A human-computer interaction method and system for emergency rescue scenarios

Abstract

This application provides a human-computer interaction method and system for emergency rescue scenarios. The method involves parsing and processing the multimedia information received from an emergency rescue requester to obtain an emergency rescue situation awareness model. Based on this model, a deep reinforcement learning algorithm is used to select possible emergency response strategies. Medical knowledge graph technology is then used to analyze the relationship between the requester's symptoms and possible causes to generate an emergency rescue guidance plan. Based on this guidance plan, a traffic flow prediction algorithm is used to optimize the rescue route in real time. Combined with road condition information collected using IoT technology, the route of the rescue vehicle is dynamically adjusted to generate an optimal rescue route and the latest rescue progress notification. Finally, the optimal rescue route and the latest progress notification are used to visualize the emergency rescue operation process, generating intuitive visual guidance. The technical solution provided by this application significantly improves the speed, accuracy, and effectiveness of emergency response.

Description

Technical Field

[0001] This application relates to the field of human-computer interaction technology, and in particular to a human-computer interaction method and system for emergency rescue scenarios. Background Technology

[0002] In emergency situations, quickly and accurately understanding the caller's condition and providing effective rescue guidance is crucial. Faced with complex emergencies, the system needs high-precision situational awareness, capable of real-time analysis and processing of multimedia information from various channels, such as voice, text, and video, while simultaneously using a geographic information system to accurately pinpoint the caller's location. Furthermore, it needs to intelligently match the most suitable emergency response strategy and dynamically adjust the rescue route to ensure that rescue vehicles can reach the scene quickly.

[0003] Current emergency response systems typically rely on traditional telephone calls and manual dispatch, supplemented by basic automated location services. Some advanced systems have begun to incorporate natural language processing technology and basic map navigation functions to analyze emergency requests and plan initial rescue routes. However, these systems have limited intelligence, primarily focusing on single information sources and lacking the ability to comprehensively analyze multimodal information. They also fail to fully utilize medical knowledge graphs for in-depth correlation analysis between symptoms and causes.

[0004] While existing emergency response systems can meet basic needs to some extent, they have significant limitations. First, traditional systems lack the ability to understand complex situations, making it difficult to comprehensively and accurately assess the specific circumstances of an emergency request. Second, existing route planning and traffic prediction algorithms are not flexible enough to adapt to constantly changing road conditions in real time, leading to prolonged rescue times. Finally, the lack of intuitive and personalized emergency guidance means that while waiting for professional rescue teams to arrive, requesters often lack effective guidance to perform correct self-rescue or mutual-rescue procedures, impacting the effectiveness of emergency care. Summary of the Invention

[0005] This application provides a human-computer interaction method and system for emergency rescue scenarios, in order to solve the problem of poor human-computer interaction effects in emergency rescue scenarios in the prior art.

[0006] In a first aspect, embodiments of this application provide a human-computer interaction method for emergency rescue scenarios, including:

[0007] Using multimodal sentiment analysis technology and geographic information system, the multimedia information received from emergency callers is analyzed and processed to obtain an emergency situation perception model;

[0008] Based on the emergency situation awareness model, a deep reinforcement learning algorithm is used to select and process possible emergency response strategies. Combined with medical knowledge graph technology, the relationship between the requester's symptoms and possible causes is analyzed to generate an emergency guidance plan.

[0009] Based on the aforementioned emergency rescue guidance scheme, the rescue route is optimized in real time using a traffic flow prediction algorithm. Combined with road condition information collected by Internet of Things technology, the driving route of the rescue vehicle is dynamically adjusted to generate the optimal rescue route and the latest rescue progress notification.

[0010] By utilizing the optimal rescue route and the latest rescue progress notifications, the emergency rescue operation process is visualized to generate intuitive visual guidance.

[0011] Optionally, the step of using a deep reinforcement learning algorithm to select possible emergency response strategies based on the emergency situation awareness model, and combining medical knowledge graph technology to analyze the relationship between the requester's symptoms and possible causes to generate an emergency guidance plan, includes:

[0012] Using the aforementioned emergency situation perception model, the emotional state and location information of the requester are extracted and processed to obtain emotional location features;

[0013] Based on the aforementioned emotional location features, a deep reinforcement learning algorithm is used to evaluate and process various emergency response strategies to obtain the optimal response strategy.

[0014] Based on the optimal response strategy and combined with medical knowledge graph technology, the requester's symptoms and possible causes are analyzed for correlation to obtain the most relevant disease type and its urgency.

[0015] Using the most relevant disease types and their urgency levels, a set of first aid guidelines is generated.

[0016] Optionally, the step of using the emergency situation awareness model to extract and process the emotional state and location information of the requester to obtain emotional location features includes:

[0017] Natural language processing technology is used to analyze and process the voice or text information of emergency callers, identify emotion-related keywords and expressions, and obtain emotional state data.

[0018] Based on the emotional state data and combined with environmental factors, the location of the requester is accurately located to generate location information.

[0019] Based on the emotional state data and location information, a comprehensive analysis is performed to assess the correlation between emotions and geographical location, and the emotional location correlation analysis results are obtained.

[0020] The emotional location association analysis results are used to perform fusion processing to obtain emotional location features.

[0021] Optionally, based on the optimal response strategy and combined with medical knowledge graph technology, the process of performing correlation analysis on the requester's symptoms and possible causes to obtain the most relevant disease types and their urgency includes:

[0022] Using the symptom identification information contained in the optimal response strategy, the requester's symptom presentation is preliminarily classified to obtain the symptom classification result;

[0023] Based on the symptom classification results, and combined with the disease and symptom relationship database in the medical knowledge graph, the association between the requester's symptoms and known diseases is matched to generate a list of potential diseases.

