyze the following EEG data and translate the emotions and thoughts of the care recipient into natural language:

[0588] Input data: {EEG data}

[0589] Additional data from the emotion engine: {emotion data}"

[0590] This will significantly improve communication between care recipients and medical professionals in medical and nursing care settings, allowing for more appropriate care and treatment to be provided.

[0591] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0592] Step 1:

[0593] The terminal collects EEG data in real time using an EEG measuring device attached to the care recipient or medical care recipient. The input is the biopotential signal from the EEG measuring device, and the output is data stored in a buffer within the terminal. The data stored in this buffer is ready to be sent to the next stage.

[0594] Step 2:

[0595] The EEG data collected by the device is sent to the server using a secure communication protocol (e.g., TLS). The input is the EEG data stored in the buffer on the device, and the output is the data sent to the server via a secure channel. Data confidentiality and integrity are guaranteed.

[0596] Step 3:

[0597] The server performs preprocessing on the received EEG data. The input is raw EEG data received through a secure communication protocol, and the output is a clean data format with noise removed. Specifically, the processing involves applying a bandpass filter to remove noise and converting the signal into frequency components. The target frequency range is generally 4Hz to 40Hz.

[0598] Step 4:

[0599] The server inputs the preprocessed data into a deep learning model. The input is preprocessed EEG data, and the output is analyzed emotion and thought data. Deep learning models (e.g., LSTM or CNN) learn from large amounts of past training data and can identify emotions and thoughts with high accuracy. This analysis outputs emotions such as "anxiety" or "relief."

[0600] Step 5:

[0601] The server inputs the analysis results into an emotion engine to obtain detailed emotion data. The input is emotion data obtained from the deep learning model, and the output is detailed emotion data complemented by the emotion engine. The emotion engine (such as EmotionAI) performs effective emotion recognition in real time, improving the accuracy of the analysis results.

[0602] Step 6:

[0603] The server uses a generative AI model (e.g., GPT-4) to generate natural language messages based on data obtained from the deep learning model and emotion engine. The input is detailed emotion data and analysis results, and the output is a sentence converted into natural language. For example, when emotion data of "anxiety" is input, a message such as "I am anxious, I am worried about something" is generated.

[0604] Step 7:

[0605] The server transmits the generated natural language message to the terminal again using a secure communication protocol. The input is the generated natural language message, and the output is the message sent to the terminal through the secure channel.

[0606] Step 8:

[0607] The terminal displays the received natural language message on the monitor. The input is the natural language message sent from the server, and the output is the text displayed on the monitor. By looking at this display, caregivers and medical professionals can instantly understand the feelings and thoughts of the person receiving care or medical care, and take appropriate action.

[0608] (Application example 2)

[0609] 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."

[0610] In manufacturing sites that use industrial robots, there is a need to improve work efficiency and ensure safety by smoothing communication between workers and robots and quickly and accurately conveying the feelings and thoughts of workers to the robots. In particular, it is challenging to convey the stress and fatigue of workers to the robots in real time, so that appropriate measures can be taken, and to notify the workers of feedback from the robots in natural language.

[0611] 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.

[0612] In this invention, the server includes means for collecting electroencephalogram data acquired by an electroencephalogram measuring device, means for preprocessing the collected electroencephalogram data and converting it into an analyzable data format, means for inputting the preprocessed electroencephalogram data into a deep learning model to analyze emotions and thoughts, means for converting the analysis results into natural language using a generative AI model, means for displaying the converted natural language, means for notifying the industrial robot of an error based on the emotion analysis results, means for collecting feedback from the industrial robot, and means for converting the collected feedback into natural language and notifying the worker. This allows the worker to communicate their emotions and state to the robot in real time, and the robot can respond appropriately accordingly, thereby improving work efficiency and ensuring safety on site.

[0613] An "EEG measuring device" is a device for acquiring electroencephalogram data, which is worn on the head.

[0614] The "collection means" refers to a mechanism or processing method for capturing data from an electroencephalogram measuring device and accumulating or temporarily storing the data.

[0615] "Preprocessing" refers to the initial stage of data processing to convert collected data into an analyzable format, and includes noise removal and data shaping.

[0616] A "deep learning model" is a machine learning model that automatically extracts features from large amounts of data and analyzes emotions, thoughts, and so on.

[0617] "Means for analyzing emotions and thoughts" refers to methods and technologies that use preprocessed data to analyze and identify the emotions, thoughts, and other information contained within it.

[0618] A "generative AI model" is an artificial intelligence model used to convert analysis results into natural language.

[0619] "Means for converting into natural language" refers to a mechanism or method for converting the analysis results of a deep learning model into language that humans can understand.

[0620] "Displaying means" refers to a device or interface for visually displaying the converted natural language.

[0621] A "secure communication protocol" is a set of communication procedures and rules that ensure data confidentiality and integrity.

[0622] A "noise reduction filter" is an algorithm or method for removing unwanted noise from collected data.

[0623] An "industrial robot" is an automated work machine used in production sites such as factories.

[0624] A "means for collecting feedback" is a method or mechanism for capturing and storing information from an industrial robot.

[0625] This invention is a system that analyzes brainwave data and converts emotions and thoughts into natural language, facilitating communication with industrial robots. The system includes a process that acquires a worker's brainwaves in real time, analyzes the data using a deep learning model, and converts emotions and thoughts into natural language. Furthermore, it can send feedback to the industrial robot based on the analysis results and communicate the robot's response to the worker.

[0626] Hardware used:

[0627] 1. Brainwave Sensor - worn on the worker's head to collect brainwave data in real time.

[0628] 2. Head-Mounted Display (HMD) - A device worn by the worker to provide visual feedback.

[0629] 3. RobotInterface - An interface device for communicating with and receiving feedback from an industrial robot.

[0630] Software used:

[0631] 1. Secure Communication Protocol - Sends EEG data to the server while maintaining data confidentiality.

[0632] 2. Deep learning model (AnalysisModel) - A model that analyzes emotions and thoughts based on preprocessed EEG data.

[0633] 3. Emotion Engine - An engine that analyzes analyzed emotional data with even greater precision.

[0634] 4. Natural Language Generation Model (NaturalLanguageGenerator) - A model for converting emotions and thoughts into natural language.

[0635] Specific processing details:

[0636] 1. EEG data collection and preprocessing:

[0637] The EEG data acquired from the EEG measurement device is transmitted via a secure communication protocol to a server, where it is preprocessed using a noise reduction filter and converted into a format suitable for analysis.

[0638] 2. Emotion and thought analysis:

[0639] The pre-processed data is then analyzed using a deep learning model, where the emotion engine analyzes the emotion data with high precision, enabling more accurate identification of emotions and thoughts.

[0640] 3. Natural Language Conversion:

[0641] Based on the analysis results, a generative AI model converts emotions and thoughts into natural language, yielding results such as "I'm feeling stressed" or "I'm tired."

[0642] 4. Notification and feedback for industrial robots:

[0643] The converted natural language messages are sent to the industrial robot through the robot interface. The robot takes appropriate action based on the information and sends feedback to the server if necessary. This feedback is also converted into natural language and communicated to the worker via a head-mounted display.

[0644] Example prompt sentence:

[0645] If a worker feels stressed: "I'm stressed. I need a break."

[0646] If the robot detects an abnormality: "An abnormality has been detected in the cooling system. Please perform maintenance."

[0647] These functions allow workers to communicate their emotions and state of mind to the robot in real time, allowing the robot to respond appropriately based on that information, improving work efficiency and ensuring safety on site.

[0648] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0649] Step 1:

[0650] EEG data collection

[0651] The terminal collects real-time brain wave data from an EEG measurement device attached to the head of the care recipient or worker. The collected data is temporarily stored in the terminal's memory.

[0652] Input: EEG data from an EEG measurement device

[0653] Output: Raw EEG data stored on the device

[0654] Specific operation: An EEG measuring device is attached to the head, and EEG data is continuously acquired.

[0655] Step 2:

[0656] Secure transmission of data

[0657] The device transmits the collected EEG data to a server using a secure communication protocol, and encryption techniques are used to ensure data confidentiality and integrity.

[0658] Input: Raw EEG data stored on the device

[0659] Output: Secure EEG data sent to a server

[0660] Specific operation: The collected EEG data is sent to a server via a secure communication protocol (encrypted communication channel).

[0661] Step 3:

[0662] Data Preprocessing

[0663] The server performs preprocessing on the EEG data it receives, applying a noise reduction filter and converting the data into an analyzable format. In preprocessing, unnecessary noise is removed and the data is shaped.

[0664] Input: Secure EEG data sent to the server

[0665] Output: Preprocessed and clean EEG data

[0666] What happens: A noise removal filter runs on the server, shaping and transforming the data.

[0667] Step 4:

[0668] Analysis using deep learning models

[0669] The pre-processed data is fed into a deep learning model to analyze emotions and thoughts. The model learns from a large amount of data and can identify emotions and thoughts with high accuracy.

[0670] Input: Preprocessed clean EEG data

[0671] Output: Analyzed emotion and thought data

[0672] Specific operation: The deep learning model works to extract features from input data and analyze emotions and thoughts with high accuracy.

[0673] Step 5:

[0674] Improved accuracy with emotion engine

[0675] The emotion engine is applied to the analysis results to improve the accuracy of the emotion data. The emotion engine has its own algorithm and provides more detailed emotion data.

[0676] Input: Analysis results from deep learning model

[0677] Output: Emotion data with improved accuracy

[0678] Specific operation: The emotion engine further analyzes the analysis results and improves the accuracy of the emotion data.

[0679] Step 6:

[0680] Natural language conversion using generative AI models

[0681] The improved emotion data is then fed into a generative AI model and converted into natural language, generating messages such as "I'm feeling stressed" or "I'm tired."

[0682] Input: Emotion data with improved accuracy

[0683] Output: Message converted into natural language

[0684] Specific operation: The generative AI model generates prompt sentences based on emotional data and outputs natural language messages.

[0685] Step 7:

[0686] Notification for industrial robots

[0687] The generated natural language message is sent to the industrial robot through the robot interface, and the robot then takes appropriate action based on the information.

[0688] Input: A message converted into natural language

[0689] Output: Feedback message sent to the industrial robot

[0690] Specific behavior: Natural language messages are sent to the robot interface, and the robot adjusts its behavior based on the information it receives.

[0691] Step 8:

[0692] Gathering feedback from industrial robots

[0693] Feedback from industrial robots is collected, converted into natural language, and communicated to workers. Feedback messages are displayed on a head-mounted display.

[0694] Input: Feedback from an industrial robot

[0695] Output: A natural language message sent to the worker

[0696] Specific operation: Feedback from the robot is sent to a server, converted into natural language through a generative AI model, and communicated to the worker via a head-mounted display.

[0697] 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.

[0698] 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.

[0699] 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.

[0700] [Third embodiment]

[0701] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0702] 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.

[0703] 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).

[0704] 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.

[0705] 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.

[0706] 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).

[0707] 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.

[0708] 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.

[0709] 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.

[0710] 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.

[0711] 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.

[0712] 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."

[0713] This invention is a system designed to support communication in medical and care settings, and analyzes the electroencephalogram data of care recipients and medical care recipients, converting their emotions and thoughts into natural language and displaying them. Below, we will explain in detail the role that the program in this system plays and the processing it performs.

[0714] First, the terminal collects EEG data in real time from an EEG measurement device ("EEG measuring device") worn on the head of the care recipient or medical care recipient. The collected data is temporarily stored in a buffer and then transmitted to a server using a secure communication protocol. The secure communication protocol guarantees the confidentiality and integrity of the data.

