KR102066225B1 - Smart health care apparatus, system and method using artificial intelligence - Google Patents

Abstract

The present invention relates to an artificial intelligence smart diagnosis apparatus, system and method. The artificial intelligence smart diagnosis apparatus according to the present invention comprises: an intelligent personal assistant data input/output module which collects audio and video data about condition of a user and outputs synthesized voice required to lead a conversation with the user; a data analysis module which analyzes a conversation period with the user, a change in voice of the user, and a conversation content by using the collected data, and outputs the analyzed data; and a diagnostic module which diagnoses a changed condition of the user by using the analyzed data. Therefore, a patient in the distance can be diagnosed and monitored by using an artificial intelligence input/output device based on voice and video data.

Description

SMART HEALTH CARE APPARATUS, SYSTEM AND METHOD USING ARTIFICIAL INTELLIGENCE

The present invention is a smart diagnostic apparatus, system and method of the artificial intelligence function, and more specifically, using a data input and output device, that is, an artificial intelligence speaker with a camera to observe the condition of the user needing protection and suspected symptoms It relates to a smart diagnostic device, system and method of artificial intelligence function for diagnosing and informing the patient occurrence.

In the past, if the interface between man and machine was based on touch relations in the third industrial revolution, the basis of the interface between man and machine is expected to be based on human voice.

What is needed to achieve an interface based on human voice is audio technology and intelligent personal assistant technology. Audio technology collects the highest quality voice information that can be recognized by the machine from human voice, and the collected voice information is interpreted into a machine-recognized language by the artificial intelligence assistant technology.

Recently, the AI speaker is located in the representative service using the artificial intelligence assistant and voice recognition. The AI speaker corresponding to the audio input / output device is operated by an AI algorithm that performs an AI assistant function. The AI speaker differs from the general speaker in that it performs a function of listening to music and searching information by using voice recognition through audio input and output.

AI speakers, which combine AI-based speech natural language processing technology with next-generation home automation technology to create an automated environment for users, are expanding their markets around Amazon and Google.

The function of AI speakers is expanding. From early music listening and weather search to smart home, online shopping, delivery food ordering, etc., it is now possible to make phone call and voice payment.

However, the service provided by the AI speaker is still limited to the home service. According to a recent survey by US Internet companies, customers of AI speakers use AI speakers mainly for general questions, weather, stream music, time / alarms, schedule checks, home automation, stream news, games, and shopping. do.

Smart diagnostic apparatus, system and method of artificial intelligence function according to an embodiment of the present invention, by diagnosing the state of the individual user through a conversation with the artificial intelligence speaker provided in the home and protects the user or the information about the user state It is characterized by providing a service that can notify the terminal of the administrator. In this regard, the smart diagnostic apparatus, system, and method for an artificial intelligence function according to an embodiment of the present invention are distinguished from the above-described prior art and disclosed to solve the problems of the prior art.

Korean Unexamined Patent Publication No. 10-2004-0007003 (2004.01.24.)

An object of the present invention is to provide an apparatus, system, and method for providing a patient diagnosis and monitoring service of an artificial intelligence function based on voice and image information.

An object of the present invention is to further solve, an artificial intelligence smart diagnostic device, system and method for analyzing and diagnosing an individual's condition through the speaker of the artificial intelligence function, and notifying analysis information and diagnostic information to a terminal such as a guardian. To provide.

The present invention to further solve the problem, through the user's voice and the image taken by the user through the speaker with the artificial intelligence shooting function, in particular to analyze the condition of the patient needing protection, remote diagnosis of hospitals and public health centers The present invention provides a smart diagnostic apparatus, system, and method of an artificial intelligence function that can diagnose through a device and notify analysis information and diagnostic information to a terminal such as a guardian.

Smart diagnostic device of the artificial intelligence function according to an embodiment of the present invention, the artificial intelligence assistant, which collects audio and video data about the user's status information, and outputs the synthesized voice required to lead a conversation with the user ( a data input / output module having an intelligent personal assistant function; A data analysis module configured to analyze a conversation period with a user, a change in a user's voice, and a conversation content using the collected data, and output analysis data; And a diagnostic module for diagnosing a changed state of the user using the analysis data.