[0024] Based on the list of potential diseases, medical knowledge graph technology is used to further analyze the typical symptoms, common complications and pathogenesis of each disease, assess the probability and urgency of each disease, and obtain a disease probability and urgency assessment report.

[0025] Using the aforementioned disease probability and urgency assessment report, the most relevant disease types and their urgency levels are obtained.

[0026] Optionally, based on the emergency rescue guidance plan, the rescue route is optimized in real time using a traffic flow prediction algorithm, and the driving route of the rescue vehicle is dynamically adjusted by combining road condition information collected by Internet of Things technology, generating the optimal rescue route and the latest rescue progress notification, including:

[0027] Using the urgency and location information provided in the emergency guidance plan, the initially planned rescue route is initialized to obtain the initial rescue route;

[0028] Based on the initial rescue route, the current and expected traffic conditions are analyzed and processed using a traffic flow prediction algorithm to predict factors that may affect the rescue time and generate traffic condition prediction results.

[0029] Based on the traffic condition prediction results and combined with real-time road condition information collected by Internet of Things technology, the driving route of the rescue vehicle is dynamically adjusted to obtain the optimal rescue route.

[0030] Using the optimal rescue route, the actual driving situation is continuously monitored, communication is maintained with the requester, and the latest rescue progress notifications are generated.

[0031] Optionally, the step of dynamically adjusting the route of the rescue vehicle based on the traffic condition prediction results and in conjunction with real-time road condition information collected using IoT technology to obtain the optimal rescue route includes:

[0032] Using the traffic condition prediction results, factors that may affect rescue time are identified and processed to obtain potential traffic influencing factors;

[0033] Based on the aforementioned potential traffic influencing factors, and combined with real-time road condition information collected using IoT technology, a comprehensive analysis and processing is performed to generate real-time road condition updates.

[0034] Based on the real-time traffic updates, the driving routes of the rescue vehicles are replanned to avoid unfavorable road conditions, resulting in adjusted driving routes.

[0035] Using the adjusted driving route, the system is continuously optimized until the shortest time and best safety conditions are achieved, thus generating the optimal rescue route.

[0036] Optionally, the process of visualizing the emergency rescue operation using the optimal rescue route and the latest rescue progress notification to generate intuitive visual guidance includes:

[0037] Using the optimal rescue route and combining it with map data, the travel route of the rescue vehicle is visualized to obtain a dynamic route map;

[0038] Based on the dynamic route map, the current location and estimated arrival time of the rescue vehicles are updated and displayed in real time according to the latest rescue progress notification, generating a real-time progress display;

[0039] Based on the real-time progress display, combined with the requester's specific symptoms and the generated first aid guidance plan, the first aid operation steps are demonstrated using augmented reality technology to obtain a first aid operation demonstration sequence.

[0040] The aforementioned first aid procedure demonstration sequence is used to generate intuitive visual instructions.

[0041] Secondly, embodiments of this application provide a human-computer interaction system for emergency rescue scenarios, comprising:

[0042] The parsing module is used to analyze and process the multimedia information received from emergency callers using multimodal sentiment analysis technology and geographic information systems to obtain an emergency situation perception model.

[0043] The selection module is used to select possible emergency response strategies based on the emergency situation perception model using a deep reinforcement learning algorithm, and to generate an emergency guidance plan by combining medical knowledge graph technology to analyze the relationship between the requester's symptoms and possible causes.

[0044] The adjustment module is used to optimize the rescue route in real time based on the emergency rescue guidance plan by using a traffic flow prediction algorithm, and dynamically adjust the driving route of the rescue vehicle by combining road condition information collected by Internet of Things technology, so as to generate the optimal rescue route and the latest rescue progress notification.

[0045] The processing module is used to visualize the emergency rescue operation process using the optimal rescue route and the latest rescue progress notification, and generate intuitive visual guidance.

[0046] Thirdly, embodiments of this application provide a computing device, including a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be invoked and executed by the processing component to implement a human-computer interaction method in an emergency rescue scenario as described in the first aspect.

[0047] Fourthly, embodiments of this application provide a computer storage medium storing a computer program, which, when executed by a computer, implements a human-computer interaction method in an emergency rescue scenario as described in the first aspect.

[0048] In this embodiment, multimodal sentiment analysis technology and geographic information systems are used to parse and process the multimedia information received from emergency callers to obtain an emergency situation awareness model. Based on the emergency situation awareness model, deep reinforcement learning algorithms are used to select possible emergency response strategies. Medical knowledge graph technology is combined to analyze the relationship between the caller's symptoms and possible causes to generate an emergency guidance plan. Based on the emergency guidance plan, a traffic flow prediction algorithm is used to optimize the rescue route in real time. Combined with road condition information collected by Internet of Things technology, the driving route of the rescue vehicle is dynamically adjusted to generate the optimal rescue route and the latest rescue progress notification. Using the optimal rescue route and the latest rescue progress notification, the emergency operation process is visualized to generate intuitive visual guidance.

[0049] The technical solution of this application has the following beneficial effects:

[0050] By analyzing and processing the multimedia information received from emergency callers using multimodal sentiment analysis technology and geographic information systems, an emergency situation awareness model is constructed. This method can more accurately understand the specific circumstances of emergency requests, thereby quickly and precisely formulating emergency response strategies, significantly improving the speed and accuracy of emergency response. Based on the emergency situation awareness model, deep reinforcement learning algorithms are used to select the optimal emergency response strategy, and medical knowledge graph technology is combined to analyze the relationship between symptoms and causes, generating personalized emergency guidance plans. This process not only considers the current health status of the emergency caller but also incorporates historical data and professional knowledge, ensuring the effectiveness and scientific nature of emergency measures and improving the quality of emergency decision-making. Based on the emergency guidance plan, traffic flow prediction algorithms are used to optimize rescue routes in real time, and road condition information collected by IoT technology is used to dynamically adjust the routes of rescue vehicles. This method can effectively avoid traffic congestion and other adverse factors, ensuring that rescue vehicles arrive at the scene in the shortest time and under the best conditions, greatly improving rescue efficiency and reducing valuable rescue time. By utilizing the optimal rescue route and the latest rescue progress notifications, the emergency operation process is visualized to generate intuitive visual guidance. This intuitive guidance helps emergency callers to perform effective self-rescue or mutual rescue before professional rescue forces arrive, improving the accuracy and success rate of emergency procedures, while reducing the caller's anxiety and enhancing their confidence in dealing with emergencies.