[0715] On the server side, the received EEG data is first preprocessed. Noise reduction filters are applied to clean up the data and convert it into a format suitable for model input. The preprocessed data is then input into a deep learning model. This model learns from large amounts of data and analyzes the emotions and thoughts of care recipients or medical care recipients with high accuracy.

[0716] The analysis results obtained by the deep learning model are input into a generative AI model, which converts emotions and thoughts into natural language. For example, if the emotion "anxiety" is analyzed, the generative AI model converts this into a natural language sentence such as "I am anxious. I am worried about something." The converted natural language message is then sent back to the device using a secure communication protocol.

[0717] Finally, the device displays the received message on a monitor, allowing caregivers and medical professionals to visually understand the emotions and thoughts of the person receiving care or medical care, and respond appropriately.

[0718] Specific examples

[0719] For example, suppose a care recipient is having difficulty communicating. The device collects the care recipient's brain wave data in real time and sends it to a server. The server performs preprocessing and uses a deep learning model to analyze the emotion "anxiety." The generative AI model converts this analysis result into natural language, saying "I'm anxious, I'm worried about something," and sends this message to the device. The device's monitor displays "I'm anxious, I'm worried about something," and the caregiver immediately understands that the care recipient is feeling anxious.

[0720] This system will significantly improve communication between care recipients and medical professionals in medical and nursing care settings, allowing for the provision of appropriate care and treatment.

[0721] The processing flow will be explained below.

[0722] Step 1:

[0723] The terminal collects real-time brain wave data from an EEG measurement device worn on the head of the care recipient or medical care recipient, samples the electrical signals from the device at regular intervals, and temporarily stores the data in a buffer.

[0724] Step 2:

[0725] The device divides the collected EEG data into packets and sends them to the server using a secure communication protocol (e.g., HTTPS). When a certain amount of data has accumulated in the buffer, it is sent sequentially.

[0726] Step 3:

[0727] The server reconstructs the received packets to obtain the raw EEG data, which is then stored in memory and sent to the next processing step.

[0728] Step 4:

[0729] The server applies noise reduction filters to clean up the EEG data, using band-pass filters and artifact removal algorithms to remove unwanted noise and outliers.

[0730] Step 5:

[0731] The server standardizes the clean EEG data and converts it into a format suitable for the analysis model, for example, by performing data normalization and reshaping operations to create a dataset for input to the deep learning model.

[0732] Step 6:

[0733] The server inputs the preprocessed data into a deep learning model to analyze emotions and thoughts. It uses techniques such as convolutional neural networks (CNNs) and recurrent neural networks (RNNs) to identify emotion categories. The analysis results are output along with a probability score.

[0734] Step 7:

[0735] The server creates a data structure to be input into a generative AI model based on the output of the deep learning model. For example, if the emotion category "anxiety" is identified, that information is used as input to a generative AI model (e.g., GPT-4).

[0736] Step 8:

[0737] The server uses a generative AI model to convert emotion categories into natural language, and the generated text is output as a natural language sentence such as "I'm anxious, I'm worried about something."

[0738] Step 9:

[0739] The server then divides the generated text message into packets and sends them to the device using a secure communication protocol, again using a protocol such as HTTPS for secure communication.

[0740] Step 10:

[0741] The terminal reconstructs the received message packets to obtain the generated natural language text, which is stored in memory.

[0742] Step 11:

[0743] The device displays the received text messages on a monitor, allowing caregivers and medical professionals to visually check the messages and understand the emotions and thoughts of the person receiving care or medical care in real time.

[0744] This series of processing steps realizes a system that analyzes the brain wave data of care recipients and medical care recipients and displays their emotions and thoughts in natural language.

[0745] Example 1

[0746] 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."

[0747] Conventional EEG data analysis systems require a lot of manual operations and time for collecting, preprocessing, analyzing, and presenting the results, making it difficult to provide fast and accurate communication support. Furthermore, they often lack important processing functions such as secure communication, noise reduction, and natural language generation, resulting in problems with data integrity and reliability. This limits their practical use in medical and nursing care settings.

[0748] 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.

[0749] In this invention, the server includes means for collecting EEG data acquired by an EEG measuring device, means for transmitting the collected EEG data to the server using a secure communication protocol, means for preprocessing the transmitted EEG data and converting it into an analyzable data format, means for applying a noise reduction filter in the preprocessing stage, means for inputting the preprocessed EEG data into a deep learning model to analyze emotions and thoughts, means for converting the analysis results into natural language using a generative AI model, means for transmitting the converted natural language message to a terminal using a secure communication protocol, and means for displaying the converted natural language. This makes it possible to analyze EEG data quickly and accurately and visually present the results in natural language.

[0750] An "EEG measuring device" is a device that measures and collects the brain wave data of care recipients and medical care recipients in real time.

[0751] A "secure communications protocol" is an encrypted communications method that ensures confidentiality and integrity during data transmission.

[0752] "Preprocessing" refers to the process of removing noise from raw EEG data and converting the data format to make it analyzable.

[0753] A "noise removal filter" is an algorithm or processing means for removing unnecessary noise contained in electroencephalogram data.

[0754] A "deep learning model" is an artificial intelligence model that learns using large amounts of data and analyzes emotions and thoughts from input data with high accuracy.

[0755] A "generative AI model" is an artificial intelligence model that converts analyzed emotions and thoughts into natural language.

[0756] "Natural language" refers to the language used by humans on a daily basis, and refers to the textual output generated by generative AI models.

[0757] "Terminal" means an electronic device that collects data from the EEG measurement device and receives and displays messages from the server.

[0758] "Display" refers to presenting the generated natural language message on a monitor or screen.

[0759] This invention is a system designed to support communication in medical and care settings, analyzing the electroencephalogram (EEG) data of care recipients and converting their emotions and thoughts into natural language for display. To implement this invention, an EEG measurement device, a terminal, a server, a deep learning model, and a generative AI model are required.

[0760] First, the terminal collects brain wave data in real time via an electroencephalogram (EEG) device worn on the head of the care recipient or medical care recipient. An example of an EEG device is an EEG device. The EEG device measures the brain waves of the care recipient or medical care recipient through electrodes and transmits them to the terminal as a digital signal.

[0761] The collected EEG data is then temporarily stored in a buffer on the device and then transmitted to the server using a secure communication protocol (e.g., HTTPS or SSL / TLS), which ensures data confidentiality and integrity.

[0762] On the server side, the received EEG data is first preprocessed. This preprocessing involves applying a noise reduction filter, specifically an algorithm such as a Butterworth filter. The noise reduction filter removes unnecessary noise from the EEG data and converts it into a clean data format.

[0763] The preprocessed data is then input into a deep learning model. The deep learning model is built using machine learning libraries such as TensorFlow and PyTorch, and analyzes the emotions and thoughts of care recipients and medical care recipients with high accuracy from large amounts of training data. The analysis results in labels for emotions and thoughts such as "anxiety," "joy," and "sadness."

[0764] The resulting analysis is then fed into a generative AI model (e.g., GPT-3), which converts emotions and thoughts into natural language. Specifically, the generative AI model is fed the following prompt:

[0765] Analyze the care recipient's EEG data. Generate a message to display if "anxiety" is detected.

[0766] Based on this prompt, the generative AI model generates a natural language message such as "I'm anxious, something is bothering me."

[0767] The generated natural language message is then sent to the device using a secure communication protocol. The device then displays the received message on a monitor, allowing caregivers and medical professionals to visually understand the emotions and thoughts of the person receiving care or medical care, enabling them to respond quickly and appropriately.

[0768] For example, suppose a care recipient is having difficulty communicating. In this case, the device collects the care recipient's brain wave data in real time and sends it to the server. The server performs preprocessing and uses a deep learning model to analyze the emotion "anxiety." The generative AI model converts this analysis result into natural language, saying "I'm anxious, I'm worried about something," and sends this message to the device. The device's monitor displays "I'm anxious, I'm worried about something," allowing the caregiver to immediately understand that the care recipient is feeling anxious.

[0769] This system will significantly improve communication between care recipients and medical professionals in medical and nursing care settings, allowing for the provision of appropriate care and treatment.

[0770] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0771] Step 1:

[0772] The terminal collects real-time brain wave data from the EEG measurement device.

[0773] Input: EEG data of care recipients and medical care recipients

[0774] How it works: The device connects to the EEG device via wireless communication such as Bluetooth or Wi-Fi, and periodically acquires EEG data. The EEG device measures the brainwaves through electrodes and transmits them to the device as digital signals.

[0775] Output: Collected EEG data

[0776] Step 2:

[0777] The terminal temporarily stores the collected brain wave data in a buffer.

[0778] Input: Collected EEG data

[0779] What it does: Stores data in temporary storage on the device and prepares it for transmission.

[0780] Output: Buffered EEG data

[0781] Step 3:

[0782] The device transmits the brainwave data to the server using a secure communication protocol.

[0783] Input: Buffered EEG data

[0784] How it works: The device sends data to the server using an encrypted channel (e.g. HTTPS or SSL / TLS) and waits for an acknowledgment to confirm the communication was successful.

[0785] Output: EEG data sent to the server

[0786] Step 4:

[0787] The server performs preprocessing on the received EEG data.

[0788] Input: Transmitted EEG data

[0789] Behavior: The server first checks the integrity of the received data and makes sure there is no missing data.

[0790] Output: Data ready for preprocessing

[0791] Step 5:

[0792] The server applies a noise reduction filter as part of the preprocessing.

[0793] Input: Data ready for preprocessing

[0794] How it works: The server uses algorithms such as Butterworth filters to remove noise.

[0795] Output: Clean EEG data

[0796] Step 6:

[0797] The server inputs the pre-processed, clean EEG data into a deep learning model to analyze emotions and thoughts.

[0798] Input: Clean EEG data

[0799] How it works: The server runs deep learning models using libraries such as TensorFlow and PyTorch. The models use pre-trained parameters to analyze input data and output emotion and thought labels.

[0800] Output: Emotion or thought analysis (e.g., "anxiety")

[0801] Step 7:

[0802] The server inputs the analysis results into a generative AI model, converting emotions and thoughts into natural language.

[0803] Input: Emotion and thought analysis results

[0804] How it works: A generative AI model (e.g., GPT-3) receives the analysis results as a prompt and generates a natural language message accordingly. A specific prompt might be something like, "The care recipient's emotion has been analyzed as 'anxiety.' Please generate a sentence that describes this emotion."

[0805] Output: Generated natural language message (e.g., "I'm anxious, something is bothering me")

[0806] Step 8:

[0807] The server transmits the generated natural language message to the terminal using a secure communication protocol.

[0808] Input: Generated natural language message

[0809] How it works: The server sends the generated message back to the device over an encrypted channel.

[0810] Output: A natural language message sent to the terminal

[0811] Step 9:

[0812] The terminal displays the received message on the monitor.

[0813] Input: A natural language message sent to the terminal

[0814] BEHAVIOR: The device receives the message, formats it appropriately as text, and displays it on the screen.

[0815] Output: Natural language message displayed on the monitor

[0816] Through this series of steps, the emotions and thoughts of care recipients and medical personnel in medical and nursing care settings can be analyzed quickly and accurately, and presented visually as natural language messages, enabling appropriate and prompt responses.

[0817] (Application example 1)

[0818] 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."

[0819] With conventional methods, it is difficult to accurately grasp the emotions and thoughts of care recipients in nursing and medical settings, which often results in delays in appropriate responses and care. Furthermore, it is difficult to understand customer emotions in real time in physical stores, making it difficult to provide personalized services to improve customer satisfaction. To solve this problem, it is necessary to analyze the brainwave data of care recipients, medical recipients, or customers, understand their emotions and thoughts in real time, and provide appropriate responses.

[0820] 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.