An intelligent personal assistant according to an embodiment of the present invention collects audio and video data regarding status information of a user and outputs a synthesized voice required to direct a conversation with the user. A data analysis module configured to analyze a conversation period with a user, a change in a user's voice, and a conversation content by using the data collected through a data input / output device having a function, and output analysis data; And a diagnostic module for diagnosing a changed state of the user using the analysis data.

An intelligent personal assistant according to an embodiment of the present invention collects audio and video data regarding status information of a user and outputs a synthesized voice required to direct a conversation with the user. A function data input / output device; A data analysis device configured to analyze a conversation period with a user, a change in a user's voice, and a conversation content using the collected data, and output analysis data; And a diagnostic device for diagnosing a changed state of the user using the analysis data.

Here, the data input / output module includes a camera for detecting a facial expression of a user, color, and a user's behavior and capturing it as an image, and the data analyzing module includes a user's appearance using image data input through the camera. And analyzing the user's behavior and outputting the analysis data.

The smart diagnostic method according to an embodiment of the present invention is a diagnostic method performed by a diagnostic apparatus for diagnosing a user's condition based on voice or image data collected with respect to the user's status information. collecting the data through a conversation with a user utilizing an assistant function; Analyzing the conversation period with the user, the change in the user's voice and the conversation content using the collected data, and outputting the analysis data; And diagnosing a state of the user based on the analysis data.

The smart diagnostic method according to an embodiment of the present invention is a diagnostic method performed by a diagnostic apparatus for diagnosing a user's condition based on voice or image data collected with respect to the user's status information. receiving the data collected through a dialogue with a user utilizing an assistant function; Analyzing the conversation period with the user, the change in the user's voice and the conversation content using the collected data, and outputting the analysis data; And diagnosing a state of the user based on the analysis data.

The collecting of the data may further include capturing image data of a user's facial expression, color, and user's behavior, and outputting the analysis data may include using the image data. Analyze the appearance and behavior of the user, and outputs the analysis data.

The receiving of the data may include receiving image data in which a facial expression of a user, a color, and a user's behavior are detected. The outputting of the analysis data may include a user's appearance using the image data. And analyzing the user's behavior and outputting the analysis data.

The outputting of the analysis data may include outputting analysis data necessary for diagnosing the mental illness of the elderly by analyzing the time of occurrence of the conversation, the time required, the change of the user's voice and the contents of the conversation, and the facial expression and color of the user. And outputting analysis data necessary for diagnosing a trauma, a skin disease, an endocrine, and a disease related to the circulatory system through the user's behavior detection and image analysis.

Here, the smart diagnostic method, characterized in that it further comprises the step of transmitting the analysis data or diagnostic results to the terminal of the guardian and the government office to protect or manage the user according to the analysis or diagnosis results.

The diagnosing of the user's state may include diagnosing the user's state through a remote medical server of a hospital or a public health center corresponding to the diagnosis apparatus using the received analysis data.

According to the present invention, it is possible to diagnose and monitor a patient who has fallen into a space using a data input / output device having an artificial intelligence function based on voice and image information.

In addition, the data input / output device having an artificial intelligence function may analyze and diagnose a changed state of an individual, and may notify analysis information and diagnostic information to a terminal such as a guardian.

In addition, based on the user's voice and images taken by the user through the artificial intelligence data input and output device, in particular, the condition of the patient in need of protection is analyzed, and diagnosed by the remote diagnosis device of the hospital and public health center, and the analysis information and The diagnostic information can be informed to the terminal such as a guardian.

1 is an exemplary diagram of a smart diagnostic apparatus of the artificial intelligence function according to an embodiment of the present invention.
1 is an exemplary diagram of a smart diagnostic apparatus of the artificial intelligence function according to an embodiment of the present invention.
3 is an exemplary view of a smart diagnostic apparatus of the artificial intelligence function according to an embodiment of the present invention.
4 is an exemplary diagram of a smart diagnostic apparatus of an artificial intelligence function according to an embodiment of the present invention.
5 is a block diagram of the apparatus shown in FIG. 4.
6 is a flowchart illustrating a smart diagnosis method of an artificial intelligence function according to an embodiment of the present invention.
7 is a block diagram illustrating an audio and video analysis process according to an embodiment of the present invention.