[0051] Furthermore, this method ensures the scientific and personalized nature of emergency response strategies through precise contextual awareness and intelligent decision-making. Specifically, an emergency contextual awareness model built using multimodal sentiment analysis technology and geographic information systems can more accurately understand the specific circumstances of an emergency request, including the requester's emotional state and geographical location, providing a solid foundation for subsequent response strategy selection. Based on emotional location features, deep reinforcement learning algorithms are used to dynamically select the optimal response strategy, and the combination of historical data and professional knowledge improves the scientific and effective nature of decision-making. Further, by incorporating medical knowledge graph technology, correlation analysis is performed on the requester's symptoms and possible causes to identify the most relevant disease types and their urgency levels, making the generated emergency guidance plan more personalized and targeted, helping emergency personnel and requesters take the most effective measures at the first moment. Finally, through the above intelligent processing steps, not only is the speed and accuracy of emergency response improved, but the overall coordination and efficiency of rescue operations are also enhanced, unnecessary delays are reduced, more precious time is gained to save lives, and the overall quality and effectiveness of emergency care are significantly improved.

[0052] Furthermore, this method ensures high flexibility and rapid response in rescue operations through intelligent route planning and dynamic adjustments. Specifically, by initializing rescue routes and accurately predicting traffic conditions, the system can avoid potential traffic obstacles in advance, significantly shortening rescue time. Combining IoT technology with real-time updates of road conditions and dynamic route optimization further improves rescue efficiency. Simultaneously, continuous monitoring of actual driving conditions and communication with the requester ensures information transparency and timely feedback, enhancing the requester's sense of security and trust. Ultimately, this achieves efficient and precise emergency dispatch, significantly improving the overall rescue effectiveness and user experience.

[0053] These or other aspects of this application will become more apparent in the following description of the embodiments. Attached Figure Description

[0054] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0055] Figure 1 A flowchart illustrating a human-computer interaction method in an emergency rescue scenario provided in this application embodiment;

[0056] Figure 2 This application provides a schematic diagram of the structure of a human-computer interaction system in an emergency rescue scenario.

[0057] Figure 3 This is a schematic diagram of the structure of a computing device provided in an embodiment of this application. Detailed Implementation

[0058] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings.

[0059] In some of the processes described in the specification, claims, and accompanying drawings of this application, multiple operations appearing in a specific order are included. However, it should be clearly understood that these operations may not be executed in the order they appear herein, or may be executed in parallel. The operation numbers, such as 101, 102, etc., are merely used to distinguish different operations and do not themselves represent any execution order. Furthermore, these processes may include more or fewer operations, and these operations may be executed sequentially or in parallel. It should be noted that the descriptions such as "first," "second," etc., in this document are used to distinguish different messages, devices, modules, etc., and do not represent a chronological order, nor do they limit "first" and "second" to different types.

[0060] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0061] Figure 1 A flowchart of a human-computer interaction method in an emergency rescue scenario is provided in this application embodiment, as shown below. Figure 1 As shown, the method includes:

[0062] 101. Using multimodal sentiment analysis technology and geographic information system, the multimedia information of the received emergency callers is analyzed and processed to obtain an emergency situation perception model;

[0063] In this step, multimodal sentiment analysis technology involves sentiment recognition and semantic understanding of various data formats, including voice, text, and video, to capture the emotional state and specific needs of emergency callers. Geographic Information Systems (GIS) then integrate geographic location data to provide precise location services and environmental information. The combined use of these two technologies enables comprehensive analysis of multimedia information from different channels, constructing a perceptual model that comprehensively reflects the emergency situation.

[0064] In this embodiment, the system first receives multimedia information from emergency callers, such as voice calls, text messages, or video streams. Then, it uses multimodal sentiment analysis technology to analyze the emotional state and urgency of these messages, and combines the location information provided by the geographic information system to generate an emergency situation perception model that includes emotional features and geographic location, providing a solid foundation for subsequent decision-making.

[0065] Suppose that in an emergency call for help following a traffic accident, the system receives a recorded phone call and photos taken at the scene. Using multimodal sentiment analysis, the system identifies the caller's extreme anxiety and describes the severity of the accident; simultaneously, a geographic information system (GIS) precisely locates the accident site. Based on this information, the system constructs an emergency situation awareness model that accurately reflects the specific circumstances of the accident and the caller's emotional state, providing crucial information for selecting the next rescue strategy.

[0066] 102. Based on the emergency situation perception model, a deep reinforcement learning algorithm is used to select and process possible emergency response strategies. Combined with medical knowledge graph technology, the relationship between the requester's symptoms and possible causes is analyzed to generate an emergency guidance plan.

[0067] In this step, deep reinforcement learning algorithms, a machine learning method, optimize the decision-making process by simulating numerous interaction scenarios; a medical knowledge graph is a knowledge base that structures medical knowledge to assist in diagnosis and treatment recommendations. The combination of these two can intelligently evaluate the effectiveness and applicability of various emergency response strategies, while simultaneously analyzing the complex relationship between symptoms and causes.

[0068] In this embodiment, after obtaining the emergency situation awareness model, the system uses a deep reinforcement learning algorithm to evaluate various possible emergency response strategies and select the optimal solution. Then, combined with medical knowledge graph technology, it deeply analyzes the requester's symptoms and potential causes to ensure that the generated emergency guidance plan is both scientific and personalized, and responds to the current emergency situation in the best way.