[0821] In this invention, the server includes means for collecting EEG data acquired by an EEG measuring device, means for preprocessing the collected EEG data and converting it into an analyzable data format, means for inputting the preprocessed EEG data into a deep learning model to analyze emotions and thoughts, means for converting the analysis results into natural language using a generative AI model, means for displaying the converted natural language, means for displaying on smart glasses, means for collecting and analyzing EEG data of customers in a physical store, and means for providing the analysis results to store staff in the physical store in real time. This not only enables the emotions and thoughts of care recipients and medical care recipients to be grasped in real time and appropriate care and responses to be provided, but also enables understanding of customer emotions in physical stores and personalized customer service.

[0822] An "EEG measuring device" is a device worn on the head to collect electroencephalogram data.

[0823] "Collecting" is the process of gathering and storing data obtained from an EEG measuring device.

[0824] "Preprocessing" is the process of converting collected raw data into an analyzable format and removing noise and unnecessary information.

[0825] A "deep learning model" is a type of artificial intelligence that learns from large amounts of data and extracts specific patterns and features from the input data.

[0826] "Emotion and thought analysis" is the process of using deep learning models to infer a subject's emotional and thought state from EEG data.

[0827] A "generative AI model" is an artificial intelligence model that converts analyzed emotional and thought data into natural language.

[0828] "Conversion to natural language" is the process of converting numerical data and analytical results into textual language that is easy for humans to understand.

[0829] "Display" refers to the act of showing the converted natural language message on an output device such as a display.

[0830] "Smart glasses" are glasses-type devices that have a display function and can display information in real time.

[0831] A "physical store" is a store that sells or provides services at a physical location.

[0832] "Providing in real time" means instantly obtaining and displaying the latest data and information on the subject and providing analysis results.

[0833] The system for implementing this invention is mainly composed of smart glasses, a server, and a secure communication infrastructure. Specific examples of the main hardware and software of this system include an EEG measurement device, deep learning libraries (TensorFlow, PyTorch), secure communication protocols (HTTPS, MQTT), a generative AI model (OpenAI GPT), and data preprocessing libraries (NumPy, SciPy).

[0834] Explanation of program processing

[0835] 1. Acquisition of EEG data

[0836] The EEG measurement device attached to the smart glasses collects the customer's brainwave data in real time, which is first temporarily stored in a buffer.

[0837] 2. Data transmission

[0838] The buffered EEG data is then sent to a server using a secure communication protocol (HTTPS), ensuring confidentiality and integrity of the data.

[0839] 3. Data Preprocessing

[0840] On the server side, a noise reduction filter (SciPy) is applied to the received EEG data to clean it up, and then the data is converted into a format suitable for the deep learning model (TensorFlow).

[0841] 4. Emotion or thought analysis

[0842] The cleaned data is then fed into a deep learning model (TensorFlow) to analyze customer emotions and thoughts. This model uses previous learnings from large amounts of data to predict emotions and thoughts with high accuracy.

[0843] 5. Natural Language Translation

[0844] The analyzed data is input into a generative AI model (OpenAI GPT) and converted into natural language. For example, if the emotion "excited" is analyzed, the generative AI model generates a text message such as "The customer is excited. They seem to have found something special."

[0845] 6. Provision and display of results

[0846] The generated text message is then sent to the smart glasses using a secure communication protocol (HTTPS), which then displays the message on the smart glasses' display, allowing store staff to understand the customer's emotions and thoughts in real time and take appropriate action.

[0847] Specific examples

[0848] Consider a scenario in which a customer enters a brick-and-mortar store and puts on smart glasses. While the customer is browsing products, an EEG measurement device captures brainwave data and sends it to a server in real time. The data is preprocessed on the server, and a deep learning model is used to analyze the data and determine that the customer is "excited." A generative AI model then generates a natural language message, such as "The customer is excited and seems to have found something special," which is displayed on the smart glasses' display. Based on this information, a store clerk can ask the customer, "Did you find anything interesting?", providing more personalized customer service.

[0849] Example prompt sentence:

[0850] Customer sentiment analysis results: Excitement

[0851] Generate natural language message: Your customer is excited, they've found something special.

[0852] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0853] Step 1:

[0854] The EEG measurement device attached to the smart glasses collects the customer's EEG data in real time. The collected EEG data is temporarily stored in a buffer inside the smart glasses. Specifically, the sensor inside the smart glasses measures bioelectric potentials, converts the data into digital format, and stores it.

[0855] Step 2:

[0856] The smart glasses transmit the buffered EEG data to the server using a secure communication protocol (HTTPS), in an encrypted format to ensure confidentiality and integrity of the data. The input EEG data is transmitted and received by the server, and the data is then transferred.

[0857] Step 3:

[0858] The server performs data preprocessing on the received EEG data. Specifically, it applies a noise reduction filter (SciPy) to remove unnecessary noise and outliers. At this stage, the input is encrypted EEG data, and the output is cleaned EEG data.

[0859] Step 4:

[0860] The pre-processed clean EEG data is input into a deep learning model (TensorFlow). The deep learning model analyzes the customer's emotions and thoughts from the EEG data based on pre-learned patterns. The input is the clean EEG data, and the output is the analysis result of emotions or thoughts. Specifically, pattern recognition based on numerical data is performed.

[0861] Step 5:

[0862] The analysis results are input into a generative AI model (OpenAI GPT) and converted into natural language. For example, if the emotion "excited" is analyzed, the generative AI model generates a natural language message such as "The customer is excited and seems to have found something special." The input is the analysis result of the emotion or thought, and the output is a natural language text message.

[0863] Step 6:

[0864] The generated natural language message is then sent to the smart glasses using a secure communication protocol (HTTPS). This message is then displayed on the smart glasses' display, allowing the store clerk to understand the customer's emotions and thoughts in real time and take appropriate action. Specifically, the input is a natural language text message, and the output is a text message displayed on the smart glasses' display.

[0865] 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.

[0866] This invention is a system designed to support communication in medical and care settings, analyzing the electroencephalogram data of care recipients and converting their emotions and thoughts into natural language for display. This system also combines an emotion engine to improve the accuracy of emotion recognition.

[0867] First, the terminal collects real-time EEG data from an EEG measuring device ("EEG measuring device") worn on the head of the care recipient or medical care recipient. The collected data is temporarily stored in a buffer and then transmitted to a server using a secure communication protocol. The secure communication protocol guarantees the confidentiality and integrity of the data.

[0868] On the server side, the received EEG data is first preprocessed. Noise reduction filters are applied to clean up the data and convert it into a format suitable for model input. The preprocessed data is then input into a deep learning model. This model learns from large amounts of data and analyzes the emotions and thoughts of care recipients or medical care recipients with high accuracy.

[0869] Meanwhile, the emotion engine has the ability to recognize user emotions in real time. The emotion data obtained by this emotion engine is also used as input data for deep learning models, improving the accuracy of emotion analysis. In addition, the data output by the emotion engine is also input into generative AI models, allowing for more appropriate adjustment of the content of generated natural language messages.

[0870] The analysis results of the deep learning model are input into a generative AI model, which converts emotions and thoughts into natural language. For example, if the emotion "anxiety" is analyzed, the generative AI model converts this into a natural language sentence such as "I am anxious. I am worried about something." The converted natural language message is then sent back to the device using a secure communication protocol.

[0871] Finally, the device displays the received message on a monitor, allowing caregivers and medical professionals to visually understand the emotions and thoughts of the person receiving care or medical care, and respond appropriately.

[0872] Specific examples

[0873] For example, suppose a care recipient is having difficulty communicating. The device collects the recipient's brainwave data in real time and sends it to the server. The server then performs preprocessing and analyzes the emotion of "anxiety" using a deep learning model. The analysis results are linked to the emotion engine, which then provides more detailed emotional data.

[0874] Based on the analysis results and the emotion engine data, the generative AI model converts the message into natural language: "I'm anxious, I'm worried about something." This message is sent to the device, and the message is displayed on the device's monitor. The caregiver can immediately understand that the person being cared for is feeling anxious and provide appropriate care.

[0875] This system will significantly improve communication between care recipients and those receiving medical care in medical and nursing care settings, enabling appropriate care and treatment to be provided. By combining it with an emotion engine, it will be possible to respond more accurately and appropriately.

[0876] The processing flow will be explained below.

[0877] Step 1:

[0878] The terminal collects real-time brain wave data from an EEG measurement device worn on the head of the care recipient or medical care recipient, samples the electrical signals from the device at regular intervals, and temporarily stores the data in a buffer.

[0879] Step 2:

[0880] The device divides the collected EEG data into packets and sends them to the server using a secure communication protocol (e.g., HTTPS). When a certain amount of data has accumulated in the buffer, it is sent sequentially.

[0881] Step 3:

[0882] The server reconstructs the received packets to obtain the raw EEG data, which is then stored in memory and sent to the next processing step.

[0883] Step 4:

[0884] The server applies noise reduction filters to clean up the EEG data, using band-pass filters and artifact removal algorithms to remove unwanted noise and outliers.

[0885] Step 5:

[0886] The server standardizes the clean EEG data and converts it into a format suitable for the analysis model, for example, by performing data normalization and reshaping operations to create a dataset for input to the deep learning model.

[0887] Step 6:

[0888] The server inputs the preprocessed data into a deep learning model to analyze emotions and thoughts. It uses techniques such as convolutional neural networks (CNNs) and recurrent neural networks (RNNs) to identify emotion categories. The analysis results are output along with a probability score.

[0889] Step 7:

[0890] The server uses an emotion engine to recognize the user's emotions in real time, analyzing facial expressions, voice, and body movements to generate emotion data.

[0891] Step 8:

[0892] The server uses the emotion data obtained by the emotion engine in conjunction with a deep learning model, which improves the accuracy of emotion analysis.

[0893] Step 9:

[0894] The server creates a data structure to be input into the generative AI model based on the analysis results of the deep learning model and the emotion data obtained from the emotion engine. For example, if the emotion category "anxiety" and detailed emotion data are identified, that information is used as input to the generative AI model.

[0895] Step 10:

[0896] The server uses a generative AI model to convert emotion categories and detailed emotion data into natural language, and the generated text is output as a natural language sentence such as "I'm anxious, I'm worried about something."

[0897] Step 11:

[0898] The server then divides the generated text message into packets and sends them to the device using a secure communication protocol, again using a protocol such as HTTPS for secure communication.

[0899] Step 12:

[0900] The terminal reconstructs the received message packets to obtain the generated natural language text, which is stored in memory.

[0901] Step 13:

[0902] The device displays the received text messages on a monitor, allowing caregivers and medical professionals to visually check the messages and understand the emotions and thoughts of the person receiving care or medical care in real time.

[0903] This series of processing steps will create a system that analyzes the brainwave data of care recipients and medical care recipients, and displays their emotions and thoughts in natural language. By combining it with an emotion engine, it will be possible to respond more accurately and appropriately.

[0904] Example 2

[0905] 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."

[0906] With conventional technology, it has been difficult to accurately grasp the emotions and thoughts of care recipients and medical care recipients. Especially in situations where communication is difficult, the lack of information necessary to provide appropriate care and treatment reduces the quality of communication, negatively impacting the patient's experience and treatment outcomes. Another problem is that security and data accuracy are not adequately guaranteed when collecting and analyzing EEG data.

[0907] The identification process by the identification 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 preprocessing transmitted electroencephalogram data and converting it into an analyzable data format, means for inputting the preprocessed electroencephalogram data into a deep learning model to analyze emotions and thoughts, and means for complementing the analysis results of the deep learning model using an emotion engine. This makes it possible to safely and accurately analyze the electroencephalogram data acquired by the electroencephalogram measurement device via a secure communication protocol, and ultimately convert the intended emotions and thoughts into natural language with high accuracy and display them.

