Intelligent accompanying diagnosis system and method based on AI and storage medium

Through the AI-based smart accompanying system, intelligent robots and interactive terminals provide automatic navigation and auxiliary services, it solves the problems of difficulty in seeking medical treatment and queue leader during peak medical treatment, and improves the effectiveness of medical accompanying and user experience.

CN118942656AInactive Publication Date: 2024-11-12THE UNIVERSITY OF HONG KONG SHENZHEN HOSPITAL

Patent Information

Application Number
CN202410983586.1
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-22
Publication Date
2024-11-12
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

Medical institutions often experience difficulties in seeking medical treatment and queuing up at peak times. The lack of accompanying patients during the treatment process leads to overloading the medical system, and the time arrangement of family members or friends cannot always be with the patient.

Method used

Adopting an AI-based smart accompanying system, including a movable robot body and interactive terminal, the user and the service terminal can be interacted and navigation through the command recognition module and navigation module, providing automatic accompanying services, combining voice and text recognition modules , laser scanning and robotic arms technologies to realize intelligent navigation and auxiliary operations.

Benefits of technology

It effectively solves the problems of difficulty in seeking medical treatment and queuing up, improves the effectiveness of medical care, reduces the work burden of outpatient nurses, provides convenient medical treatment process and personalized interactive popular science, and enhances users' health awareness and medical experience.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118942656A_ABST
    Figure CN118942656A_ABST
Patent Text Reader

Abstract

The invention discloses an AI-based intelligent accompanying diagnosis system and method, and a storage medium. The AI-based intelligent accompanying diagnosis system comprises a service terminal; the diagnosis accompanying robot comprises a robot body and an interaction terminal arranged on the robot body, and the interaction terminal is in communication connection with the service terminal; the interaction terminal comprises an instruction identification module and a navigation module; the instruction identification module is used for receiving and analyzing an interaction instruction issued by a user to the interaction terminal, so that the interaction terminal communicates with the service terminal and / or displays interaction information to the user according to an analysis result; the navigation module is used for receiving an inquiry instruction issued by a user to the interaction terminal and generating an accompanying diagnosis route according to the inquiry instruction; the robot body comprises a control module used for controlling the robot body to move, and the control module is communicated with the interaction terminal. The method solves the problems of difficult medical treatment and long queuing in a medical accompanying diagnosis mode, and overcomes the defects of lack of accompanying diagnosis and the like.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of medical care technology, and in particular to an AI-based intelligent accompanying medical system, method and storage medium. Background Art

[0002] At present, people's pursuit of healthy quality of life is gradually improving, which has led to an increasing number of patients visiting various medical institutions. Due to the high randomness of the time of patient arrival and consultation and medication, medical institutions often face difficulties in seeing a doctor and long queues during peak consultation periods, which leads to the medical system being overloaded for a long time. In addition, in the existing medical environment, patients often need someone to accompany them during the medical consultation process to provide help when needed. However, due to various reasons, such as the schedule of family members or friends, they may not always be with the patient. Summary of the invention

[0003] The embodiments of the present invention provide an AI-based intelligent medical accompanying system, method and storage medium, which aim to solve the problems of difficulty in seeing a doctor and long queues through medical accompanying, overcome the defects such as the lack of medical accompanying, and improve the effect of medical consultation.

[0004] In a first aspect, an embodiment of the present invention provides an AI-based intelligent medical companion system, comprising:

[0005] Service terminal;

[0006] The accompanying robot comprises a movable robot body and an interactive terminal arranged on the robot body, wherein the interactive terminal is communicatively connected with the service terminal;

[0007] The interactive terminal includes an instruction recognition module and a navigation module; the instruction recognition module is used to receive and analyze the interactive instructions sent by the user to the interactive terminal, so that the interactive terminal communicates with the service terminal and / or displays interactive information to the user according to the analysis result; the navigation module is used to receive the consultation instruction sent by the user to the interactive terminal, and generate a medical consultation route according to the consultation instruction;

[0008] The robot body includes a control module for controlling the movement of the robot body, and the control module is connected to the interactive terminal.

[0009] In a second aspect, an embodiment of the present invention provides an AI-based intelligent accompanying diagnosis method, which is applied to the AI-based intelligent accompanying diagnosis system as described in the first aspect, including:

[0010] In response to the interactive instruction issued by the user, the interactive terminal parses and identifies the interactive instruction through the instruction recognition module, obtains the interactive analysis result, and interacts with the service terminal according to the interactive analysis result, and / or returns interactive information to the user; wherein the interactive instruction includes a medical consultation instruction;

[0011] When the interactive instruction is a medical consultation instruction, the interactive terminal sends the interactive analysis result of the medical consultation instruction to the navigation module, and the navigation module generates a medical consultation route according to the interactive analysis result;

[0012] The interactive terminal obtains the medical consultation route generated by the navigation module and returns it to the user;

[0013] In response to the automatic driving instruction issued by the user based on the accompanying route, the interactive terminal sends the accompanying route to the control module of the robot body, and the control module controls the robot body to drive automatically according to the accompanying route.