[0069] A drowning incident occurred in a coastal city. The person seeking help sent a voice message and their current location through the emergency call center. The system first used natural language processing (NLP) and geographic information systems (GIS) to analyze the person's extreme panic and precise location (e.g., near a beach), combining this with tidal conditions and weather conditions (e.g., high waves) to construct a comprehensive emotional location profile. Next, the system used deep reinforcement learning algorithms to evaluate various emergency response strategies, such as dispatching the nearest coast guard or instructing on-site personnel on CPR. Ultimately, the system selected the most suitable strategy—dispatching the coast guard—and combined this with medical knowledge graph technology to deeply analyze the relationship between the symptoms described by the person seeking help (e.g., loss of consciousness, difficulty breathing) and the potential drowning, confirming the injury as severe drowning requiring immediate treatment. Finally, the system generated a detailed first aid guide, including initial first aid measures (e.g., clearing airway obstructions), precautions before rescue vehicle arrival, and a subsequent hospital transport plan. Through these steps, the system not only improved the speed and accuracy of emergency response but also provided clear operational guidelines for on-site personnel, significantly enhancing rescue efficiency and success rates.

[0070] Optionally, step 102, based on the emergency situation awareness model, uses a deep reinforcement learning algorithm to select possible emergency response strategies, and combines medical knowledge graph technology to analyze the relationship between the requester's symptoms and possible causes to generate an emergency guidance plan, including:

[0071] Using the aforementioned emergency situation awareness model, the emotional state and location information of the requester are extracted and processed to obtain emotional location features. Based on these emotional location features, a deep reinforcement learning algorithm is used to evaluate various emergency response strategies to obtain the optimal response strategy. According to the optimal response strategy, combined with medical knowledge graph technology, the requester's symptoms and possible causes are analyzed for correlation to obtain the most relevant disease types and their urgency levels. Using the most relevant disease types and their urgency levels, an emergency guidance plan is generated.

[0072] In this step, emotional location features encompass the requester's emotional state and geographic location information extracted from the emergency situation awareness model, used to more comprehensively understand emergency needs; deep reinforcement learning algorithms are an advanced machine learning method that optimizes the decision-making process by simulating a large number of interaction scenarios to select the most appropriate emergency response strategy; and medical knowledge graph technology is a structured medical knowledge base used to assist in diagnosis and treatment recommendations, analyzing the complex relationship between symptoms and causes.

[0073] In this embodiment, firstly, an emergency situation awareness model is used to analyze the requester's multimedia information, extracting emotional state and location information to form emotional location features. Secondly, based on these features, a deep reinforcement learning algorithm is used to evaluate various possible emergency response strategies and select the optimal solution. Thirdly, medical knowledge graph technology is combined to deeply analyze the requester's symptoms and potential causes, determining the most relevant disease types and their urgency. Finally, based on the above analysis results, a scientific and personalized emergency guidance plan is generated to ensure the effectiveness and relevance of emergency measures.

[0074] Suppose a serious car accident occurs in a remote mountainous area, and the person seeking help sends a voice message with their current location via mobile phone. The system first receives and analyzes the caller's voice, identifying their extreme panic, and then uses a geographic information system to pinpoint the accident location. Next, the system uses deep reinforcement learning algorithms to evaluate different response strategies, such as dispatching helicopter rescue, ground ambulances, or assistance from nearby volunteers, ultimately selecting the most suitable helicopter rescue option. Then, the system combines medical knowledge graph technology to analyze the relationship between the symptoms described by the person seeking help (such as severe bleeding, confusion, etc.) and potential traumatic injuries, confirming the injury as severe and requiring immediate treatment. Finally, the system generates a detailed first aid guide, including initial on-site hemostasis measures, how to prepare for helicopter landing, and a subsequent hospital transfer plan. Through these steps, the system not only improves the speed and accuracy of emergency response but also provides clear operational guidelines for on-site personnel, significantly enhancing rescue efficiency and success rate.

[0075] Optionally, the step of using the emergency situation awareness model to extract and process the emotional state and location information of the requester to obtain emotional location features includes:

[0076] Natural language processing (NLP) technology is used to parse and process the voice or text information of emergency callers, identify emotion-related keywords and expressions, and obtain emotion state data. Based on the emotion state data and environmental factors, the location of the caller is accurately located to generate location information. The emotion state data and location information are then comprehensively analyzed to assess the correlation between emotion and geographical location, resulting in an emotion-location correlation analysis. Finally, the emotion-location correlation analysis results are fused to obtain emotion-location features.

[0077] Optionally, based on the optimal response strategy and combined with medical knowledge graph technology, the process of performing correlation analysis on the requester's symptoms and possible causes to obtain the most relevant disease types and their urgency includes:

[0078] Using the symptom identification information contained in the optimal response strategy, the requester's symptoms are initially classified to obtain symptom classification results. Based on the symptom classification results, and combined with the disease and symptom relationship database in the medical knowledge graph, the association between the requester's symptoms and known diseases is matched to generate a potential disease list. According to the potential disease list, the typical symptoms, common complications, and pathogenesis of each disease are further analyzed using medical knowledge graph technology to assess the probability and urgency of each disease, resulting in a disease probability and urgency assessment report. Using the disease probability and urgency assessment report, the most relevant disease type and its urgency are obtained.