[0908] An "EEG measuring device" is a device that is worn on the head of a care recipient or medical care recipient and measures the bioelectric potential signals generated by the brain in real time.

[0909] A "secure communication protocol" is a communication method used to maintain the confidentiality and integrity of communication data, and includes encryption technologies such as TLS and SSL.

[0910] "Preprocessing" is the process of preparing collected raw data for analysis by performing processes such as noise removal and data format conversion.

[0911] A "deep learning model" is a machine learning algorithm that uses large amounts of data to learn and analyze complex nonlinear relationships with high precision, and includes neural networks.

[0912] The "emotion engine" is a system that recognizes the user's emotions in real time and supplements them with analytical data.

[0913] A "generative AI model" is an artificial intelligence model that generates new information based on learned data, and has the ability to generate sentences in natural language processing tasks, for example.

[0914] "Natural language" refers to a language used by humans on a daily basis, specifically spoken or written language, as opposed to programming languages ​​or codes.

[0915] A "monitor" is a device connected to a computer or other electronic device for displaying information.

[0916] This invention is a system for supporting communication in medical and nursing care settings. This system analyzes the emotions and thoughts of care recipients and medical care recipients with high accuracy, converts them into natural language, and displays them.

[0917] First, a terminal collects EEG data in real time using an electroencephalogram (EEG) measuring device (for example, a generic name "EEG measuring device") attached to the head of the care recipient or medical care recipient. The EEG device acquires biopotential signals from the brain through electrodes and transmits the data to the terminal via Bluetooth. This collected data is temporarily stored in a buffer within the terminal.

[0918] The device then transmits the collected EEG data to a server using a secure communication protocol such as TLS, which ensures data confidentiality and integrity.

[0919] The EEG data received by the server is first preprocessed by applying a noise reduction filter. For example, a band-pass filter is used to remove noise and extract only the components in the desired frequency range (4 Hz to 40 Hz). This preprocessed data is then input into a deep learning model (e.g., LSTM or CNN).

[0920] A deep learning model analyzes emotions and thoughts from preprocessed EEG data. The model learns from large amounts of past data and can identify the emotions and thoughts of care recipients and medical care recipients with high accuracy. For example, the analysis can identify the emotion "anxiety."

[0921] The server inputs the analysis results into an emotion engine (e.g., a generic name for this is an "emotion recognition system") to obtain detailed emotion data in real time. This emotion engine further improves the accuracy of the analysis results.

[0922] Next, based on the analysis results and data obtained from the emotion engine, a generative AI model (such as GPT-4) is used to convert emotions and thoughts into natural language messages. For example, based on the emotion data "anxiety," a natural language sentence such as "I am anxious, I am worried about something" is generated.

[0923] The generated natural language message is then sent from the server to the device using a secure communication protocol. The device then displays the received message on a monitor. This allows caregivers and medical professionals to instantly understand the emotions and thoughts of the person receiving care or medical care, enabling them to respond appropriately.

[0924] A specific example is shown below. If a care recipient is experiencing difficulty communicating, the device collects the care recipient's brainwave data in real time and sends it to the server. The server preprocesses the data, removing noise and converting the data format. The deep learning model then analyzes the emotion "anxiety." Based on the results of this analysis, the emotion engine provides detailed emotion data, and finally, the generative AI model generates a natural language message such as "I'm anxious. I'm worried about something." This message is sent to the device and displayed on the monitor.

[0925] An example prompt is:

[0926] "Please analyze the following EEG data and translate the emotions and thoughts of the care recipient into natural language:

[0927] Input data: {EEG data}

[0928] Additional data from the emotion engine: {emotion data}"

[0929] This will significantly improve communication between care recipients and medical professionals in medical and nursing care settings, allowing for more appropriate care and treatment to be provided.

[0930] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0931] Step 1:

[0932] The terminal collects EEG data in real time using an EEG measuring device attached to the care recipient or medical care recipient. The input is the biopotential signal from the EEG measuring device, and the output is data stored in a buffer within the terminal. The data stored in this buffer is ready to be sent to the next stage.

[0933] Step 2:

[0934] The EEG data collected by the device is sent to the server using a secure communication protocol (e.g., TLS). The input is the EEG data stored in the buffer on the device, and the output is the data sent to the server via a secure channel. Data confidentiality and integrity are guaranteed.

[0935] Step 3:

[0936] The server performs preprocessing on the received EEG data. The input is raw EEG data received through a secure communication protocol, and the output is a clean data format with noise removed. Specifically, the processing involves applying a bandpass filter to remove noise and converting the signal into frequency components. The target frequency range is generally 4Hz to 40Hz.

[0937] Step 4:

[0938] The server inputs the preprocessed data into a deep learning model. The input is preprocessed EEG data, and the output is analyzed emotion and thought data. Deep learning models (e.g., LSTM or CNN) learn from large amounts of past training data and can identify emotions and thoughts with high accuracy. This analysis outputs emotions such as "anxiety" or "relief."

[0939] Step 5:

[0940] The server inputs the analysis results into an emotion engine to obtain detailed emotion data. The input is emotion data obtained from the deep learning model, and the output is detailed emotion data complemented by the emotion engine. The emotion engine (such as EmotionAI) performs effective emotion recognition in real time, improving the accuracy of the analysis results.

[0941] Step 6:

[0942] The server uses a generative AI model (e.g., GPT-4) to generate natural language messages based on data obtained from the deep learning model and emotion engine. The input is detailed emotion data and analysis results, and the output is a sentence converted into natural language. For example, when emotion data of "anxiety" is input, a message such as "I am anxious, I am worried about something" is generated.

[0943] Step 7:

[0944] The server transmits the generated natural language message to the terminal again using a secure communication protocol. The input is the generated natural language message, and the output is the message sent to the terminal through the secure channel.

[0945] Step 8:

[0946] The terminal displays the received natural language message on the monitor. The input is the natural language message sent from the server, and the output is the text displayed on the monitor. By looking at this display, caregivers and medical professionals can instantly understand the feelings and thoughts of the person receiving care or medical care, and take appropriate action.

[0947] (Application example 2)

[0948] 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."

[0949] In manufacturing sites that use industrial robots, there is a need to improve work efficiency and ensure safety by smoothing communication between workers and robots and quickly and accurately conveying the feelings and thoughts of workers to the robots. In particular, it is challenging to convey the stress and fatigue of workers to the robots in real time, so that appropriate measures can be taken, and to notify the workers of feedback from the robots in natural language.

[0950] 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.

[0951] In this invention, the server includes means for collecting electroencephalogram data acquired by an electroencephalogram measuring device, means for preprocessing the collected electroencephalogram data and converting it into an analyzable data format, means for inputting the preprocessed electroencephalogram data into a deep learning model to analyze emotions and thoughts, means for converting the analysis results into natural language using a generative AI model, means for displaying the converted natural language, means for notifying the industrial robot of an error based on the emotion analysis results, means for collecting feedback from the industrial robot, and means for converting the collected feedback into natural language and notifying the worker. This allows the worker to communicate their emotions and state to the robot in real time, and the robot can respond appropriately accordingly, thereby improving work efficiency and ensuring safety on site.

[0952] An "EEG measuring device" is a device for acquiring electroencephalogram data, which is worn on the head.

[0953] The "collection means" refers to a mechanism or processing method for capturing data from an electroencephalogram measuring device and accumulating or temporarily storing the data.

[0954] "Preprocessing" refers to the initial stage of data processing to convert collected data into an analyzable format, and includes noise removal and data shaping.

[0955] A "deep learning model" is a machine learning model that automatically extracts features from large amounts of data and analyzes emotions, thoughts, and so on.

[0956] "Means for analyzing emotions and thoughts" refers to methods and technologies that use preprocessed data to analyze and identify the emotions, thoughts, and other information contained within it.

[0957] A "generative AI model" is an artificial intelligence model used to convert analysis results into natural language.

[0958] "Means for converting into natural language" refers to a mechanism or method for converting the analysis results of a deep learning model into language that humans can understand.

[0959] "Displaying means" refers to a device or interface for visually displaying the converted natural language.

[0960] A "secure communication protocol" is a set of communication procedures and rules that ensure data confidentiality and integrity.

[0961] A "noise reduction filter" is an algorithm or method for removing unwanted noise from collected data.

[0962] An "industrial robot" is an automated work machine used in production sites such as factories.

[0963] A "means for collecting feedback" is a method or mechanism for capturing and storing information from an industrial robot.

[0964] This invention is a system that analyzes brainwave data and converts emotions and thoughts into natural language, facilitating communication with industrial robots. The system includes a process that acquires a worker's brainwaves in real time, analyzes the data using a deep learning model, and converts emotions and thoughts into natural language. Furthermore, it can send feedback to the industrial robot based on the analysis results and communicate the robot's response to the worker.

[0965] Hardware used:

[0966] 1. Brainwave Sensor - worn on the worker's head to collect brainwave data in real time.

[0967] 2. Head-Mounted Display (HMD) - A device worn by the worker to provide visual feedback.

[0968] 3. RobotInterface - An interface device for communicating with and receiving feedback from an industrial robot.

[0969] Software used:

[0970] 1. Secure Communication Protocol - Sends EEG data to the server while maintaining data confidentiality.

[0971] 2. Deep learning model (AnalysisModel) - A model that analyzes emotions and thoughts based on preprocessed EEG data.

[0972] 3. Emotion Engine - An engine that analyzes analyzed emotional data with even greater precision.

[0973] 4. Natural Language Generation Model (NaturalLanguageGenerator) - A model for converting emotions and thoughts into natural language.

[0974] Specific processing details:

[0975] 1. EEG data collection and preprocessing:

[0976] The EEG data acquired from the EEG measurement device is transmitted via a secure communication protocol to a server, where it is preprocessed using a noise reduction filter and converted into a format suitable for analysis.

[0977] 2. Emotion and thought analysis:

[0978] The pre-processed data is then analyzed using a deep learning model, where the emotion engine analyzes the emotion data with high precision, enabling more accurate identification of emotions and thoughts.

[0979] 3. Natural Language Conversion:

[0980] Based on the analysis results, a generative AI model converts emotions and thoughts into natural language, yielding results such as "I'm feeling stressed" or "I'm tired."

[0981] 4. Notification and feedback for industrial robots:

[0982] The converted natural language messages are sent to the industrial robot through the robot interface. The robot takes appropriate action based on the information and sends feedback to the server if necessary. This feedback is also converted into natural language and communicated to the worker via a head-mounted display.

[0983] Example prompt sentence:

[0984] If a worker feels stressed: "I'm stressed. I need a break."

[0985] If the robot detects an abnormality: "An abnormality has been detected in the cooling system. Please perform maintenance."

[0986] These functions allow workers to communicate their emotions and state of mind to the robot in real time, allowing the robot to respond appropriately based on that information, improving work efficiency and ensuring safety on site.

[0987] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0988] Step 1:

[0989] EEG data collection

[0990] The terminal collects real-time brain wave data from an EEG measurement device attached to the head of the care recipient or worker. The collected data is temporarily stored in the terminal's memory.

[0991] Input: EEG data from an EEG measurement device

[0992] Output: Raw EEG data stored on the device

[0993] Specific operation: An EEG measuring device is attached to the head, and EEG data is continuously acquired.

[0994] Step 2:

[0995] Secure transmission of data

[0996] The device transmits the collected EEG data to a server using a secure communication protocol, and encryption techniques are used to ensure data confidentiality and integrity.

[0997] Input: Raw EEG data stored on the device

[0998] Output: Secure EEG data sent to a server

[0999] Specific operation: The collected EEG data is sent to a server via a secure communication protocol (encrypted communication channel).