Hereinafter, exemplary embodiments of a smart diagnosis apparatus, a system, and a method of an artificial intelligence function will be described in detail with reference to the accompanying drawings.

Like reference numerals in the drawings denote like elements. In addition, specific structural to functional descriptions of one embodiment of the present invention are only illustrated for the purpose of describing the embodiment according to the present invention, and unless otherwise defined, all terms used herein including technical or scientific terms are used. The terms have the same meaning as commonly understood by one of ordinary skill in the art. Terms such as those defined in the commonly used dictionaries should be construed as having meanings consistent with the meanings in the context of the related art, and are not construed in ideal or excessively formal meanings unless expressly defined herein. It is preferable not to.

Hereinafter, a smart diagnosis apparatus and system of an artificial intelligence function according to an embodiment of the present invention will be described.

Smart diagnostic apparatus 100 and system 1000 according to an embodiment of the present invention is a means for collecting data for diagnosis 200, the means for analyzing the collected data (300, 400) for the purpose of remote diagnosis and prescription And means 100 for making a diagnosis and prescription using the analyzed data and a database 600 for diagnosis. Collected data includes audio and video data. In one embodiment according to the present invention, the means 300 for processing and analyzing voice data and the means 400 for processing and analyzing image data may be shared or independently configured. In addition, the means 100 to 400 and the database 600 are connected through the network 700, and means other than the means 200 for collecting data for diagnosis among the devices and the database 600 are located at the same place. It can be located at or in separate locations. Hereinafter, specific configurations of the respective means 100 to 400 and the database 600 will be described.

1 and 2 are exemplary diagrams of a smart diagnostic apparatus of an artificial intelligence function according to an embodiment of the present invention.

Referring to FIG. 1, the smart diagnostic apparatus 100 having an artificial intelligence function according to an embodiment of the present invention may include a data input / output module 110, a data analysis module 120, and a diagnostic module 130 as its components. It includes. In addition, the smart diagnosis apparatus 100, the management server or the user terminal 10, and the database 600 of the artificial intelligence function are connected through the network 700 to enable communication with each other.

In the embodiment shown in FIG. 1, the smart diagnostic apparatus 100 starts with a process of collecting voice data and image data about a user, analyzes the data, and diagnoses a user's state using the analysis data. It is characterized by processing. As the processing power of the processor performing the calculation is improved, the smart diagnostic apparatus 100 illustrated in FIG. 1 may be implemented using an artificial intelligence speaker equipped with a camera and a PC.

The smart diagnostic apparatus 100 may use a database 600 included in itself or a database 600 connected through a network. The database 600 includes data necessary for speech recognition, speech synthesis, image analysis, and condition diagnosis using artificial intelligence.

Referring to FIG. 2, the smart diagnostic apparatus 100 according to an embodiment of the present invention may include an independent data input / output device 200 corresponding to the data input / output module 110 as compared with the smart diagnostic apparatus 100 of FIG. 1. It may be configured not to include. Therefore, the smart diagnostic apparatus 100 of FIG. 2 receives the audio and image data collected from the data input / output device 200.

In the embodiment shown in FIG. 2, the smart diagnosis apparatus 100 collects voice data and image data about a user from an independent device 200, and analyzes the data, and diagnoses the user's state using the analysis data. Characterized in that the processing process. As the processing power of the processor performing the operation is improved, the data input / output device 200 including the microphone, the speaker, and the camera illustrated in FIG. 2 may be implemented in the form of a client, and the smart diagnostic apparatus 100 may be implemented in the form of a server. .

The smart diagnostic apparatus 100 may use a database 600 included in itself or a database 600 connected through a network. The database 600 includes data necessary for speech recognition, speech synthesis, image analysis, and condition diagnosis using artificial intelligence.