[0014] In a third aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the AI-based intelligent accompanying medical consultation method as described in the second aspect is implemented.

[0015] The embodiments of the present invention provide an AI-based intelligent accompanying medical system, method and storage medium. The embodiments of the present invention can solve the problems of difficulty in seeing a doctor and long queues by accompanying the patient in medical consultation. At the same time, the system can interact with the user through an interactive terminal and use a movable accompanying robot to accompany the user in medical consultation, thereby improving the effect of medical accompanying. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying any creative work.

[0017] Figure 1 A system architecture diagram of an AI-based intelligent accompanying medical system provided in an embodiment of the present invention;

[0018] Figure 2 A flowchart of an AI-based intelligent accompanying diagnosis method provided by an embodiment of the present invention;

[0019] Figure 3 A sub-process diagram of an AI-based intelligent medical accompanying method provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0020] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0021] It should be understood that when used in this specification and the appended claims, the terms "include" and "comprises" indicate the presence of described features, integers, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or combinations thereof.

[0022] It should also be understood that the terms used in this specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the specification of the present invention and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

[0023] It should be further understood that the term "and / or" used in the present description and the appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0024] See below Figure 1 The embodiment of the present invention provides an AI-based intelligent accompanying medical system, comprising:

[0025] Service terminal 100;

[0026] The accompanying robot comprises a movable robot body 200 and an interactive terminal 300 disposed on the robot body 200, wherein the interactive terminal 300 is communicatively connected with the service terminal 100;

[0027] The interactive terminal 300 includes an instruction recognition module and a navigation module; the instruction recognition module is used to receive and analyze the interactive instructions issued by the user to the interactive terminal 300, so that the interactive terminal 300 communicates with the service terminal 100 and / or displays interactive information to the user according to the analysis result; the navigation module is used to receive the consultation instruction issued by the user to the interactive terminal 300, and generate an accompanying consultation route according to the consultation instruction;

[0028] The robot body 200 includes a control module for controlling the movement of the robot body 200 , and the control module is connected to the interactive terminal 300 .

[0029] In this embodiment, the AI-based intelligent accompanying medical system includes a service terminal 100 and an accompanying medical robot, wherein the accompanying medical robot includes a robot body 200 provided with a control module and an interactive terminal 300, such as Figure 1 As shown, the robot body 200 is specifically an intelligent wheelchair. When the user issues an interactive instruction to the interactive terminal 300, the interactive terminal 300 first parses and identifies the received interactive instruction through the instruction recognition module to obtain an interactive analysis result. For the interactive analysis result, the interactive terminal 300 can send it to the service terminal 100, and then the service terminal 100 returns the corresponding interactive content, or directly returns the corresponding interactive content to the user. For example, when the user expects to obtain medical knowledge or interact with medical personnel through the interactive terminal 300, the interactive terminal 300 can send the interactive analysis result to the service terminal 100, so that the corresponding medical knowledge or the medical personnel's reply is returned to the user through the service terminal 100. For example, when the user expects to view the medical personnel information or the outpatient location through the interactive terminal 300, the interactive terminal 300 can return the corresponding query result to the user based on the locally pre-stored information.

[0030] There can be many interactive instructions in this embodiment, and considering that this embodiment is an AI-based intelligent medical companion system, the most commonly used interactive instruction is the medical consultation instruction. Therefore, when the interactive terminal 300 receives the medical consultation instruction issued by the user, it can parse and identify the medical consultation instruction according to the instruction recognition module, and then send the interactive analysis result to the navigation module. The navigation module can extract the corresponding medical consultation information from the interactive analysis result, and thus generate a corresponding navigation route according to the medical consultation information. After obtaining the navigation route, the user can choose whether to be carried to the destination by the medical companion robot. When the user chooses to be carried by the medical companion robot, the interactive terminal 300 will send corresponding information to the control module on the medical companion robot, such as sending the medical companion route and driving instructions, and then the control module will control the robot body 200 to drive automatically, so as to carry the user to the medical consultation location.

[0031] This embodiment uses the interactive terminal 300 to conduct interactive questions and answers with the user, and uses a movable accompanying robot to accompany the user to consult and see a doctor, thereby improving the effect of medical accompanying. It can be understood that the interactive terminal 300 is provided with a wireless communication module for sending and receiving data, so that the interactive terminal 300 can communicate with the service terminal 100, the control module, etc. In actual application scenarios, video communication software can also be set on the interactive terminal 300, so that after the interactive terminal 300 is connected to the Internet, the user can communicate with other people through video or voice through the video communication software of the interactive terminal 300, such as medical personnel, family members, etc., which can improve the efficiency of consultation.