[1000] Step 3:

[1001] Data Preprocessing

[1002] The server performs preprocessing on the EEG data it receives, applying a noise reduction filter and converting the data into an analyzable format. In preprocessing, unnecessary noise is removed and the data is shaped.

[1003] Input: Secure EEG data sent to the server

[1004] Output: Preprocessed and clean EEG data

[1005] What happens: A noise removal filter runs on the server, shaping and transforming the data.

[1006] Step 4:

[1007] Analysis using deep learning models

[1008] The pre-processed data is fed into a deep learning model to analyze emotions and thoughts. The model learns from a large amount of data and can identify emotions and thoughts with high accuracy.

[1009] Input: Preprocessed clean EEG data

[1010] Output: Analyzed emotion and thought data

[1011] Specific operation: The deep learning model works to extract features from input data and analyze emotions and thoughts with high accuracy.

[1012] Step 5:

[1013] Improved accuracy with emotion engine

[1014] The emotion engine is applied to the analysis results to improve the accuracy of the emotion data. The emotion engine has its own algorithm and provides more detailed emotion data.

[1015] Input: Analysis results from deep learning model

[1016] Output: Emotion data with improved accuracy

[1017] Specific operation: The emotion engine further analyzes the analysis results and improves the accuracy of the emotion data.

[1018] Step 6:

[1019] Natural language conversion using generative AI models

[1020] The improved emotion data is then fed into a generative AI model and converted into natural language, generating messages such as "I'm feeling stressed" or "I'm tired."

[1021] Input: Emotion data with improved accuracy

[1022] Output: Message converted into natural language

[1023] Specific operation: The generative AI model generates prompt sentences based on emotional data and outputs natural language messages.

[1024] Step 7:

[1025] Notification for industrial robots

[1026] The generated natural language message is sent to the industrial robot through the robot interface, and the robot then takes appropriate action based on the information.

[1027] Input: A message converted into natural language

[1028] Output: Feedback message sent to the industrial robot

[1029] Specific behavior: Natural language messages are sent to the robot interface, and the robot adjusts its behavior based on the information it receives.

[1030] Step 8:

[1031] Gathering feedback from industrial robots

[1032] Feedback from industrial robots is collected, converted into natural language, and communicated to workers. Feedback messages are displayed on a head-mounted display.

[1033] Input: Feedback from an industrial robot

[1034] Output: A natural language message sent to the worker

[1035] Specific operation: Feedback from the robot is sent to a server, converted into natural language through a generative AI model, and communicated to the worker via a head-mounted display.

[1036] 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.

[1037] 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.

[1038] 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.

[1039] [Fourth embodiment]

[1040] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1041] 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.

[1042] 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).

[1043] 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.

[1044] 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.

[1045] 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).

[1046] 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.

[1047] 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.

[1048] 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.

[1049] 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.

[1050] 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.

[1051] 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.

[1052] 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."

[1053] This invention is a system designed to support communication in medical and care settings, and analyzes the electroencephalogram data of care recipients and medical care recipients, converting their emotions and thoughts into natural language and displaying them. Below, we will explain in detail the role that the program in this system plays and the processing it performs.

[1054] First, the terminal collects EEG data in real time from an EEG measurement device ("EEG measuring device") worn on the head of the care recipient or medical care recipient. The collected data is temporarily stored in a buffer and then transmitted to a server using a secure communication protocol. The secure communication protocol guarantees the confidentiality and integrity of the data.

[1055] On the server side, the received EEG data is first preprocessed. Noise reduction filters are applied to clean up the data and convert it into a format suitable for model input. The preprocessed data is then input into a deep learning model. This model learns from large amounts of data and analyzes the emotions and thoughts of care recipients or medical care recipients with high accuracy.

[1056] The analysis results obtained by the deep learning model are input into a generative AI model, which converts emotions and thoughts into natural language. For example, if the emotion "anxiety" is analyzed, the generative AI model converts this into a natural language sentence such as "I am anxious. I am worried about something." The converted natural language message is then sent back to the device using a secure communication protocol.

[1057] Finally, the device displays the received message on a monitor, allowing caregivers and medical professionals to visually understand the emotions and thoughts of the person receiving care or medical care, and respond appropriately.

[1058] Specific examples

[1059] For example, suppose a care recipient is having difficulty communicating. The device collects the care recipient's brain wave data in real time and sends it to a server. The server performs preprocessing and uses a deep learning model to analyze the emotion "anxiety." The generative AI model converts this analysis result into natural language, saying "I'm anxious, I'm worried about something," and sends this message to the device. The device's monitor displays "I'm anxious, I'm worried about something," and the caregiver immediately understands that the care recipient is feeling anxious.

[1060] This system will significantly improve communication between care recipients and medical professionals in medical and nursing care settings, allowing for the provision of appropriate care and treatment.

[1061] The processing flow will be explained below.

[1062] Step 1:

[1063] The terminal collects real-time brain wave data from an EEG measurement device worn on the head of the care recipient or medical care recipient, samples the electrical signals from the device at regular intervals, and temporarily stores the data in a buffer.

[1064] Step 2:

[1065] The device divides the collected EEG data into packets and sends them to the server using a secure communication protocol (e.g., HTTPS). When a certain amount of data has accumulated in the buffer, it is sent sequentially.

[1066] Step 3:

[1067] The server reconstructs the received packets to obtain the raw EEG data, which is then stored in memory and sent to the next processing step.

[1068] Step 4:

[1069] The server applies noise reduction filters to clean up the EEG data, using band-pass filters and artifact removal algorithms to remove unwanted noise and outliers.

[1070] Step 5:

[1071] The server standardizes the clean EEG data and converts it into a format suitable for the analysis model, for example, by performing data normalization and reshaping operations to create a dataset for input to the deep learning model.

[1072] Step 6:

[1073] The server inputs the preprocessed data into a deep learning model to analyze emotions and thoughts. It uses techniques such as convolutional neural networks (CNNs) and recurrent neural networks (RNNs) to identify emotion categories. The analysis results are output along with a probability score.

[1074] Step 7:

[1075] The server creates a data structure to be input into a generative AI model based on the output of the deep learning model. For example, if the emotion category "anxiety" is identified, that information is used as input to a generative AI model (e.g., GPT-4).

[1076] Step 8:

[1077] The server uses a generative AI model to convert emotion categories into natural language, and the generated text is output as a natural language sentence such as "I'm anxious, I'm worried about something."

[1078] Step 9:

[1079] The server then divides the generated text message into packets and sends them to the device using a secure communication protocol, again using a protocol such as HTTPS for secure communication.

[1080] Step 10:

[1081] The terminal reconstructs the received message packets to obtain the generated natural language text, which is stored in memory.

[1082] Step 11:

[1083] The device displays the received text messages on a monitor, allowing caregivers and medical professionals to visually check the messages and understand the emotions and thoughts of the person receiving care or medical care in real time.

[1084] This series of processing steps realizes a system that analyzes the brain wave data of care recipients and medical care recipients and displays their emotions and thoughts in natural language.

[1085] Example 1

[1086] 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."

[1087] Conventional EEG data analysis systems require a lot of manual operations and time for collecting, preprocessing, analyzing, and presenting the results, making it difficult to provide fast and accurate communication support. Furthermore, they often lack important processing functions such as secure communication, noise reduction, and natural language generation, resulting in problems with data integrity and reliability. This limits their practical use in medical and nursing care settings.

[1088] 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.

[1089] In this invention, the server includes means for collecting EEG data acquired by an EEG measuring device, means for transmitting the collected EEG data to the server using a secure communication protocol, means for preprocessing the transmitted EEG data and converting it into an analyzable data format, means for applying a noise reduction filter in the preprocessing stage, means for inputting the preprocessed EEG data into a deep learning model to analyze emotions and thoughts, means for converting the analysis results into natural language using a generative AI model, means for transmitting the converted natural language message to a terminal using a secure communication protocol, and means for displaying the converted natural language. This makes it possible to analyze EEG data quickly and accurately and visually present the results in natural language.

[1090] An "EEG measuring device" is a device that measures and collects the brain wave data of care recipients and medical care recipients in real time.

[1091] A "secure communications protocol" is an encrypted communications method that ensures confidentiality and integrity during data transmission.

[1092] "Preprocessing" refers to the process of removing noise from raw EEG data and converting the data format to make it analyzable.

[1093] A "noise removal filter" is an algorithm or processing means for removing unnecessary noise contained in electroencephalogram data.

[1094] A "deep learning model" is an artificial intelligence model that learns using large amounts of data and analyzes emotions and thoughts from input data with high accuracy.

[1095] A "generative AI model" is an artificial intelligence model that converts analyzed emotions and thoughts into natural language.

[1096] "Natural language" refers to the language used by humans on a daily basis, and refers to the textual output generated by generative AI models.

[1097] "Terminal" means an electronic device that collects data from the EEG measurement device and receives and displays messages from the server.

[1098] "Display" refers to presenting the generated natural language message on a monitor or screen.

[1099] This invention is a system designed to support communication in medical and care settings, analyzing the electroencephalogram (EEG) data of care recipients and converting their emotions and thoughts into natural language for display. To implement this invention, an EEG measurement device, a terminal, a server, a deep learning model, and a generative AI model are required.

[1100] First, the terminal collects brain wave data in real time via an electroencephalogram (EEG) device worn on the head of the care recipient or medical care recipient. An example of an EEG device is an EEG device. The EEG device measures the brain waves of the care recipient or medical care recipient through electrodes and transmits them to the terminal as a digital signal.

[1101] The collected EEG data is then temporarily stored in a buffer on the device and then transmitted to the server using a secure communication protocol (e.g., HTTPS or SSL / TLS), which ensures data confidentiality and integrity.

[1102] On the server side, the received EEG data is first preprocessed. This preprocessing involves applying a noise reduction filter, specifically an algorithm such as a Butterworth filter. The noise reduction filter removes unnecessary noise from the EEG data and converts it into a clean data format.

[1103] The preprocessed data is then input into a deep learning model. The deep learning model is built using machine learning libraries such as TensorFlow and PyTorch, and analyzes the emotions and thoughts of care recipients and medical care recipients with high accuracy from large amounts of training data. The analysis results in labels for emotions and thoughts such as "anxiety," "joy," and "sadness."

[1104] The resulting analysis is then fed into a generative AI model (e.g., GPT-3), which converts emotions and thoughts into natural language. Specifically, the generative AI model is fed the following prompt:

[1105] Analyze the care recipient's EEG data. Generate a message to display if "anxiety" is detected.

[1106] Based on this prompt, the generative AI model generates a natural language message such as "I'm anxious, something is bothering me."

[1107] The generated natural language message is then sent to the device using a secure communication protocol. The device then displays the received message on a monitor, allowing caregivers and medical professionals to visually understand the emotions and thoughts of the person receiving care or medical care, enabling them to respond quickly and appropriately.

[1108] For example, suppose a care recipient is having difficulty communicating. In this case, the device collects the care recipient's brain wave data in real time and sends it to the server. The server performs preprocessing and uses a deep learning model to analyze the emotion "anxiety." The generative AI model converts this analysis result into natural language, saying "I'm anxious, I'm worried about something," and sends this message to the device. The device's monitor displays "I'm anxious, I'm worried about something," allowing the caregiver to immediately understand that the care recipient is feeling anxious.

[1109] This system will significantly improve communication between care recipients and medical professionals in medical and nursing care settings, allowing for the provision of appropriate care and treatment.

[1110] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1111] Step 1:

[1112] The terminal collects real-time brain wave data from the EEG measurement device.