3 and 4 are exemplary diagrams of a smart diagnosis system of an artificial intelligence function according to an embodiment of the present invention.

Referring to FIG. 3, the smart diagnosis system 1000 of the artificial intelligence function according to an embodiment of the present invention may include a diagnosis apparatus 100, a data input / output device 200, a data analysis apparatus 500, and a database 600. And network 700.

Referring to FIG. 4, the smart diagnostic system 1000 of the artificial intelligence function according to an embodiment of the present invention may be compared with the configuration of FIG. 3, instead of the data analyzing apparatus 500, the voice analysis apparatus 300 and the image analysis. Device 400. Hereinafter, the configuration modules and the configuration apparatuses corresponding to the data input / output means, the data analysis means, and the diagnostic means shown in FIGS. 1 to 4 will be described with reference to FIG. 4.

Referring back to FIG. 4, the data input / output device 200 communicates with the user while acting as one entity of conversation using an artificial intelligence function, observes the user's change through the conversation, and changes the user's state. It collects audio and video information from the user. In order to collect audio information, the data input / output device 200 receives a voice in a conversation with a user and performs a function of outputting a synthesized voice in response to the voice. In order to collect video information, the data input / output device 200 performs a function of capturing an image using a camera during a conversation with a user.

The image analyzing apparatus 400 outputs analysis data, that is, image data, required for diagnosis by using the collected image, and transmits the analysis data to the diagnostic apparatus.

The speech analyzing apparatus 300 generates a natural language necessary for dialogue through speech recognition, and similarly outputs analysis data necessary for diagnosis, that is, audio data and transmits the same to the diagnostic apparatus.

The smart diagnostic apparatus 100 performs a role of diagnosing a user's state by using analysis data, that is, image and audio data. Here, the database 600 provides data about various diseases required for diagnosis. In particular, the smart diagnostic apparatus 100 may include a diagnostic server for performing a direct diagnosis and a user terminal indirectly assisting the diagnosis by a doctor.

The network 700 may include wired and wireless networks, such as the Internet, intranets and extranets, cellular systems, such as 3G mobile, GSM, WCDMA, CDMA 2000, LTE, 5G mobile, LAN, WAN, WiFi networks, It may be any suitable communication network, including ad hoc networks and combinations thereof.

Network 700 may include a connection of network elements such as hubs, bridges, routers, switches, and gateways. Network 700 may include one or more connected networks, such as a multi-network environment, including a public network such as the Internet and a private network such as a secure corporate private network. Access to the network 700 may be provided through one or more wired or wireless access networks. Hereinafter, specific configurations of the data input / output device 200, the voice analysis device 300, and the image analysis device 400 will be described.

5 is a block diagram of each device included in the smart diagnostic system of the artificial intelligence function of FIG.

Referring to component 200 in FIG. 5, the data input / output device 200 includes a communication interface 210, an audio input / output interface 220, a voice synthesis module 230, and a camera module 240.

The data input / output device 200 is responsible for input / output of data regarding user voice information and input of image data regarding user image information. The user voice information is a result of a conversation process between the user and the data input / output device 200, for example, a preprocessing step, a voice recognition step, and voice recognition of a voice spoken by an elderly person living alone who is to be protected and observed. Is output as a text, after which the output text is interpreted as a natural language, and a response thereto is processed through the conversation management, natural language generation, and translation steps, and the completed text response is transmitted through the data input / output device 200 through speech synthesis. Can be collected through the output step.

The data input / output device 200 may perform a home gateway function using the communication interface 210. Here, the home gateway refers to a kind of hub that connects a home LAN to an external network, for example, the Internet. The data input / output device 200 may communicate with various other devices through the communication interface 210. The data input / output device 200 may be implemented in the form of an artificial intelligence speaker 200 equipped with a camera.

In addition, the data input / output device 200 may perform a router function using the communication interface 210. That is, the data input / output device 200 may provide a wireless LAN service such as Wi-Fi to the communication terminal of the customer through the communication interface 210.