[0032] In particular, based on the service terminal 100 and the interactive terminal 300 provided in this embodiment, the Internet of Things technology can also be used to export the user's vital signs data, including body temperature, blood pressure, pulse, respiration and heart rate, to the electronic medical record system, which can greatly reduce the workload of outpatient nurses and allow them to focus more on providing high-quality medical services. At the same time, users can also easily complete payment on the interactive terminal 300, and obtain call reminders in time through the interactive terminal 300 to enjoy a convenient medical treatment process. In addition, by deeply mining the big data of the interactive terminal 300, it is possible to accurately understand the needs and preferences of users, and then during the waiting period, personalized interactive popular science content can be provided to users based on this information, which not only enhances the user's health awareness, but also greatly enriches the user's medical experience. This user-centered service model closely combines technology with humanistic care, bringing users more intimate and efficient medical services.

[0033] In a specific embodiment, the instruction recognition module includes a speech recognition module and a text recognition module;

[0034] The voice recognition module is used to receive the user's voice interaction instructions, and the text recognition module is used to receive the user's text interaction instructions.

[0035] In order to efficiently and accurately receive instructions issued by users, this embodiment provides two recognition methods: a voice recognition module and a text recognition module. That is, users can issue interactive instructions through voice or input interactive instructions through text. The interactive terminal 300 will determine which recognition module to use for instruction recognition based on the user's input method.

[0036] In another specific embodiment, the robot body 200 also includes armrests 210 located on the left and right sides, wherein a laser scanning device 211 and a camera device 212 respectively connected to the control module are disposed on the front end of at least one side of the armrest 210, and a movable robotic arm 220 connected to the control module is disposed on the side of at least one side of the armrest 210.

[0037] In this embodiment, a laser scanning device 211 is provided on the robot body 200 to scan obstacles, so as to prevent the accompanying robot from being disturbed by obstacles during the movement, resulting in obstruction of the movement. The laser scanning device 211 can be an infrared obstacle avoidance sensor. In this embodiment, a camera device 212 is provided on the robot body 200 to capture and collect surrounding image information. For example, the laser scanning device 211 can be combined to capture obstacle information, so that the interactive terminal 300 can judge whether the accompanying route needs to be adjusted according to the obstacle information. For example, the camera device 212 can be used to obtain position information, environmental information, etc.

[0038] In addition, this embodiment also installs a robotic arm 220 on the robot body 200, so that the robotic arm 220 can complete some operations for the user, such as using the robotic arm 220 to help the user perform elevator button operations, or using the robotic arm 220 to help the user carry items, etc.

[0039] Since the embodiments of the system part correspond to the embodiments of the method part, please refer to the description of the embodiments of the method part for the embodiments of the system part, which will not be repeated here.

[0040] Figure 2 A flowchart of an AI-based smart medical accompanying method provided in an embodiment of the present invention is provided. The method is applied to the AI-based smart medical accompanying system as described above, and specifically includes steps S201 to S204.

[0041] S201, in response to an interaction instruction issued by a user, the interaction terminal 300 parses and identifies the interaction instruction through an instruction recognition module, obtains an interaction analysis result, and interacts with the service terminal 100 according to the interaction analysis result, and / or returns interaction information to the user; wherein the interaction instruction includes a medical consultation instruction;

[0042] S202: When the interactive instruction is a medical consultation instruction, the interactive terminal 300 sends the interactive analysis result of the medical consultation instruction to the navigation module, and the navigation module generates a medical consultation route according to the interactive analysis result;

[0043] S203, the interactive terminal 300 obtains the medical consultation route generated by the navigation module and returns it to the user;

[0044] S204. In response to the automatic driving instruction issued by the user based on the accompanying route, the interactive terminal 300 sends the accompanying route to the control module of the robot body 200, and the control module controls the robot body 200 to drive automatically according to the accompanying route.

[0045] In this embodiment, when the user issues an interactive instruction to the interactive terminal 300, the interactive terminal 300 first parses and identifies the received interactive instruction through the instruction recognition module to obtain an interactive analysis result. For the interactive analysis result, the interactive terminal 300 can send it to the service terminal 100, and then the service terminal 100 returns the corresponding interactive content, or directly returns the corresponding interactive content to the user. For example, when the user expects to obtain medical knowledge or interact with medical personnel through the interactive terminal 300, the interactive terminal 300 can send the interactive analysis result to the service terminal 100, so that the corresponding medical knowledge or the medical personnel's reply is returned to the user through the service terminal 100. For example, when the user expects to view the medical personnel information or the clinic location through the interactive terminal 300, the interactive terminal 300 can return the corresponding query result to the user based on the locally pre-stored information.