[0079] In this embodiment, firstly, natural language processing technology is used to parse the voice or text information of the emergency caller, identify emotion-related keywords and expressions, and generate emotion state data. Secondly, combined with environmental factors, such as traffic conditions and weather conditions, the caller's location is precisely located to generate location information. Thirdly, based on the emotion state data and location information, a comprehensive analysis is performed to assess the correlation between emotion and geographical location, resulting in an emotion-location correlation analysis. Finally, these analysis results are fused to construct emotion-location features. For the optimal response strategy, firstly, the symptom identification information contained therein is used to perform preliminary classification of the caller's symptoms, resulting in symptom classification results. Secondly, based on the symptom classification results, combined with the disease and symptom relationship database in the medical knowledge graph, the correlation between the caller's symptoms and known diseases is matched to generate a potential disease list. Thirdly, based on the potential disease list, the typical symptoms, common complications, and pathogenesis of each disease are further analyzed to assess the probability and urgency, generating a disease probability and urgency assessment report. Finally, this assessment report is used to determine the most relevant disease type and its urgency.

[0080] Suppose a traffic accident occurs in a busy city center, and the person seeking help sends a voice message with their current location via mobile phone. The system first receives and analyzes the caller's voice, using natural language processing (NLP) to identify their extreme anxiety and extract keywords describing the severity of the accident. Next, combining this with the traffic conditions and weather, the system uses a geographic information system (GIS) to precisely locate the accident site and generate location information. Then, based on the emotional state data and location information, the system performs a comprehensive analysis, assessing the correlation between the caller's emotions and geographical location. It finds that the accident occurred during peak traffic hours and the caller exhibited high levels of anxiety, increasing the urgency of the rescue. Finally, the system integrates these analytical results to construct an emotional location feature, providing a solid foundation for subsequent decision-making. Subsequently, based on the symptom identification information in the optimal response strategy, the system performs a preliminary classification of the symptoms described by the person seeking help (such as severe pain and difficulty breathing), obtaining symptom classification results. Then, based on these symptom classification results, the system combines the disease and symptom relationship database in the medical knowledge graph to match possible diseases, such as multiple fractures and internal bleeding, generating a list of potential diseases. The system then conducts in-depth analysis of the typical symptoms, common complications, and pathogenesis of these potential diseases, assessing the probability and urgency of each disease. Ultimately, it confirms that the most relevant disease types are multiple fractures and internal bleeding, indicating that both are extremely urgent. Through these steps, the system not only improves the speed and accuracy of emergency response but also provides clear operational guidelines for on-site personnel, significantly improving rescue efficiency and success rate.

[0081] This application considers that, in existing technologies, emergency response systems typically rely on traditional telephone call reception and manual dispatch, supplemented by preliminary automatic location services. However, these systems have significant limitations: first, they lack the ability to understand complex situations, making it difficult to comprehensively and accurately assess the specific circumstances of an emergency request; second, their route planning and traffic prediction algorithms are not flexible enough to adapt to constantly changing road conditions in real time, leading to prolonged rescue times; and finally, they lack intuitive and personalized emergency guidance, often leaving requesters unable to perform correct self-rescue or mutual rescue operations before professional rescue forces arrive, thus affecting the effectiveness of emergency response. Therefore, this invention proposes this alternative solution to address the above problems, ensuring the speed, accuracy, and effectiveness of emergency response through more precise situational awareness, intelligent decision-making, and dynamic route optimization.

[0082] Optionally, the step of evaluating multiple first aid response strategies based on the emotional location features and obtaining the optimal response strategy using a deep reinforcement learning algorithm includes:

[0083] Before calculating the quality function Q(s,a), the requester's emotional state and location information are first analyzed using natural language processing technology and geographic information system, and environmental factors are combined to provide a comprehensive and accurate state description for the subsequent calculation of the quality function; the environmental factors include traffic conditions and weather conditions.

[0084] Q(s,a)=∑ s′ P(s′|s,a)[R(s,a,s′)+γmax a′ Q(s′,a′)+β·I(s,a,s′)];

[0085] Where Q(s,a) is the quality function value of taking action a in state s, representing the expected long-term reward after taking action a; P(s'|s,a) is the state transition probability, referring to the probability of transitioning from the current state s to the next state s' by taking action a; R(s,a,s') represents the immediate reward obtained by reaching state s' after taking action a from state s; γ is a discount factor used to weigh the importance of future rewards, ranging from 0 to 1; max a' Q(s',a') is the maximum quality function value that can be achieved by taking the best action a' in the new state s'; β is an influence coefficient used to adjust the degree of influence of the immediate information I(s,a,s') on the final decision; I(s,a,s') represents the immediate situational information brought about by the emotional position characteristics when arriving at state s' from state s through action a;

[0086] After calculating Q(s,α), the urgency coefficient α needs to be dynamically adjusted based on the emotional position characteristics of the requester, and the situational consistency C(s,a) after taking different actions needs to be evaluated to ensure that the selected actions are not only efficient but also situationally stable, thereby determining the optimal action α. * ;

[0087] a * =argmax a {Q(s,a)+α·ΔE(s,a)+λ·C(s,a)};

[0088] Where, α * Q(s,a) is the optimal action selected based on the current state s, which maximizes the combined value of the quality function plus urgency adjustment and situational consistency. Q(s,a) is the quality function value of taking action a in state s. α is the urgency adjustment coefficient, used to adjust the urgency of the strategy based on the emotional and locational characteristics of the requester, reflecting the influence of emotional and locational information on the priority of emergency response strategies. ΔE(s,a) is the energy difference, reflecting the potential impact after strategy execution by analyzing the energy difference between emotional and locational characteristics and the expected changes after taking action a. λ is the situational consistency coefficient, used to measure the consistency level of the situation after taking action a, where the situation includes emotional stability and the rescue environment. C(s,a) represents the situational consistency score after taking action a in state s, which helps ensure that the selected action considers not only efficiency but also situational stability.

[0089] Determine the optimal action a * Then, it is transformed into a specific first aid guidance plan, and the feedback information during the implementation process is continuously monitored to dynamically adjust the strategy, and finally a scientific and efficient first aid response strategy is generated.