The data input / output device 200 receives a user's voice through the audio input / output interface 220 and outputs a synthesized voice. The audio input / output interface 220 may be implemented as a microphone and a speaker, and the microphone may receive a voice and change it into an electrical signal, and the electrical signal may be streamed or stored as a file and transmitted to the voice analysis apparatus 300. .

The data input / output device 200 receives a text response from the voice analysis device 300 to complete a response required for a conversation with a user, and uses the speech synthesis module 230 to convert a text response required for a customer's conversation into voice. The converted voice is output through the voice input / output interface 220.

The communication interface 210 includes devices for short range wireless communication, wireless data communication, and wireless voice communication in addition to wired data communication. Specifically, the communication interface 210 includes a communication module corresponding to various communication networks of the network 700, for example, a Bluetooth module, a Wi-Fi module, an Ethernet interface card, a USB module, a cellular wireless communication module, a modem, and a wireless router. However, it is not limited thereto.

Referring to component 400 in FIG. 5, the image analysis apparatus 400 may include an image preprocessing module 410, a storage device 420, an image analysis module 430, a communication device 440, an input device 450, and an output. Device 460. Hereinafter, the image analyzing apparatus 400 according to an exemplary embodiment will be described.

The image preprocessing module 410 performs preprocessing on the digitally converted image data. The preprocessing includes various image processing processes such as brightness adjustment, gamma value adjustment and binarization.

The storage device 420 stores image data captured by the camera module 240 and analysis data corresponding to a result of the analysis.

The image analysis module 430 detects a user's facial expression, color, and user's behavior using preprocessed image data, and analyzes data necessary for diagnosing a user's skin disease, endocrine, and circulatory disease through image analysis. Outputs

The communication device 440 includes a device for short range wireless communication, wireless data communication, and wireless voice communication in addition to wired data communication. In detail, the communication device 440 includes a communication module corresponding to various communication networks of the network 700, for example, a Bluetooth module, a Wi-Fi module, an Ethernet interface card, a USB module, a cellular wireless communication module, a modem, and a wireless router. However, it is not limited thereto.

The input device 450 receives various commands for analyzing image information according to a user's input. The input device 450 may be implemented by a keyboard, a touch screen, a mouse, an electronic pen, a bar code input device, and the like, but is not limited thereto.

The output device 460 may be implemented by a display device for outputting an image information analysis process or the like. Hereinafter, the speech analysis apparatus 300 will be described.

Referring to component 300 in FIG. 2, the speech analysis apparatus 300 includes an artificial intelligence and natural language processing module 310, a speech analysis module 320, and a data / knowledge base 330. Here, the artificial intelligence and natural language processing module 310 may use the published artificial intelligence and natural language processing API.

The preprocessing module 311 performs a voice preprocessing process of removing noise from the input sound, extracting a human voice, and amplifying and converting it. The voice recognition module 312 performs a voice recognition function of converting the voice signal received from the preprocessing module 310 into text. The natural language interpretation module 313 interprets the meaning of the natural language included in the text. The dialogue manager 314 determines the best dialogue strategy for the intention of the user speech in consideration of the dialogue flow and the situation to generate the meaning and expression of the system speech. The natural language generation module 315 then generates a natural language text response including the meaning and expression generated by the conversation manager 314 as a response to the natural language. Here, the translation module 316 may be provided to serve to translate natural language text responses into foreign languages.

The voice analysis module 320 analyzes the conversation period with the user, the change in the user's voice, and the conversation content based on the conversation content of the user. The voice analysis device 300 outputs analysis data necessary for diagnosing mental and elderly diseases of the user through the analysis result.

The data / knowledge base 330 includes learning data necessary for recognizing the user's voice and analyzing the user's voice using an artificial intelligence algorithm.

6 is a flowchart illustrating a smart diagnosis method of an artificial intelligence function according to an embodiment of the present invention.

Referring to FIG. 6, the smart diagnostic method of an artificial intelligence function according to an embodiment of the present invention includes steps S110 to S150. And each step is characterized in that performed by the configuration device included in the smart diagnostic device 100 or system 1000 of the artificial intelligence function according to an embodiment of the present invention depicted in FIG.