[0046] There can be many interactive instructions in this embodiment, and considering that this embodiment is an AI-based intelligent accompanying medical method, the most commonly used interactive instruction is the consultation instruction. Therefore, when the interactive terminal 300 receives the consultation instruction issued by the user, it can parse and identify the consultation instruction according to the instruction recognition module, and then send the interactive analysis result to the navigation module. The navigation module can extract the corresponding consultation information from the interactive analysis result, and thus generate a corresponding navigation route according to the consultation information. After obtaining the navigation route, the user can choose whether to be carried to the destination by the accompanying medical robot. When the user chooses to be carried by the accompanying medical robot, the interactive terminal 300 will send corresponding information to the control module on the accompanying medical robot, such as sending the accompanying medical route and driving instructions, and then the control module will control the robot body 200 to drive automatically, so as to carry the user to the consultation location.

[0047] This embodiment uses the interactive terminal 300 to conduct interactive questions and answers with the user, and uses a movable accompanying robot to accompany the user to consult and see a doctor, thereby improving the effect of medical accompanying. It can be understood that the interactive terminal 300 is provided with a wireless communication module for sending and receiving data, so that the interactive terminal 300 can communicate with the service terminal 100, the control module, etc. In actual application scenarios, video communication software can also be set on the interactive terminal 300, so that after the interactive terminal 300 is connected to the Internet, the user can communicate with other people through video or voice through the video communication software of the interactive terminal 300, such as medical personnel, family members, etc., which can improve the efficiency of consultation.

[0048] In one embodiment, if Figure 3 As shown, the step S201 includes: steps S301 to S304.

[0049] S301: When the interaction instruction sent by the user is a voice interaction instruction, the interaction terminal 300 receives the voice interaction instruction through a voice recognition module, and performs spatial mapping processing on the voice interaction instruction to obtain a voice sequence;

[0050] S302, extracting context features from the speech sequence to obtain corresponding context features;

[0051] S303, combining the context features, using a speech recognition model to recognize the speech interaction instruction, and obtaining a speech recognition result;

[0052] S304: extracting features from the speech recognition result through a deep learning neural network, and using the feature extraction result as the interactive analysis result.

[0053] In this embodiment, since the user can not only control the robot body 200 to accompany the patient through the interactive terminal 300, but also determine the corresponding consultation location through the interactive terminal 300, for example, if the user has certain physical symptoms but is not sure which department to go to for consultation, he can input a paragraph of speech to the interactive terminal 300 to describe his own symptoms, and then the interactive terminal 300 extracts keywords and other content based on the voice input by the user, and then outputs the corresponding reference department based on the extracted content. Considering that the user has a certain accent when inputting voice, resulting in inaccurate recognition results, the interactive terminal 300 first performs accent calibration on the voice interaction instruction issued by the user, and then extracts features of the voice interaction instruction after accent calibration through a deep learning neural network, thereby obtaining an interactive analysis result.

[0054] Specifically, when performing accent calibration, the voice interaction instructions are first spatially mapped, so that the voice sequence corresponding to the voice interaction instructions can be obtained, and then the context feature extraction is performed on the voice sequence, and then the extracted context feature is combined to perform voice recognition, so that a more accurate voice recognition result can be obtained. Here, when obtaining the voice sequence, the voice interaction instructions can be sampled according to the preset frame interval to obtain the initial voice sequence, and then each initial voice sequence data in the initial voice sequence is spatially mapped by the encoder to obtain the voice sequence. When performing context feature extraction, firstly, for each voice variable in the voice sequence, its similarity with each phoneme vector in the preset phoneme codebook is calculated, and then the target phoneme vector corresponding to the voice variable is determined according to the phoneme vector with the highest similarity, and then the target phoneme vector is spliced ​​into a target phoneme sequence, and then the context feature is extracted from the target phoneme sequence using the context network.

[0055] In one embodiment, the step S201 further includes:

[0056] When the interaction instruction sent by the user is a text interaction instruction, the interaction terminal 300 receives the text interaction instruction through a text recognition module, and performs text preprocessing on the text interaction instruction to obtain an interaction text;

[0057] The interactive text is segmented into words and sentences to obtain a plurality of sub-texts, and then prompt words about the medical consultation are set for each sub-text, and after the setting is completed, each sub-text is input into the generative artificial intelligence big model, so that the generative artificial intelligence big model outputs a plurality of corresponding error correction sub-texts;

[0058] splicing the multiple error correction subtexts into an error correction text, and comparing the difference between the error correction text and the interactive text to obtain a comparison result;

[0059] Performing word segmentation mapping on the error correction text according to the comparison result, and performing rule matching on the interactive text;

[0060] Combining the results of word segmentation mapping and rule matching, a target error correction word for the interactive text is obtained, and then the interactive text is annotated with error correction based on the target error correction word;

[0061] A standard text is obtained based on the results of the error correction annotation, and features are extracted from the standard text through a deep learning neural network, and then the result of the feature extraction is used as the interactive parsing result.