[0090] This formula aims to overcome the limitations of existing technologies by introducing a deep reinforcement learning algorithm combined with emotional location features to select the optimal response strategy. Specifically, it evaluates the effectiveness of different first aid response strategies by calculating a quality function Q(s,a), and determines the optimal action a based on this. * This leads to the generation of scientific and efficient emergency response strategies. This approach not only improves the scientific rigor and personalization of decision-making but also enhances the effectiveness and specificity of emergency measures, ultimately achieving efficient and precise emergency dispatch and providing strong technical support for saving lives.

[0091] The following is a brief introduction to the design rationale behind each term of the formula:

[0092] Q(s,a)=∑ s′ P(s′|s,a)[R(s,a,s′)+γmax a′Q(s′,a′)+β·I(s,a,s′)]

[0093] State transition probability P(s'|s,a): Represents the probability of transitioning from the current state to the next state, used to measure the likelihood of reaching different states after taking an action; Immediate reward R(s,a,s'): Represents the reward obtained immediately after taking an action, reflecting the direct benefit brought by the action; γmax a' Q(s',a'): This term is designed to measure long-term benefits; β·I(s,a,s'): This term is designed to capture and utilize the impact of immediate contextual information on the final decision.

[0094] The following is a brief introduction to how the parameters of this formula are obtained:

[0095] P(s'|s,a): Derived through statistical analysis of historical data; R(s,a,s'): Set according to actual conditions, such as a positive reward for reduced arrival time; γ: Preset value, usually close to but less than 1; max a' Q(s',a'): Calculated recursively; β: Adjusted based on experiments to ensure that real-time information has an appropriate impact on decision-making; I(s,a,s'): Calculated by analyzing the requester's emotional state and location information using natural language processing technology and geographic information systems, combined with environmental factors (such as traffic conditions and weather conditions).

[0096] The following is a brief introduction to the design rationale behind each term of the formula:

[0097] a * =arg max a {Q(s,a)+α·ΔE(s,a)+λ·C(s,a)};

[0098] Quality function Q(s,a): The quality function value of taking action a in state s; Urgency adjustment coefficient α: Used to adjust the urgency of the strategy based on the emotional and locational characteristics of the requester, reflecting the influence of emotional and locational information on the priority of emergency response strategies; Energy difference ΔE(s,a): Reflects the potential impact after strategy execution by analyzing the energy difference between emotional and locational characteristics and the expected changes after taking the action; Situational consistency coefficient λ: Used to measure the consistency level of the situation after taking the action, where the situation includes emotional stability and the rescue environment; Situational consistency score C(s,a): Represents the situational consistency score after taking action a in state s, which helps ensure that the selected action considers not only efficiency but also situational stability.

[0099] The following is a brief introduction to how the parameters of this formula are obtained:

[0100] α: Dynamically adjusted based on emotional location characteristics to ensure that the degree of urgency is appropriately reflected; ΔE(s,a): Determined by analyzing the energy difference between emotional location characteristics and expected changes after taking action; λ: Adjusted based on experiments to ensure that situational consistency has an appropriate impact on decision-making; C(s,a): Determined by comprehensively evaluating emotional stability and the rescue environment.

[0101] Suppose a traffic accident occurs in a busy city center, and the person seeking help sends a voice message and their current location via mobile phone. First, the system uses natural language processing technology and geographic information systems to analyze the caller's voice, identifying their extreme anxiety. Combined with the traffic conditions (peak hour) and weather conditions (clear skies), it provides a comprehensive and accurate state description for subsequent quality function calculations. Then, the system calculates the quality function Q(s,a)=∑ s' P(s'|s,a)[R(s,a,s')+0.9max a' Q(s',a')+0.8·I(s,a,s')], where P(s'|s,a) is 0.7 obtained from historical data statistics, R(s,a,s') is set to 5 (timely arrival receives a positive reward), γ=0.9,max a' Q(s',a') is calculated recursively to be 10, β = 0.8, and I(s,a,s') is calculated to be 3 through emotional location feature analysis. Next, the system dynamically adjusts the urgency coefficient α = 1.2 based on the emotional location features of the person seeking help, and evaluates the situational consistency C(s,a) after taking different actions, ultimately determining the optimal action a. * =argmax a The formula is {Q(s,a)+1.2·ΔE(s,a)+0.6·C(s,a)}, where ΔE(s,a) is calculated as 2 by analyzing the energy difference between the emotional location characteristics and the expected change after taking action, λ=0.6, and C(s,a) is calculated as 4 by comprehensively evaluating emotional stability and the rescue environment. Finally, the system converts the optimal action into a specific first aid guidance plan, continuously monitors feedback information during implementation, dynamically adjusts the strategy, and ultimately generates a scientific and efficient first aid response strategy. Through these steps, the system not only improves the speed and accuracy of first aid response but also ensures information transparency and smooth communication, significantly improving the overall first aid effect and success rate.

[0102] This approach ensures the scientific and personalized nature of emergency response strategies through precise contextual awareness and intelligent decision-making. Specifically, by initializing rescue routes and accurately predicting traffic conditions, the system can avoid potential traffic obstacles in advance, significantly shortening rescue time. Combined with IoT technology to update road condition information in real time and dynamically optimize driving routes, it further improves rescue efficiency. At the same time, continuous monitoring of actual driving conditions and communication with the requester ensures information transparency and timely feedback, enhancing the requester's sense of security and trust. Ultimately, it achieves efficient and precise emergency dispatch, significantly improving the overall rescue effect and user experience.

[0103] 103. Based on the aforementioned emergency rescue guidance scheme, the rescue route is optimized in real time using a traffic flow prediction algorithm. Combined with road condition information collected by Internet of Things technology, the driving route of the rescue vehicle is dynamically adjusted to generate the optimal rescue route and the latest rescue progress notification.