[1113] Input: EEG data of care recipients and medical care recipients

[1114] How it works: The device connects to the EEG device via wireless communication such as Bluetooth or Wi-Fi, and periodically acquires EEG data. The EEG device measures the brainwaves through electrodes and transmits them to the device as digital signals.

[1115] Output: Collected EEG data

[1116] Step 2:

[1117] The terminal temporarily stores the collected brain wave data in a buffer.

[1118] Input: Collected EEG data

[1119] What it does: Stores data in temporary storage on the device and prepares it for transmission.

[1120] Output: Buffered EEG data

[1121] Step 3:

[1122] The device transmits the brainwave data to the server using a secure communication protocol.

[1123] Input: Buffered EEG data

[1124] How it works: The device sends data to the server using an encrypted channel (e.g. HTTPS or SSL / TLS) and waits for an acknowledgment to confirm the communication was successful.

[1125] Output: EEG data sent to the server

[1126] Step 4:

[1127] The server performs preprocessing on the received EEG data.

[1128] Input: Transmitted EEG data

[1129] Behavior: The server first checks the integrity of the received data and makes sure there is no missing data.

[1130] Output: Data ready for preprocessing

[1131] Step 5:

[1132] The server applies a noise reduction filter as part of the preprocessing.

[1133] Input: Data ready for preprocessing

[1134] How it works: The server uses algorithms such as Butterworth filters to remove noise.

[1135] Output: Clean EEG data

[1136] Step 6:

[1137] The server inputs the pre-processed, clean EEG data into a deep learning model to analyze emotions and thoughts.

[1138] Input: Clean EEG data

[1139] How it works: The server runs deep learning models using libraries such as TensorFlow and PyTorch. The models use pre-trained parameters to analyze input data and output emotion and thought labels.

[1140] Output: Emotion or thought analysis (e.g., "anxiety")

[1141] Step 7:

[1142] The server inputs the analysis results into a generative AI model, converting emotions and thoughts into natural language.

[1143] Input: Emotion and thought analysis results

[1144] How it works: A generative AI model (e.g., GPT-3) receives the analysis results as a prompt and generates a natural language message accordingly. A specific prompt might be something like, "The care recipient's emotion has been analyzed as 'anxiety.' Please generate a sentence that describes this emotion."

[1145] Output: Generated natural language message (e.g., "I'm anxious, something is bothering me")

[1146] Step 8:

[1147] The server transmits the generated natural language message to the terminal using a secure communication protocol.

[1148] Input: Generated natural language message

[1149] How it works: The server sends the generated message back to the device over an encrypted channel.

[1150] Output: A natural language message sent to the terminal

[1151] Step 9:

[1152] The terminal displays the received message on the monitor.

[1153] Input: A natural language message sent to the terminal

[1154] BEHAVIOR: The device receives the message, formats it appropriately as text, and displays it on the screen.

[1155] Output: Natural language message displayed on the monitor

[1156] Through this series of steps, the emotions and thoughts of care recipients and medical personnel in medical and nursing care settings can be analyzed quickly and accurately, and presented visually as natural language messages, enabling appropriate and prompt responses.

[1157] (Application example 1)

[1158] 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."

[1159] With conventional methods, it is difficult to accurately grasp the emotions and thoughts of care recipients in nursing and medical settings, which often results in delays in appropriate responses and care. Furthermore, it is difficult to understand customer emotions in real time in physical stores, making it difficult to provide personalized services to improve customer satisfaction. To solve this problem, it is necessary to analyze the brainwave data of care recipients, medical recipients, or customers, understand their emotions and thoughts in real time, and provide appropriate responses.

[1160] 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.

[1161] In this invention, the server includes means for collecting EEG data acquired by an EEG measuring device, means for preprocessing the collected EEG data and converting it into an analyzable data format, means for inputting the preprocessed EEG data into a deep learning model to analyze emotions and thoughts, means for converting the analysis results into natural language using a generative AI model, means for displaying the converted natural language, means for displaying on smart glasses, means for collecting and analyzing EEG data of customers in a physical store, and means for providing the analysis results to store staff in the physical store in real time. This not only enables the emotions and thoughts of care recipients and medical care recipients to be grasped in real time and appropriate care and responses to be provided, but also enables understanding of customer emotions in physical stores and personalized customer service.

[1162] An "EEG measuring device" is a device worn on the head to collect electroencephalogram data.

[1163] "Collecting" is the process of gathering and storing data obtained from an EEG measuring device.

[1164] "Preprocessing" is the process of converting collected raw data into an analyzable format and removing noise and unnecessary information.

[1165] A "deep learning model" is a type of artificial intelligence that learns from large amounts of data and extracts specific patterns and features from the input data.

[1166] "Emotion and thought analysis" is the process of using deep learning models to infer a subject's emotional and thought state from EEG data.

[1167] A "generative AI model" is an artificial intelligence model that converts analyzed emotional and thought data into natural language.

[1168] "Conversion to natural language" is the process of converting numerical data and analytical results into textual language that is easy for humans to understand.

[1169] "Display" refers to the act of showing the converted natural language message on an output device such as a display.

[1170] "Smart glasses" are glasses-type devices that have a display function and can display information in real time.

[1171] A "physical store" is a store that sells or provides services at a physical location.

[1172] "Providing in real time" means instantly obtaining and displaying the latest data and information on the subject and providing analysis results.

[1173] The system for implementing this invention is mainly composed of smart glasses, a server, and a secure communication infrastructure. Specific examples of the main hardware and software of this system include an EEG measurement device, deep learning libraries (TensorFlow, PyTorch), secure communication protocols (HTTPS, MQTT), a generative AI model (OpenAI GPT), and data preprocessing libraries (NumPy, SciPy).

[1174] Explanation of program processing

[1175] 1. Acquisition of EEG data

[1176] The EEG measurement device attached to the smart glasses collects the customer's brainwave data in real time, which is first temporarily stored in a buffer.

[1177] 2. Data transmission

[1178] The buffered EEG data is then sent to a server using a secure communication protocol (HTTPS), ensuring confidentiality and integrity of the data.

[1179] 3. Data Preprocessing

[1180] On the server side, a noise reduction filter (SciPy) is applied to the received EEG data to clean it up, and then the data is converted into a format suitable for the deep learning model (TensorFlow).

[1181] 4. Emotion or thought analysis

[1182] The cleaned data is then fed into a deep learning model (TensorFlow) to analyze customer emotions and thoughts. This model uses previous learnings from large amounts of data to predict emotions and thoughts with high accuracy.

[1183] 5. Natural Language Translation

[1184] The analyzed data is input into a generative AI model (OpenAI GPT) and converted into natural language. For example, if the emotion "excited" is analyzed, the generative AI model generates a text message such as "The customer is excited. They seem to have found something special."

[1185] 6. Provision and display of results

[1186] The generated text message is then sent to the smart glasses using a secure communication protocol (HTTPS), which then displays the message on the smart glasses' display, allowing store staff to understand the customer's emotions and thoughts in real time and take appropriate action.

[1187] Specific examples

[1188] Consider a scenario in which a customer enters a brick-and-mortar store and puts on smart glasses. While the customer is browsing products, an EEG measurement device captures brainwave data and sends it to a server in real time. The data is preprocessed on the server, and a deep learning model is used to analyze the data and determine that the customer is "excited." A generative AI model then generates a natural language message, such as "The customer is excited and seems to have found something special," which is displayed on the smart glasses' display. Based on this information, a store clerk can ask the customer, "Did you find anything interesting?", providing more personalized customer service.

[1189] Example prompt sentence:

[1190] Customer sentiment analysis results: Excitement

[1191] Generate natural language message: Your customer is excited, they've found something special.

[1192] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1193] Step 1:

[1194] The EEG measurement device attached to the smart glasses collects the customer's EEG data in real time. The collected EEG data is temporarily stored in a buffer inside the smart glasses. Specifically, the sensor inside the smart glasses measures bioelectric potentials, converts the data into digital format, and stores it.

[1195] Step 2:

[1196] The smart glasses transmit the buffered EEG data to the server using a secure communication protocol (HTTPS), in an encrypted format to ensure confidentiality and integrity of the data. The input EEG data is transmitted and received by the server, and the data is then transferred.

[1197] Step 3:

[1198] The server performs data preprocessing on the received EEG data. Specifically, it applies a noise reduction filter (SciPy) to remove unnecessary noise and outliers. At this stage, the input is encrypted EEG data, and the output is cleaned EEG data.

[1199] Step 4:

[1200] The pre-processed clean EEG data is input into a deep learning model (TensorFlow). The deep learning model analyzes the customer's emotions and thoughts from the EEG data based on pre-learned patterns. The input is the clean EEG data, and the output is the analysis result of emotions or thoughts. Specifically, pattern recognition based on numerical data is performed.

[1201] Step 5:

[1202] The analysis results are input into a generative AI model (OpenAI GPT) and converted into natural language. For example, if the emotion "excited" is analyzed, the generative AI model generates a natural language message such as "The customer is excited and seems to have found something special." The input is the analysis result of the emotion or thought, and the output is a natural language text message.

[1203] Step 6:

[1204] The generated natural language message is then sent to the smart glasses using a secure communication protocol (HTTPS). This message is then displayed on the smart glasses' display, allowing the store clerk to understand the customer's emotions and thoughts in real time and take appropriate action. Specifically, the input is a natural language text message, and the output is a text message displayed on the smart glasses' display.

[1205] 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.

[1206] This invention is a system designed to support communication in medical and care settings, analyzing the electroencephalogram data of care recipients and converting their emotions and thoughts into natural language for display. This system also combines an emotion engine to improve the accuracy of emotion recognition.

[1207] First, the terminal collects real-time EEG data from an EEG measuring device ("EEG measuring device") worn on the head of the care recipient or medical care recipient. The collected data is temporarily stored in a buffer and then transmitted to a server using a secure communication protocol. The secure communication protocol guarantees the confidentiality and integrity of the data.

[1208] On the server side, the received EEG data is first preprocessed. Noise reduction filters are applied to clean up the data and convert it into a format suitable for model input. The preprocessed data is then input into a deep learning model. This model learns from large amounts of data and analyzes the emotions and thoughts of care recipients or medical care recipients with high accuracy.

[1209] Meanwhile, the emotion engine has the ability to recognize user emotions in real time. The emotion data obtained by this emotion engine is also used as input data for deep learning models, improving the accuracy of emotion analysis. In addition, the data output by the emotion engine is also input into generative AI models, allowing for more appropriate adjustment of the content of generated natural language messages.

[1210] The analysis results of the deep learning model are input into a generative AI model, which converts emotions and thoughts into natural language. For example, if the emotion "anxiety" is analyzed, the generative AI model converts this into a natural language sentence such as "I am anxious. I am worried about something." The converted natural language message is then sent back to the device using a secure communication protocol.

[1211] Finally, the device displays the received message on a monitor, allowing caregivers and medical professionals to visually understand the emotions and thoughts of the person receiving care or medical care, and respond appropriately.

[1212] Specific examples

[1213] For example, suppose a care recipient is having difficulty communicating. The device collects the recipient's brainwave data in real time and sends it to the server. The server then performs preprocessing and analyzes the emotion of "anxiety" using a deep learning model. The analysis results are linked to the emotion engine, which then provides more detailed emotional data.

[1214] Based on the analysis results and the emotion engine data, the generative AI model converts the message into natural language: "I'm anxious, I'm worried about something." This message is sent to the device, and the message is displayed on the device's monitor. The caregiver can immediately understand that the person being cared for is feeling anxious and provide appropriate care.

[1215] This system will significantly improve communication between care recipients and those receiving medical care in medical and nursing care settings, enabling appropriate care and treatment to be provided. By combining it with an emotion engine, it will be possible to respond more accurately and appropriately.