[0062] In this embodiment, when the user inputs a text to the interactive terminal 300 to describe his own symptoms, the interactive terminal 300 can extract keywords and other contents according to the text input by the user, and then output the corresponding reference department according to the extracted contents. Specifically, when the user sends a text interaction instruction to the interactive terminal 300, the interactive terminal 300 performs text preprocessing on the received text interaction instruction, such as removing stop words in the text interaction instruction, and performing missing value detection on the text interaction instruction, etc., and then performs error correction processing on the preprocessed text to avoid the influence of text errors on the extraction results in the subsequent feature extraction process (such as keywords). In the error correction process, the interactive text is first divided into multiple sub-texts. Of course, if the interactive text content is small, it is not necessary to divide it. Then, prompt words are set for the divided sub-texts. The prompt words need to be related to the medical consultation. In this way, it can be ensured that the error correction content is related to the consultation and the error correction efficiency is improved. Then, the sub-text with the prompt words is predicted and output through the generative artificial intelligence model to obtain the corresponding error correction sub-text, and then all the error correction sub-texts are spliced ​​together to obtain the error correction text corresponding to the interactive text. The error correction text is then compared with the interactive text to determine whether there is a difference between the two. The error correction text is segmented and mapped according to the judgment result, and the interactive text is matched with rules at the same time. In this way, the words that need to be corrected in the interactive text can be determined, so that the interactive text can be annotated and corrected, and the corresponding standard text can be obtained. For the standard text, feature extraction can be performed on it through a deep learning neural network to obtain the corresponding feature extraction result, and the feature extraction result can be used as the interactive analysis result for the text interaction instruction input by the user.

[0063] Here, word segmentation mapping refers to mapping the editing operation to the corresponding word segmentation and obtaining the editing mapping result of the word segmentation granularity. The purpose of this is to make it easier to understand which word in which word needs to be modified. Rule matching refers to matching the interactive text with the preset full-scale error-prone word library, obtaining the error correction result of the interactive text under the rule matching, and setting the error correction result as the rule matching result. It can also be understood that whether it is the generative artificial intelligence large model or the deep learning neural network, before using them for prediction or extraction, they will be trained and learned using the corresponding data sets to improve the processing effect of the two.

[0064] In one embodiment, the step S202 includes:

[0065] Based on the interactive analysis result, extract the user's current location and consultation location;

[0066] Determine a starting reference point closest to the current position, and use the starting reference point as the starting position;

[0067] Combining the starting location and the consultation location, generating a consultation route including at least one reference point;

[0068] When traveling along the first accompanying consultation route, the user's location information is acquired in real time, and the accompanying consultation route is dynamically adjusted according to the location information until the user reaches the consultation location.

[0069] In this embodiment, when the accompanying route is generated by the navigation module, considering that the accompanying route will involve indoor navigation, it is necessary to make the generated accompanying route have a certain accuracy. Specifically, the starting position and consultation position of the accompanying route are first determined. Here, since the user's position, that is, the current position, may not be iconic, a reference point of the distance is selected according to the current position as the starting position. Of course, if the user's current position is the reference point, then the current position can be directly used as the starting position. The user's consultation position normally belongs to a specific position, so there is no need to make special settings for it. For example, if the user is between two buildings, then the nearest building or a house number or department under the building can be selected as the starting position. At the same time, if the user's consultation position is Department A, then Department A can be directly used as the end point of the accompanying route. Then, between the starting position and the consultation position, the reference points are obtained, and then the accompanying route is generated in combination with these reference points. Furthermore, in order to ensure the accuracy of the accompanying route, the accompanying route can be adjusted in real time according to the user's location information while the user is traveling, so that the user can reach the consultation location safely and quickly.