That is, each step constituting the smart diagnostic method (S100) of the artificial intelligence function is one device 100, two devices (100, 200), three devices (100, 200, 500) or four devices It is characterized in that performed by the (100, 200, 300, 400). Hereinafter, a smart diagnostic method performed by four devices will be described as an example.

First, voice data spoken by a user is collected through a conversation with a user using a data input / output device 200 having an intelligent personal assistant function (S110). The data input / output device 200 distinguishes between the general noise and the user's voice by using an artificial intelligence algorithm, records the user's voice in the always-recording mode, and the recorded audio data is stored in the voice analysis device 300. The full-fledged conversation in which the user and the data input / output device 200 alternately utter a voice is initiated by the operation of a physical key or a driving command by the user.

In addition, the user's facial expression, color, and user's behavior are detected through the camera, and an image related thereto is captured (S120). Here, in steps S110 and S120, both steps may be performed without a subsequent order, or at the same time or only one step may be performed. By presetting, the data input / output device 200 may periodically collect audio and video data of the user through greetings and conversations with the user at a predetermined time, for example, after waking up or after breakfast. However, video data about a user may be limited in time and space because the user must be in a position where the camera module 240 can capture the user's view.

Next, the dialogue period with the user, the change of the user's voice, the contents of the conversation are analyzed by using the voice data collected through the voice and video analysis apparatuses 300 and 400, and the analysis data is output (S130). In addition, the user's appearance and user's behavior are analyzed using the image data input through the camera module 240, and the analysis data is output. Here, the user's voice change, conversation contents, user's appearance, and user's behavior analysis are performed through a model trained through an artificial intelligence algorithm based on the user's voice and image data collected in the past.

In addition, analysis data and an alarm signal may be transmitted to the guardian terminal protecting the user and the manager terminal 10 managing the user according to the analysis result through the voice and projection analysis apparatus 300 or 400 (S140).

Next, the changed state of the user is diagnosed based on a database of diseases established in the Internet portal server or the disease management server using the analysis data (S150). The analysis data includes a result of analyzing audio and video about a user. For example, the analysis data may include changes in the user's voice, degree of voice change, characteristics or changes in word selection during conversation, whether or not the utterance is in conformity with the grammar compared to the usual utterance, the change in the facial expression of the user, the change in the color, It includes data on the presence or absence of a trauma to the face or body, changes in the body, singularity of the user's behavior compared to the existing behavior.

Here, the step of outputting the analysis data (S130), characterized in that to output the analysis data necessary for the diagnosis of mental and elderly diseases of the user through the analysis of the conversation period, the change of the user's voice and the conversation content with the user do. For example, it is possible to advance whether or not the user invented dementia based on a change in a language, a change in a tone, a change in a speech rate, etc. used by the user.

Here, the step of outputting the analysis data (S130), through the detection and image analysis of the user's facial expression, color, and the user's behavior outputs the analysis data necessary for diagnosing the user's skin disease, endocrine and diseases related to the circulatory system Characterized in that. For example, an endocrine disorder may be diagnosed based on a changed color of a user, or a physical infringement due to an accident may be diagnosed based on a trauma or physical change found in a user's body.

In particular, the step (S150) of diagnosing the user's condition may include directly diagnosing the user's condition through a remote medical server of a hospital or a public health center using the received audio and image analysis data; And indirectly diagnosing the state of the user based on the decision of the doctor through the user terminal. For example, the analysis data is input to the diagnostic server, and based on the artificial intelligence algorithm, the diagnosis of dementia is automatically or directly probable by the diagnostic server based on the AI algorithm, or the analysis data is transmitted to the user terminal possessed by a doctor. In addition, whether or not the invention of dementia can be indirectly advanced by the judgment of the physician based on the analysis data.

The method of processing data using a learned machine called machine learning or artificial intelligence may be implemented by hardware and software means. Hardware and software means include neural networks, cloud computing, artificial intelligence algorithms and microprocessors.

Hereinafter, an artificial intelligence function performed in an embodiment of the present invention will be described.