[0104] In this step, traffic flow prediction algorithms are used to predict vehicle flow on the road network to help plan the shortest and safest travel routes; Internet of Things (IoT) technology collects real-time traffic information, such as traffic flow, weather conditions, and emergencies, through sensors and other devices to ensure the flexibility and accuracy of route planning.

[0105] In this embodiment, based on the generated emergency medical guidance plan, the system optimizes the rescue route in real time through a traffic flow prediction algorithm and dynamically adjusts the route of the rescue vehicle by combining the latest road condition information collected by IoT technology, ensuring that it avoids congested sections and reaches the destination quickly. Simultaneously, the system continuously updates the rescue progress and sends notifications to the requester, maintaining information transparency and smooth communication.

[0106] Suppose a traffic accident occurs in a suburban area of ​​a city, and the person seeking help sends an emergency call via mobile phone. The system first receives and parses the urgency and location information from the emergency guidance plan, initializing the planned rescue route to determine the initial path from the nearest fire station to the accident site. Second, based on this initial path, the system analyzes current and anticipated traffic conditions using a traffic flow prediction algorithm. It detects nearby road construction causing increased traffic flow, predicting this will significantly prolong rescue time, thus generating a traffic condition prediction result. Third, based on these predictions, the system combines real-time road condition information collected using IoT technology to identify a relatively smooth alternative route, dynamically adjusting the rescue vehicle's route to avoid construction areas, thus obtaining the optimal rescue route. Finally, the system uses this optimal rescue route to continuously monitor the actual driving situation of the rescue vehicle, while maintaining close communication with the person seeking help, promptly sending the latest rescue progress notifications, informing them of the estimated arrival time and precautions. Through these steps, the system not only improves the speed and accuracy of the rescue response but also ensures information transparency and smooth communication, significantly improving overall rescue efficiency and success rate.

[0107] Optionally, step 103, based on the emergency rescue guidance plan, involves real-time optimization of the rescue route using a traffic flow prediction algorithm, dynamically adjusting the rescue vehicle's route using road condition information collected via IoT technology, and generating an optimal rescue route and the latest rescue progress notification, including:

[0108] Using the urgency and location information provided in the emergency rescue guidance plan, the initially planned rescue route is initialized to obtain an initial rescue route. Based on the initial rescue route, a traffic flow prediction algorithm is used to analyze and process the current and expected traffic conditions, predict factors that may affect the rescue time, and generate a traffic condition prediction result. According to the traffic condition prediction result, combined with real-time road condition information collected by IoT technology, the driving route of the rescue vehicle is dynamically adjusted to obtain the optimal rescue route. Using the optimal rescue route, the actual driving situation is continuously monitored, communication is maintained with the requester, and the latest rescue progress notification is generated.

[0109] In this step, the emergency guidance plan includes a preliminary rescue strategy generated based on the urgency and location information of the requester to guide the rescue operation; a traffic flow prediction algorithm is a technology that analyzes vehicle flow on the road network to help plan the optimal driving route; Internet of Things (IoT) technology collects real-time road condition information, such as traffic flow, weather conditions, and emergencies, through sensors and other devices to ensure the flexibility and accuracy of route planning; and rescue progress notifications refer to the system sending the requester the latest information on the rescue progress, maintaining information transparency and smooth communication.

[0110] In this embodiment, the initial rescue route is first initialized using the urgency and location information provided in the emergency guidance plan. Then, based on this initial rescue route, a traffic flow prediction algorithm is used to analyze the current and anticipated traffic conditions, predict factors that may affect rescue time, and generate traffic condition prediction results. Next, based on these prediction results and combined with real-time road condition information collected using IoT technology, the route of the rescue vehicle is dynamically adjusted to optimize the route and avoid congestion and obstacles. Finally, the optimized rescue route is used to continuously monitor the actual driving situation and maintain communication with the requester, providing timely updates on the latest rescue progress.

[0111] Suppose a sudden heart attack occurs in a busy commercial area of ​​a city, and the person seeking help sends an emergency call via mobile phone. The system first receives and parses the urgency and location information from the emergency guidance plan, initializing the planned rescue route to determine the initial path from the nearest hospital to the incident location. Next, based on this initial route, the system analyzes current and anticipated traffic conditions using a traffic flow prediction algorithm. It detects a large event nearby causing increased traffic flow, predicting this will significantly prolong rescue time, thus generating a traffic condition prediction result. Then, based on these predictions, the system combines real-time traffic information collected using IoT technology to identify a smoother alternative route, dynamically adjusting the rescue vehicle's route to avoid congested areas, thus obtaining the optimal rescue route. Finally, the system uses this optimal rescue route to continuously monitor the actual driving situation of the rescue vehicle, while maintaining close communication with the person seeking help, sending timely updates on the rescue progress, including estimated arrival time and precautions. Through these steps, the system not only improves the speed and accuracy of the rescue response but also ensures information transparency and smooth communication, significantly improving overall rescue efficiency and success rate.

[0112] Optionally, the step of dynamically adjusting the route of the rescue vehicle based on the traffic condition prediction results and in conjunction with real-time road condition information collected using IoT technology to obtain the optimal rescue route includes:

[0113] Using the traffic condition prediction results, factors that may affect rescue time are identified and processed to obtain potential traffic influencing factors. Based on these potential traffic influencing factors, and combined with real-time road condition information collected using IoT technology, a comprehensive analysis is performed to generate real-time road condition updates. According to the real-time road condition updates, the driving route of the rescue vehicle is replanned to avoid unfavorable road conditions, resulting in an adjusted driving route. Using the adjusted driving route, continuous optimization is performed until the shortest time and best safety conditions are achieved, generating the optimal rescue path.