[0070] In some optional embodiments, when combining the starting position and the consultation position to generate an accompanying route containing at least one reference point, it can include: first obtaining all reference points, and then setting a reference point number for each reference point; then connecting adjacent reference points to form corresponding reference edges, and assigning corresponding attributes to the corresponding reference edges according to the attributes of the reference points; quantifying the attributes of each reference edge into a weight, and then taking the starting position as the starting point and the consultation position as the end point, generating the best accompanying route based on the A-star algorithm and according to the weight of each reference edge. Here, specifically, a reference path of the user's location is obtained, such as the only connecting bridge between two buildings or a direct elevator across multiple floors, etc., and then the positional relationship between the reference path and the starting position and the consultation position is determined; if the reference path includes the starting position and / or the consultation position, or the reference path is located between the starting position and the consultation position, then the starting position, the consultation position and the reference path are combined to generate the best accompanying route; if the reference path does not include the starting position or the consultation position, or the reference path is not located between the starting position and the consultation position, or the reference path is not obtained, then all reference points between the starting position and the consultation position are obtained, and then the starting position is used as the starting point, and the distance score of each reference point adjacent to the starting position is calculated using the estimation function, and the calculation formula is: f i =a i +b i , where f i is the distance score of the i-th reference point adjacent to the starting position, a i is the weight of the reference edge obtained by connecting the starting position with the i-th reference point adjacent to the starting position, b i is the estimated weight from the i-th reference point adjacent to the starting position to the consultation position. Then, the reference point with the smallest distance score is selected from all the reference points adjacent to the starting position as the candidate reference point, and so on, until the candidate reference point adjacent to the consultation position is selected, and then the route connecting the starting position, the consultation position and all the candidate reference points in sequence is used as the best accompanying route.

[0071] In one embodiment, the step S204 includes:

[0072] When the robot body 200 is in the process of driving, the control module uses the laser scanning device 211 on the robot body 200 to scan the obstacle on the front route;

[0073] When an obstacle is scanned, the control module obtains obstacle information of the route ahead through the camera device 212 and sends it to the interactive terminal 300;

[0074] After receiving the obstacle information, the interactive terminal 300 sends the obstacle information to the navigation module, so that the navigation module dynamically adjusts the accompanying route according to the obstacle information;

[0075] The interactive terminal 300 sends the dynamically adjusted accompanying diagnosis route to the control module, so that the control module controls the robot body 200 to avoid obstacles on the route ahead according to the dynamically adjusted accompanying diagnosis route.

[0076] In this embodiment, since the scene is a medical institution, there are likely to be some obstacles in the accompanying route generated by the navigation module. Therefore, during the driving process of the robot body 200, the laser scanning device 211 is used to perform real-time scanning to determine whether there are obstacles on the route ahead. If an obstacle is scanned, the corresponding obstacle image is captured by the camera device 212, and then the interactive terminal 300 analyzes and processes the obstacle image, such as determining the size of the obstacle, so as to determine whether the accompanying route needs to be changed. When it is determined that the accompanying route needs to be changed, the accompanying route is dynamically adjusted by the navigation module so that the accompanying route can avoid the obstacles ahead. If it is determined that the accompanying route does not need to be changed, the travel speed of the robot body 200 can be adjusted by the control module. Preferably, a deceleration device may be provided on the robot body 200, and the deceleration device is connected to the laser scanning device 211. When the laser scanning device 211 scans an obstacle, the deceleration device is triggered to decelerate the robot body 200 and upload obstacle information at the same time, so that the interactive terminal 300 and the control module have more processing time to avoid the robot body 200 approaching or even contacting the obstacle due to untimely processing. Further, the distance between the robot body 200 and the obstacle is determined by the laser scanning device 211, and when the distance between the robot body 200 and the obstacle reaches a preset safety distance (for example, 30 cm), the robot body 200 is controlled to stop moving forward.

[0077] In a specific embodiment, when scanning is performed by the laser scanning device 211, the control module processes the scanned data, such as filtering the scanned data to select valid data, or calculating the distance, angle, etc. between the robot body 200 and the obstacle based on the scanned data, such as calculating the distance between the robot body 200 and the obstacle by measuring the time taken for the laser to be emitted and reflected back. In actual application scenarios, in addition to scanning the obstacle on the route ahead, the laser scanning device 211 can also emit laser beams to both sides of the robot body 200 (with a maximum angle of 180 degrees), that is, it can cover the width of the robot body 200 in a fan shape, thereby improving the obstacle scanning effect.

[0078] In one embodiment, the step S204 further includes:

[0079] When the robot body 200 moves into the elevator, the interactive terminal 300 obtains the corresponding consultation floor based on the interactive analysis result, and sends the consultation floor to the control module;

[0080] The control module obtains the height of the floor button in the elevator through the camera device 212, and controls the height of the mechanical arm 220 on the robot body 200 in combination with the consultation floor, so that the mechanical arm 220 performs button operation;

[0081] The control module continues to obtain the arrival floor of the elevator through the camera device 212, and determines whether the arrival floor is the same as the consultation floor;

[0082] When it is determined that the arrival floor is the same as the consultation floor, the control module controls the robot body 200 to drive out of the elevator.