7 is a block diagram illustrating an audio and video analysis process according to an embodiment of the present invention.

Referring to FIG. 7, a data processing and data analysis process performed by the data input / output device 200, the image analysis device 400, and the voice analysis device 300 is illustrated.

The data input / output device 200 converts a voice signal input through the audio input / output interface 220 into an electrical signal, transmits it to the voice analysis device 300, and outputs a response composing a conversation. ) Converts the text response received from the speech analysis apparatus 300 into speech and outputs the speech.

The speech synthesis module 230 converts the natural language text response into a voice of a preset tone and outputs it. In particular, the voice synthesis module 230 may be included in the data input / output device 200. However, the voice synthesis module 230 may be included in the voice analysis device 300 together with the artificial intelligence and natural language processing module 310. .

The camera module 240 collects image information from the data input / output device 200. A plurality of camera modules 240 may be provided. For example, a first camera may be provided in the data input / output device 200 to capture a face of a user in a conversation, and a second camera may be provided on a wall or a ceiling to observe a user's behavior. The second camera may communicate with the data input / output device 200 and transmit data through a wireless or wired connection. Here, the first camera may be a camera module used in a smartphone, and the second camera may be implemented as an RTZ camera capable of rotating, tilting and zooming.

The speech analysis apparatus 300 may include a speech recognition server and a natural language processing server for each function. The speech recognition server recognizes a speech using an artificial intelligence algorithm based on the speech data received from the data input / output device 200, outputs a speech recognition result in a text form, and transmits the result to the conversation natural language processing server. The natural language processing server outputs the text response through natural language interpretation, conversation management, natural language generation and translation using the speech recognition result. Artificial intelligence algorithms are also used in natural language interpretation, conversation management, and natural language generation. The output text response is transmitted to the data input / output device 200. Although the speech synthesis process may be performed by the server, the data input / output device 200 may receive a text response having a smaller capacity than the speech file, and synthesize the speech response and output the speech response.

In addition, the speech synthesis, which does not have a relatively large computing capacity, is performed by the data input / output device 200 rather than the server, and as a result, the synthesized voice may be finally output through the speaker of the data input / output device 200.

The data / knowledge base 330 may be provided by using another server, and the data / knowledge base 330 may analyze the natural language mechanically and make it in a form that can be understood by a computer, or by understanding a computer processing result. It may include various engines, dictionaries, and knowledge bases related to natural language processing, which is a technique for making a language.

The process of analyzing video data may include an image preprocessing process 410 and an image analysis process 430. The image preprocessing process includes image processing operations for removing noise from an image of a photographed user and various kinds of corrections.

In the image analysis process, the database 440 necessary for image analysis may be utilized. The database 440 includes historical image data of the user. Image analysis required for diagnosis may be performed by comparing the image data of the users recorded at different times with each other or by comparing the image of the user with an image related to a disease.

The process performed by the diagnostic apparatus includes a diagnosis process by the diagnosis unit 410 and a prescription process by the prescriber 430. The database 600 includes data on diseases necessary for diagnosis, for example, data of the disease management headquarters, and various data about diseases of the port site.

The speech preprocessing, speech recognition, natural language analysis, conversation management, natural language generation and translation services provided by the speech analysis apparatus 300 are characterized by the use of artificial intelligence AIP and natural language processing API. Here, a published API may be used to process operations related to machine learning or deep learning.

In the configuration of the data input / output device 200, the image analysis device 400, and the audio analysis device 300 illustrated in FIG. 2, each device or module is a functionally divided configuration for convenience of description. It can be configured as a logical function to be processed by one processor, not limited to the present invention by the presented division.

In addition, although the connection relationship between each configuration is shown as a connection line, or the connection relationship between each other is not expressed individually, communication or transmission for control or data exchange may occur even between the configuration where the connection line is not shown, and only as shown. It does not limit the invention.

The human voice input through the data input / output device 200 is defined as one sound. And the three elements that characterize sound are frequency, amplitude, and timbre. Frequency is related to the pitch of the note, amplitude is related to the loudness, and tone is one of the characteristics of the note.