[1216] The processing flow will be explained below.

[1217] Step 1:

[1218] The terminal collects real-time brain wave data from an EEG measurement device worn on the head of the care recipient or medical care recipient, samples the electrical signals from the device at regular intervals, and temporarily stores the data in a buffer.

[1219] Step 2:

[1220] The device divides the collected EEG data into packets and sends them to the server using a secure communication protocol (e.g., HTTPS). When a certain amount of data has accumulated in the buffer, it is sent sequentially.

[1221] Step 3:

[1222] The server reconstructs the received packets to obtain the raw EEG data, which is then stored in memory and sent to the next processing step.

[1223] Step 4:

[1224] The server applies noise reduction filters to clean up the EEG data, using band-pass filters and artifact removal algorithms to remove unwanted noise and outliers.

[1225] Step 5:

[1226] The server standardizes the clean EEG data and converts it into a format suitable for the analysis model, for example, by performing data normalization and reshaping operations to create a dataset for input to the deep learning model.

[1227] Step 6:

[1228] The server inputs the preprocessed data into a deep learning model to analyze emotions and thoughts. It uses techniques such as convolutional neural networks (CNNs) and recurrent neural networks (RNNs) to identify emotion categories. The analysis results are output along with a probability score.

[1229] Step 7:

[1230] The server uses an emotion engine to recognize the user's emotions in real time, analyzing facial expressions, voice, and body movements to generate emotion data.

[1231] Step 8:

[1232] The server uses the emotion data obtained by the emotion engine in conjunction with a deep learning model, which improves the accuracy of emotion analysis.

[1233] Step 9:

[1234] The server creates a data structure to be input into the generative AI model based on the analysis results of the deep learning model and the emotion data obtained from the emotion engine. For example, if the emotion category "anxiety" and detailed emotion data are identified, that information is used as input to the generative AI model.

[1235] Step 10:

[1236] The server uses a generative AI model to convert emotion categories and detailed emotion data into natural language, and the generated text is output as a natural language sentence such as "I'm anxious, I'm worried about something."

[1237] Step 11:

[1238] The server then divides the generated text message into packets and sends them to the device using a secure communication protocol, again using a protocol such as HTTPS for secure communication.

[1239] Step 12:

[1240] The terminal reconstructs the received message packets to obtain the generated natural language text, which is stored in memory.

[1241] Step 13:

[1242] The device displays the received text messages on a monitor, allowing caregivers and medical professionals to visually check the messages and understand the emotions and thoughts of the person receiving care or medical care in real time.

[1243] This series of processing steps will create a system that analyzes the brainwave data of care recipients and medical care recipients, and displays their emotions and thoughts in natural language. By combining it with an emotion engine, it will be possible to respond more accurately and appropriately.

[1244] Example 2

[1245] 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."

[1246] With conventional technology, it has been difficult to accurately grasp the emotions and thoughts of care recipients and medical care recipients. Especially in situations where communication is difficult, the lack of information necessary to provide appropriate care and treatment reduces the quality of communication, negatively impacting the patient's experience and treatment outcomes. Another problem is that security and data accuracy are not adequately guaranteed when collecting and analyzing EEG data.

[1247] The identification process by the identification 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 preprocessing transmitted electroencephalogram data and converting it into an analyzable data format, means for inputting the preprocessed electroencephalogram data into a deep learning model to analyze emotions and thoughts, and means for complementing the analysis results of the deep learning model using an emotion engine. This makes it possible to safely and accurately analyze the electroencephalogram data acquired by the electroencephalogram measurement device via a secure communication protocol, and ultimately convert the intended emotions and thoughts into natural language with high accuracy and display them.

[1248] An "EEG measuring device" is a device that is worn on the head of a care recipient or medical care recipient and measures the bioelectric potential signals generated by the brain in real time.

[1249] A "secure communication protocol" is a communication method used to maintain the confidentiality and integrity of communication data, and includes encryption technologies such as TLS and SSL.

[1250] "Preprocessing" is the process of preparing collected raw data for analysis by performing processes such as noise removal and data format conversion.

[1251] A "deep learning model" is a machine learning algorithm that uses large amounts of data to learn and analyze complex nonlinear relationships with high precision, and includes neural networks.

[1252] The "emotion engine" is a system that recognizes the user's emotions in real time and supplements them with analytical data.

[1253] A "generative AI model" is an artificial intelligence model that generates new information based on learned data, and has the ability to generate sentences in natural language processing tasks, for example.

[1254] "Natural language" refers to a language used by humans on a daily basis, specifically spoken or written language, as opposed to programming languages ​​or codes.

[1255] A "monitor" is a device connected to a computer or other electronic device for displaying information.

[1256] This invention is a system for supporting communication in medical and nursing care settings. This system analyzes the emotions and thoughts of care recipients and medical care recipients with high accuracy, converts them into natural language, and displays them.

[1257] First, a terminal collects EEG data in real time using an electroencephalogram (EEG) measuring device (for example, a generic name "EEG measuring device") attached to the head of the care recipient or medical care recipient. The EEG device acquires biopotential signals from the brain through electrodes and transmits the data to the terminal via Bluetooth. This collected data is temporarily stored in a buffer within the terminal.

[1258] The device then transmits the collected EEG data to a server using a secure communication protocol such as TLS, which ensures data confidentiality and integrity.

[1259] The EEG data received by the server is first preprocessed by applying a noise reduction filter. For example, a band-pass filter is used to remove noise and extract only the components in the desired frequency range (4 Hz to 40 Hz). This preprocessed data is then input into a deep learning model (e.g., LSTM or CNN).

[1260] A deep learning model analyzes emotions and thoughts from preprocessed EEG data. The model learns from large amounts of past data and can identify the emotions and thoughts of care recipients and medical care recipients with high accuracy. For example, the analysis can identify the emotion "anxiety."

[1261] The server inputs the analysis results into an emotion engine (e.g., a generic name for this is an "emotion recognition system") to obtain detailed emotion data in real time. This emotion engine further improves the accuracy of the analysis results.

[1262] Next, based on the analysis results and data obtained from the emotion engine, a generative AI model (such as GPT-4) is used to convert emotions and thoughts into natural language messages. For example, based on the emotion data "anxiety," a natural language sentence such as "I am anxious, I am worried about something" is generated.

[1263] The generated natural language message is then sent from the server to the device using a secure communication protocol. The device then displays the received message on a monitor. This allows caregivers and medical professionals to instantly understand the emotions and thoughts of the person receiving care or medical care, enabling them to respond appropriately.

[1264] A specific example is shown below. If a care recipient is experiencing difficulty communicating, the device collects the care recipient's brainwave data in real time and sends it to the server. The server preprocesses the data, removing noise and converting the data format. The deep learning model then analyzes the emotion "anxiety." Based on the results of this analysis, the emotion engine provides detailed emotion data, and finally, the generative AI model generates a natural language message such as "I'm anxious. I'm worried about something." This message is sent to the device and displayed on the monitor.

[1265] An example prompt is:

[1266] "Please analyze the following EEG data and translate the emotions and thoughts of the care recipient into natural language:

[1267] Input data: {EEG data}

[1268] Additional data from the emotion engine: {emotion data}"

[1269] This will significantly improve communication between care recipients and medical professionals in medical and nursing care settings, allowing for more appropriate care and treatment to be provided.

[1270] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1271] Step 1:

[1272] The terminal collects EEG data in real time using an EEG measuring device attached to the care recipient or medical care recipient. The input is the biopotential signal from the EEG measuring device, and the output is data stored in a buffer within the terminal. The data stored in this buffer is ready to be sent to the next stage.

[1273] Step 2:

[1274] The EEG data collected by the device is sent to the server using a secure communication protocol (e.g., TLS). The input is the EEG data stored in the buffer on the device, and the output is the data sent to the server via a secure channel. Data confidentiality and integrity are guaranteed.

[1275] Step 3:

[1276] The server performs preprocessing on the received EEG data. The input is raw EEG data received through a secure communication protocol, and the output is a clean data format with noise removed. Specifically, the processing involves applying a bandpass filter to remove noise and converting the signal into frequency components. The target frequency range is generally 4Hz to 40Hz.

[1277] Step 4:

[1278] The server inputs the preprocessed data into a deep learning model. The input is preprocessed EEG data, and the output is analyzed emotion and thought data. Deep learning models (e.g., LSTM or CNN) learn from large amounts of past training data and can identify emotions and thoughts with high accuracy. This analysis outputs emotions such as "anxiety" or "relief."

[1279] Step 5:

[1280] The server inputs the analysis results into an emotion engine to obtain detailed emotion data. The input is emotion data obtained from the deep learning model, and the output is detailed emotion data complemented by the emotion engine. The emotion engine (such as EmotionAI) performs effective emotion recognition in real time, improving the accuracy of the analysis results.

[1281] Step 6:

[1282] The server uses a generative AI model (e.g., GPT-4) to generate natural language messages based on data obtained from the deep learning model and emotion engine. The input is detailed emotion data and analysis results, and the output is a sentence converted into natural language. For example, when emotion data of "anxiety" is input, a message such as "I am anxious, I am worried about something" is generated.

[1283] Step 7:

[1284] The server transmits the generated natural language message to the terminal again using a secure communication protocol. The input is the generated natural language message, and the output is the message sent to the terminal through the secure channel.

[1285] Step 8:

[1286] The terminal displays the received natural language message on the monitor. The input is the natural language message sent from the server, and the output is the text displayed on the monitor. By looking at this display, caregivers and medical professionals can instantly understand the feelings and thoughts of the person receiving care or medical care, and take appropriate action.

[1287] (Application example 2)

[1288] 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 robot 414 will be referred to as a "terminal."

[1289] In manufacturing sites that use industrial robots, there is a need to improve work efficiency and ensure safety by smoothing communication between workers and robots and quickly and accurately conveying the feelings and thoughts of workers to the robots. In particular, it is challenging to convey the stress and fatigue of workers to the robots in real time, so that appropriate measures can be taken, and to notify the workers of feedback from the robots in natural language.

[1290] 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.

[1291] In this invention, the server includes means for collecting electroencephalogram data acquired by an electroencephalogram measuring device, means for preprocessing the collected electroencephalogram data and converting it into an analyzable data format, means for inputting the preprocessed electroencephalogram data into a deep learning model to analyze emotions and thoughts, means for converting the analysis results into natural language using a generative AI model, means for displaying the converted natural language, means for notifying the industrial robot of an error based on the emotion analysis results, means for collecting feedback from the industrial robot, and means for converting the collected feedback into natural language and notifying the worker. This allows the worker to communicate their emotions and state to the robot in real time, and the robot can respond appropriately accordingly, thereby improving work efficiency and ensuring safety on site.

[1292] An "EEG measuring device" is a device for acquiring electroencephalogram data, which is worn on the head.

[1293] The "collection means" refers to a mechanism or processing method for capturing data from an electroencephalogram measuring device and accumulating or temporarily storing the data.

[1294] "Preprocessing" refers to the initial stage of data processing to convert collected data into an analyzable format, and includes noise removal and data shaping.

[1295] A "deep learning model" is a machine learning model that automatically extracts features from large amounts of data and analyzes emotions, thoughts, and so on.

[1296] "Means for analyzing emotions and thoughts" refers to methods and technologies that use preprocessed data to analyze and identify the emotions, thoughts, and other information contained within it.

[1297] A "generative AI model" is an artificial intelligence model used to convert analysis results into natural language.

[1298] "Means for converting into natural language" refers to a mechanism or method for converting the analysis results of a deep learning model into language that humans can understand.

[1299] "Displaying means" refers to a device or interface for visually displaying the converted natural language.