[0083] In this embodiment, when the user goes to the consultation location through the robot body 200, considering that the user's starting position and the consultation location may not be on the same floor, it is necessary to travel to the corresponding floor through the elevator, and it may be inconvenient for the user to touch the elevator button when sitting on the robot body 200. At this time, the robot arm 220 on the robot body 200 can help the user to achieve the button operation, thereby improving the accompanying diagnosis effect. Specifically, when the robot body 200 is in the elevator, the floor button position of the elevator can be obtained through the camera device 212, and then the control module controls the robot arm 220 to lift the robot arm 220 to the floor button position to achieve the button operation. At the same time, after the button operation is completed, the camera device 212 is continued to collect the elevator's arrival floor image, and then the control module performs image analysis processing on the arrival floor image to extract the corresponding arrival floor number from it, and then judge whether the arrival floor number is consistent with the consultation floor. If it is judged that the arrival floor is inconsistent with the consultation floor, continue to collect and judge until it is determined that the arrival floor is consistent with the consultation floor, and then control the robot body 200 to drive out of the elevator.

[0084] In some preferred embodiments, an image acquisition device may also be provided at the front end of the robotic arm 220, so that during the lifting process of the robotic arm 220, the position of the floor button may be more accurately and reliably acquired through the image acquisition device, thereby avoiding accidental pressing of other floor buttons.

[0085] In some optional embodiments, in addition to using the robotic arm 220 to help the user perform key operations, a robotic claw can also be set at the front end of the robotic arm 220, and then the robotic claw can be used to help extract some items, so that the user can interact more conveniently through the interactive terminal 300.

[0086] The embodiment of the present invention further provides a computer-readable storage medium on which a computer program is stored, and when the computer program is executed, the steps provided in the above embodiment can be implemented. The storage medium may include: a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and other media that can store program codes.

[0087] The various embodiments in the specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same and similar parts between the various embodiments can be referred to each other. For the system disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part description. It should be pointed out that for ordinary technicians in this technical field, without departing from the principles of this application, several improvements and modifications can be made to the present application, and these improvements and modifications also fall within the scope of protection of the claims of this application.

[0088] It should also be noted that, in this specification, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "comprises", "comprising" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the statement "comprising a ..." does not exclude the presence of other identical elements in the process, method, article or device including the element.

Claims

1. An AI-based intelligent medical companion system, characterized in that: include: Service terminal; The accompanying robot comprises a movable robot body and an interactive terminal arranged on the robot body, wherein the interactive terminal is communicatively connected with the service terminal; The interactive terminal includes an instruction recognition module and a navigation module; the instruction recognition module is used to receive and analyze the interactive instructions sent by the user to the interactive terminal, so that the interactive terminal communicates with the service terminal and / or displays interactive information to the user according to the analysis result; the navigation module is used to receive the consultation instruction sent by the user to the interactive terminal, and generate a medical consultation route according to the consultation instruction; The robot body includes a control module for controlling the movement of the robot body, and the control module is connected to the interactive terminal.

2. The AI-based intelligent medical companion system according to claim 1 is characterized in that: The instruction recognition module includes a speech recognition module and a text recognition module; The voice recognition module is used to receive the user's voice interaction instructions, and the text recognition module is used to receive the user's text interaction instructions.

3. The AI-based intelligent medical companion system according to claim 1 is characterized in that: The robot body also includes armrests located on the left and right sides, wherein a laser scanning device and a camera device respectively connected to the control module are arranged on the front end of at least one side of the armrest, and a movable mechanical arm connected to the control module is arranged on the side of at least one side of the armrest.

4. An AI-based intelligent accompanying diagnosis method, applied to the AI-based intelligent accompanying diagnosis system as claimed in any one of claims 1 to 3, characterized in that: include: In response to the interactive instruction issued by the user, the interactive terminal parses and identifies the interactive instruction through the instruction recognition module, obtains the interactive analysis result, and interacts with the service terminal according to the interactive analysis result, and / or returns interactive information to the user; wherein the interactive instruction includes a medical consultation instruction; When the interactive instruction is a medical consultation instruction, the interactive terminal sends the interactive analysis result of the medical consultation instruction to the navigation module, and the navigation module generates a medical consultation route according to the interactive analysis result; The interactive terminal obtains the medical consultation route generated by the navigation module and returns it to the user; In response to the automatic driving instruction issued by the user based on the accompanying route, the interactive terminal sends the accompanying route to the control module of the robot body, and the control module controls the robot body to drive automatically according to the accompanying route.