Thus, when the input speech signal is stored in a data file through digital conversion, that is, sampling, quantizing, and encoding, a model for identifying the speech of the usage is generated, The user's characteristics are learned to the model through the reproduction of the stored voice file using the artificial intelligence learning function. Using this process, the user's voice can be recognized using artificial intelligence, and a database on the recognized user's personal information can be constructed. It is important here that the characteristics of the human voice are not limited to the timbre, but are all in frequency, amplitude, and timbre, so that the characteristics of the frequency, amplitude, and timbre can be used for user recognition by the speech.

Smart diagnostic method (S100) of the artificial intelligence function according to an embodiment described through the drawings, is implemented in the form of a recording medium including a command set executable by a computer, such as a program module executed by a computer Can be. Computer readable media can be any available media that can be accessed by a computer and includes both volatile and nonvolatile media, removable and non-removable media. In addition, computer readable media may include both computer storage media and communication media. Computer storage media includes both volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data. Communication media typically includes computer readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave, or other transmission mechanism, and includes any information delivery media.

Dementia management policy programs promoted by the government and local governments include early detection and prevention, tailored care and protection, infrastructure expansion, family support and social communication. In the case of the smart diagnostic system and method of the artificial intelligence function according to an embodiment of the present invention, it is possible to help in early detection and prevention of dementia among the above programs, and effective in expanding family support and social communication. do.

It is possible to detect dementia early by detecting language features and abnormal behavior through dialogue with the user, and to prevent dementia by talking to the user through artificial intelligence. In the case of elderly dementia living remotely from children, it is possible to expand family support and social communication through the monitoring function and the dialogue function according to the smart diagnosis system and method of the artificial intelligence function according to an embodiment of the present invention.

As described above, according to an exemplary embodiment of the present disclosure, a patient who has fallen into a distance can be diagnosed and monitored using an input / output device having an artificial intelligence function based on voice and image data.

In addition, an I / O device using an artificial intelligence function may be used to analyze and diagnose an individual's state and inform the terminal such as a guardian of the analysis information and the diagnostic information.

In addition, based on the user's voice and the image taken by the user through the input and output device using the artificial intelligence function, in particular, analyzes the condition of the patient needing protection, and diagnoses through the remote diagnosis device of the hospital and public health center, analysis information and The diagnostic information can be informed to the terminal such as a guardian.

The foregoing description of the present invention is intended for illustration, and it will be understood by those skilled in the art that the present invention may be easily modified in other specific forms without changing the technical spirit or essential features of the present invention. will be. Therefore, it should be understood that the embodiments described above are exemplary in all respects and not restrictive. For example, each component described as a single type may be implemented in a distributed manner, and similarly, components described as distributed may be implemented in a combined form.

The scope of the present invention is shown by the following claims rather than the above description, and all changes or modifications derived from the meaning and scope of the claims and their equivalents should be construed as being included in the scope of the present invention. do.

100: smart diagnostic device
200: data input / output device
300: speech analysis device
400: video analysis device
500: data analysis device
600: database
700: network
1000: smart diagnostic system

Claims (11)

A data input / output module of an intelligent personal assistant function, which collects audio and video data on status information of a user using a camera, a microphone, and a speaker, and outputs synthesized voice for use in a conversation with the user;
A data analysis module for analyzing the collected data; And
A diagnostic module for diagnosing a changed state of a user using the analysis data,
The data analysis module includes voice analysis means and image analysis means.
The voice analysis means uses the synthesized voice output from the data input / output module of the artificial intelligence assistant function to change the language, tone change and speech rate used by the user received through the dialogue with the user. Analyze,
The image analyzing means analyzes the user's facial expression, color, and user's behavior,
And when the analysis result corresponds to a preset criterion, transmitting the analysis result or an alarm signal to a guardian of the user.
The method of claim 1,
The data input / output module, the data analysis module, and the diagnostic module of the artificial intelligence function are separately configured, and the smart diagnostic device, characterized in that interconnected via a network.
delete delete delete delete delete delete delete delete delete

Images