[1300] A "secure communication protocol" is a set of communication procedures and rules that ensure data confidentiality and integrity.

[1301] A "noise reduction filter" is an algorithm or method for removing unwanted noise from collected data.

[1302] An "industrial robot" is an automated work machine used in production sites such as factories.

[1303] A "means for collecting feedback" is a method or mechanism for capturing and storing information from an industrial robot.

[1304] This invention is a system that analyzes brainwave data and converts emotions and thoughts into natural language, facilitating communication with industrial robots. The system includes a process that acquires a worker's brainwaves in real time, analyzes the data using a deep learning model, and converts emotions and thoughts into natural language. Furthermore, it can send feedback to the industrial robot based on the analysis results and communicate the robot's response to the worker.

[1305] Hardware used:

[1306] 1. Brainwave Sensor - worn on the worker's head to collect brainwave data in real time.

[1307] 2. Head-Mounted Display (HMD) - A device worn by the worker to provide visual feedback.

[1308] 3. RobotInterface - An interface device for communicating with and receiving feedback from an industrial robot.

[1309] Software used:

[1310] 1. Secure Communication Protocol - Sends EEG data to the server while maintaining data confidentiality.

[1311] 2. Deep learning model (AnalysisModel) - A model that analyzes emotions and thoughts based on preprocessed EEG data.

[1312] 3. Emotion Engine - An engine that analyzes analyzed emotional data with even greater precision.

[1313] 4. Natural Language Generation Model (NaturalLanguageGenerator) - A model for converting emotions and thoughts into natural language.

[1314] Specific processing details:

[1315] 1. EEG data collection and preprocessing:

[1316] The EEG data acquired from the EEG measurement device is transmitted via a secure communication protocol to a server, where it is preprocessed using a noise reduction filter and converted into a format suitable for analysis.

[1317] 2. Emotion and thought analysis:

[1318] The pre-processed data is then analyzed using a deep learning model, where the emotion engine analyzes the emotion data with high precision, enabling more accurate identification of emotions and thoughts.

[1319] 3. Natural Language Conversion:

[1320] Based on the analysis results, a generative AI model converts emotions and thoughts into natural language, yielding results such as "I'm feeling stressed" or "I'm tired."

[1321] 4. Notification and feedback for industrial robots:

[1322] The converted natural language messages are sent to the industrial robot through the robot interface. The robot takes appropriate action based on the information and sends feedback to the server if necessary. This feedback is also converted into natural language and communicated to the worker via a head-mounted display.

[1323] Example prompt sentence:

[1324] If a worker feels stressed: "I'm stressed. I need a break."

[1325] If the robot detects an abnormality: "An abnormality has been detected in the cooling system. Please perform maintenance."

[1326] These functions allow workers to communicate their emotions and state of mind to the robot in real time, allowing the robot to respond appropriately based on that information, improving work efficiency and ensuring safety on site.

[1327] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1328] Step 1:

[1329] EEG data collection

[1330] The terminal collects real-time brain wave data from an EEG measurement device attached to the head of the care recipient or worker. The collected data is temporarily stored in the terminal's memory.

[1331] Input: EEG data from an EEG measurement device

[1332] Output: Raw EEG data stored on the device

[1333] Specific operation: An EEG measuring device is attached to the head, and EEG data is continuously acquired.

[1334] Step 2:

[1335] Secure transmission of data

[1336] The device transmits the collected EEG data to a server using a secure communication protocol, and encryption techniques are used to ensure data confidentiality and integrity.

[1337] Input: Raw EEG data stored on the device

[1338] Output: Secure EEG data sent to a server

[1339] Specific operation: The collected EEG data is sent to a server via a secure communication protocol (encrypted communication channel).

[1340] Step 3:

[1341] Data Preprocessing

[1342] The server performs preprocessing on the EEG data it receives, applying a noise reduction filter and converting the data into an analyzable format. In preprocessing, unnecessary noise is removed and the data is shaped.

[1343] Input: Secure EEG data sent to the server

[1344] Output: Preprocessed and clean EEG data

[1345] What happens: A noise removal filter runs on the server, shaping and transforming the data.

[1346] Step 4:

[1347] Analysis using deep learning models

[1348] The pre-processed data is fed into a deep learning model to analyze emotions and thoughts. The model learns from a large amount of data and can identify emotions and thoughts with high accuracy.

[1349] Input: Preprocessed clean EEG data

[1350] Output: Analyzed emotion and thought data

[1351] Specific operation: The deep learning model works to extract features from input data and analyze emotions and thoughts with high accuracy.

[1352] Step 5:

[1353] Improved accuracy with emotion engine

[1354] The emotion engine is applied to the analysis results to improve the accuracy of the emotion data. The emotion engine has its own algorithm and provides more detailed emotion data.

[1355] Input: Analysis results from deep learning model

[1356] Output: Emotion data with improved accuracy

[1357] Specific operation: The emotion engine further analyzes the analysis results and improves the accuracy of the emotion data.

[1358] Step 6:

[1359] Natural language conversion using generative AI models

[1360] The improved emotion data is then fed into a generative AI model and converted into natural language, generating messages such as "I'm feeling stressed" or "I'm tired."

[1361] Input: Emotion data with improved accuracy

[1362] Output: Message converted into natural language

[1363] Specific operation: The generative AI model generates prompt sentences based on emotional data and outputs natural language messages.

[1364] Step 7:

[1365] Notification for industrial robots

[1366] The generated natural language message is sent to the industrial robot through the robot interface, and the robot then takes appropriate action based on the information.

[1367] Input: A message converted into natural language

[1368] Output: Feedback message sent to the industrial robot

[1369] Specific behavior: Natural language messages are sent to the robot interface, and the robot adjusts its behavior based on the information it receives.

[1370] Step 8:

[1371] Gathering feedback from industrial robots

[1372] Feedback from industrial robots is collected, converted into natural language, and communicated to workers. Feedback messages are displayed on a head-mounted display.

[1373] Input: Feedback from an industrial robot

[1374] Output: A natural language message sent to the worker

[1375] Specific operation: Feedback from the robot is sent to a server, converted into natural language through a generative AI model, and communicated to the worker via a head-mounted display.

[1376] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice 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 voice data.

[1377] 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.

[1378] 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 robot 414.

[1379] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1380] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1381] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[1382] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[1383] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1384] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[1385] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[1386] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1387] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1388] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[1389] Alternatively, the specific processing program 56 may be stored in a storage device such as a server connected to the data processing device 12 via the network 54, and the specific processing program 56 may be downloaded and installed on the computer 22 in response to a request from the data processing device 12.

[1390] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[1391] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1392] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1393] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1394] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1395] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1396] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[1397] The following is further disclosed regarding the above embodiment.

[1398] (Claim 1)

[1399] means for collecting electroencephalogram data obtained by an electroencephalogram measuring device;

[1400] A means for preprocessing the collected EEG data and converting it into an analyzable data format;

[1401] A means of inputting preprocessed EEG data into a deep learning model to analyze emotions and thoughts, and

[1402] A means of converting the analysis results into natural language using a generative AI model;

[1403] The system includes a means for displaying the converted natural language.

[1404] (Claim 2)

[1405] 10. The system of claim 1, further comprising means for transmitting the collected electroencephalogram data using a secure communication protocol.

[1406] (Claim 3)

[1407] 10. The system of claim 1, further comprising means for applying a noise reduction filter in a pre-processing stage.

[1408] "Example 1"

[1409] (Claim 1)

[1410] means for collecting electroencephalogram data obtained by an electroencephalogram measuring device;

[1411] means for transmitting the collected electroencephalogram data to a server using a secure communication protocol;

[1412] means for preprocessing the transmitted EEG data and converting it into an analyzable data format;

[1413] means for applying a denoising filter in a pre-processing stage;

[1414] A means of inputting preprocessed EEG data into a deep learning model to analyze emotions and thoughts, and

[1415] A means of converting the analysis results into natural language using a generative AI model;

[1416] means for transmitting the converted natural language message to the terminal using a secure communication protocol;

[1417] means for displaying the converted natural language;

[1418] A system including:

[1419] (Claim 2)

[1420] 10. The system of claim 1, further comprising means for applying a noise reduction filter in a pre-processing stage.

[1421] (Claim 3)

[1422] 10. The system of claim 1, further comprising means for transmitting the collected electroencephalogram data using a secure communication protocol.

[1423] "Application Example 1"

[1424] (Claim 1)

[1425] means for collecting electroencephalogram data obtained by an electroencephalogram measuring device;

[1426] A means for preprocessing the collected EEG data and converting it into an analyzable data format;

[1427] A means of inputting preprocessed EEG data into a deep learning model to analyze emotions and thoughts, and

[1428] A means of converting the analysis results into natural language using a generative AI model;

[1429] means for displaying the converted natural language;

[1430] a means for displaying on the smart glasses;

[1431] A means for collecting and analyzing customer brain wave data in physical stores;

[1432] A system that includes a means for providing analysis results to store staff in physical stores in real time.

[1433] (Claim 2)

[1434] 10. The system of claim 1, further comprising means for transmitting the collected electroencephalogram data using a secure communication protocol.

[1435] (Claim 3)

[1436] 10. The system of claim 1, further comprising means for applying a noise reduction filter in a pre-processing stage.

[1437] "Example 2: Combining Emotion Engines"

[1438] (Claim 1)

[1439] means for collecting electroencephalogram data obtained by an electroencephalogram measuring device;

[1440] A means for transmitting the collected EEG data to a server using a secure communication protocol;

[1441] means for preprocessing the transmitted EEG data and converting it into an analyzable data format;

[1442] A means of inputting preprocessed EEG data into a deep learning model to analyze emotions and thoughts, and

[1443] A means to complement the analysis results of deep learning models with an emotion engine; and

[1444] A means of converting the analysis results into natural language using a generative AI model;

[1445] The system includes a means for displaying the converted natural language.

[1446] (Claim 2)

[1447] 10. The system of claim 1, further comprising means for transmitting the collected electroencephalogram data using a secure communication protocol.

[1448] (Claim 3)

[1449] 10. The system of claim 1, further comprising means for applying a noise reduction filter in a pre-processing stage.

[1450] "Application example 2 when combining emotion engines"

[1451] (Claim 1)

[1452] means for collecting electroencephalogram data obtained by an electroencephalogram measuring device;

[1453] A means for preprocessing the collected EEG data and converting it into an analyzable data format;

[1454] A means of inputting preprocessed EEG data into a deep learning model to analyze emotions and thoughts, and

[1455] A means of converting the analysis results into natural language using a generative AI model;

[1456] means for displaying the converted natural language;

[1457] a means for notifying the industrial robot of the mistake based on the emotion analysis result;

[1458] a means for collecting feedback from the industrial robot;

[1459] The system includes a means for converting collected feedback into natural language and notifying the worker.

[1460] (Claim 2)

[1461] 10. The system of claim 1, further comprising means for transmitting the collected electroencephalogram data using a secure communication protocol.

[1462] (Claim 3)

[1463] 10. The system of claim 1, further comprising means for applying a noise reduction filter in a pre-processing stage. [Explanation of symbols]

[1464] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. means for collecting electroencephalogram data obtained by an electroencephalogram measuring device; A means for preprocessing the collected EEG data and converting it into an analyzable data format; A means of inputting preprocessed EEG data into a deep learning model to analyze emotions and thoughts, and A means of converting the analysis results into natural language using a generative AI model; The system includes a means for displaying the converted natural language.

2. 10. The system of claim 1, further comprising means for transmitting the collected electroencephalogram data using a secure communication protocol.

3. 10. The system of claim 1, further comprising means for applying a noise reduction filter in a pre-processing stage.

Images