5. The AI-based intelligent accompanying diagnosis method according to claim 4 is applied to the AI-based intelligent accompanying diagnosis system according to claim 2, characterized in that: The interactive terminal parses and identifies the interactive instruction through an instruction recognition module to obtain an interactive analysis result, including: When the interaction instruction sent by the user is a voice interaction instruction, the interaction terminal receives the voice interaction instruction through a voice recognition module, and performs spatial mapping processing on the voice interaction instruction to obtain a voice sequence; Extracting context features from the speech sequence to obtain corresponding context features; In combination with the context features, the voice interaction instruction is recognized using a voice recognition model to obtain a voice recognition result; The speech recognition result is subjected to feature extraction through a deep learning neural network, and the result of the feature extraction is used as the interactive analysis result.

6. The AI-based intelligent accompanying diagnosis method according to claim 4 is applied to the AI-based intelligent accompanying diagnosis system according to claim 2, characterized in that: The instruction recognition module in the interactive terminal parses and recognizes the interactive instruction to obtain an interactive analysis result, and further includes: When the interaction instruction sent by the user is a text interaction instruction, the interaction terminal receives the text interaction instruction through a text recognition module, and performs text preprocessing on the text interaction instruction to obtain an interaction text; The interactive text is segmented into words and sentences to obtain a plurality of sub-texts, and then prompt words about the medical consultation are set for each sub-text, and after the setting is completed, each sub-text is input into the generative artificial intelligence big model, so that the generative artificial intelligence big model outputs a plurality of corresponding error correction sub-texts; splicing the multiple error correction subtexts into an error correction text, and comparing the difference between the error correction text and the interactive text to obtain a comparison result; Performing word segmentation mapping on the error correction text according to the comparison result, and performing rule matching on the interactive text; Combining the results of word segmentation mapping and rule matching, a target error correction word for the interactive text is obtained, and then the interactive text is annotated with error correction based on the target error correction word; A standard text is obtained based on the results of the error correction annotation, and features are extracted from the standard text through a deep learning neural network, and then the result of the feature extraction is used as the interactive parsing result.

7. The AI-based intelligent accompanying diagnosis method according to claim 4 is characterized in that: The interactive terminal sends the interactive analysis result of the medical consultation instruction to the navigation module, and the navigation module generates a medical consultation route according to the interactive analysis result, including: Based on the interactive analysis result, extract the user's current location and consultation location; Determine a starting reference point closest to the current position, and use the starting reference point as the starting position; Combining the starting location and the consultation location, generating a consultation route including at least one reference point; When traveling along the first accompanying diagnosis route, the user's location information is obtained in real time, and the accompanying diagnosis route is dynamically adjusted according to the location information until the user reaches the consultation location.

8. The AI-based intelligent accompanying diagnosis method according to claim 4 is applied to the AI-based intelligent accompanying diagnosis system according to claim 3, characterized in that: The control module controls the robot body to automatically drive according to the accompanying medical route, including: When the robot body is in the process of driving, the control module uses the laser scanning device on the robot body to scan the obstacle on the front route; When an obstacle is scanned, the control module obtains obstacle information of the route ahead through the camera device and sends it to the interactive terminal; After receiving the obstacle information, the interactive terminal sends the obstacle information to the navigation module, so that the navigation module dynamically adjusts the accompanying route according to the obstacle information; The interactive terminal sends the dynamically adjusted accompanying route to the control module, so that the control module controls the robot body to avoid obstacles on the route ahead according to the dynamically adjusted accompanying route.

9. The AI-based intelligent accompanying diagnosis method according to claim 4 is characterized in that: Applied to the AI-based intelligent medical companion system as claimed in claim 3, the control module controls the robot body to automatically drive according to the medical companion route, and further includes: When the robot body moves into the elevator, the interactive terminal obtains the corresponding consultation floor based on the interactive analysis result, and sends the consultation floor to the control module; The control module obtains the height of the floor buttons in the elevator through the camera device, and controls the height of the mechanical arm on the robot body in combination with the consultation floor, so that the mechanical arm performs button operation; The control module continues to obtain the arrival floor of the elevator through the camera device, and determines whether the arrival floor is the same as the consultation floor; When it is determined that the arrival floor is the same as the consultation floor, the control module controls the robot body to drive out of the elevator.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the AI-based intelligent medical consultation method as described in any one of claims 4 to 9 is implemented.

Citation Information

Patent Citations

  • Omni-directional intelligent electric wheel chair

    CN102600026A

  • Intelligent medical guiding system and method thereof

    CN113724703A

  • Acent speech recognition method and device based on artificial intelligence, equipment and medium

    CN116386628A

  • Medical text error correction system and medical query prompt text display method and device

    CN118093789A

  • Text error correction method and device based on AIGC, computer equipment and storage medium

    CN118607516A

Cited By

  • A smart hospital accompanying diagnosis wheelchair robot scheduling system and method

    CN122930434A