[0114] In this step, traffic condition prediction results include analysis of current and expected traffic flow, accidents, weather, and other influencing factors to predict factors that may affect rescue time; IoT technology collects specific road condition information in real time through sensors and other devices, such as traffic flow, road construction, and traffic accidents, to ensure the flexibility and accuracy of route planning; real-time traffic updates refer to generating the latest road condition report by combining traffic condition prediction results and data collected by IoT technology; the optimal rescue route is the path that achieves the shortest time and best safety conditions through continuous optimization of the driving route.

[0115] In this embodiment, firstly, traffic condition prediction results are used to identify factors that may affect rescue time, such as traffic congestion, accidents, or severe weather, thus obtaining potential traffic influencing factors. Secondly, based on these potential factors, combined with real-time road condition information collected by IoT technology, a comprehensive analysis is performed to generate real-time road condition updates. Thirdly, based on the real-time road condition updates, the driving route of the rescue vehicle is replanned to avoid unfavorable road conditions, and the adjusted driving route is more conducive to reaching the destination quickly. Finally, using this adjusted driving route, the system continuously optimizes until the shortest time and best safety conditions are achieved, ultimately generating the optimal rescue path.

[0116] Suppose an emergency medical incident occurs in a city center, and the person seeking help is located near an area hosting a large event. The system first uses traffic condition predictions to identify factors that may affect rescue time, such as traffic congestion and temporary road closures caused by the event, thus identifying potential traffic influencing factors. Second, based on these potential factors, and combined with real-time road condition information collected using IoT technology, such as traffic camera images and data from intelligent transportation systems, the system performs comprehensive analysis, discovering that several major roads are closed, generating real-time traffic updates. Third, based on these updates, the system replans the rescue vehicle's route, selecting an unaffected alley as an alternative route to avoid adverse road conditions, resulting in an adjusted route. Finally, using this adjusted route, the system continuously optimizes and evaluates road capacity at different times until it finds the path with the shortest time and best safety conditions, ultimately generating the optimal rescue route. Through these steps, the system not only ensures that rescue vehicles can arrive at the scene quickly and safely but also effectively avoids delays caused by traffic problems, significantly improving rescue efficiency and success rate.

[0117] This application recognizes that in existing technologies, emergency response systems typically rely on static route planning and traffic prediction algorithms, which have significant limitations: First, they lack the ability to understand complex situations, making it difficult to comprehensively and accurately assess the specific circumstances of an emergency request; second, existing route planning and traffic prediction algorithms are not flexible enough to adapt to constantly changing road conditions in real time, leading to prolonged rescue times; and finally, they lack a dynamic adjustment mechanism, making it impossible to update rescue strategies in a timely manner in the face of sudden traffic incidents. Therefore, this invention proposes this alternative solution to address the above problems, ensuring the speed, accuracy, and efficiency of emergency response through more accurate traffic flow prediction and dynamic route optimization.

[0118] Optionally, based on the initial rescue route, the current and expected traffic conditions are analyzed and processed using a traffic flow prediction algorithm to predict factors that may affect the rescue time, generating traffic condition prediction results, including:

[0119] Before calculating the expected traffic flow Q(t), it is necessary to collect and analyze historical traffic data, real-time traffic information, weather forecasts, and reports of events that may affect traffic. Combine this with a geographic information system to assess the impact of these factors on a specific road segment, providing a comprehensive basis for subsequent traffic flow prediction.

[0120]

[0121] Where Q(t) is the expected traffic flow at time t; β0 is the baseline traffic flow level; β i Influencing factor X i (t) is the influence coefficient on traffic flow, X i (t) represents the influencing factors that change over time, including weather conditions and special events; γ, ω, φ, and λ represent the amplitude, angular frequency, phase shift, and attenuation rate of the periodic component, respectively, used to capture the periodic and trend components in traffic flow; δ is the amplitude of the S-shaped function, used to simulate the impact of sudden events on traffic flow, including traffic accidents; μ is the slope of the S-shaped function, controlling the response speed; τ is the time shift of the S-shaped function, indicating the specific time point when the event occurs.

[0122] After calculating the expected traffic flow Q(t), this result is used in conjunction with the total distance D along the rescue route and the average vehicle speed V. avg By combining the parameters of (t) and velocity change ΔV(t), the vehicle travel time T(t) is predicted to ensure that the rescue operation can reach the scene in a timely manner;

[0123]

[0124] Where T(t) is the estimated travel time from the starting point to the destination; D is the total distance along the rescue route; Vavg (t) is the average vehicle speed on the road at time t; α is an adjustment parameter reflecting the basic impact of traffic flow on travel time; κ is the cumulative impact coefficient, describing the cumulative effect of traffic flow on travel time; ρ is the speed change impact coefficient, used to measure the impact of the ratio of current speed to maximum permissible speed on travel time; ΔV(t) is the speed change at time t; V max It is the maximum permissible speed on the road;

[0125] After calculating the estimated travel time T(t), the rescue route is dynamically adjusted to avoid high-traffic or congested areas, and the deviation between the actual traffic conditions and the prediction is continuously monitored to update the prediction model; real-time traffic condition updates and estimated arrival times are sent to emergency requesters and response units to generate detailed traffic condition prediction results, supporting efficient rescue dispatch.

[0126] This application aims to overcome the limitations of existing technologies by introducing a dynamic path optimization scheme based on traffic flow prediction algorithms. Specifically, it first collects and analyzes historical traffic data, real-time traffic information, weather forecasts, and reports of events that may affect traffic, combining this with a geographic information system to assess the impact of these factors on specific road segments, providing a comprehensive foundation for subsequent traffic flow prediction. Then, using the calculation of expected traffic flow Q(t) and vehicle travel time T(t), it predicts factors that may affect rescue time and generates detailed traffic condition prediction results. This approach not only improves the scientific rigor and accuracy of decision-making but also enhances the effectiveness and targeted nature of emergency response measures, ultimately achieving efficient and precise emergency dispatch and providing strong technical support for saving lives.

[0127] The following is a brief introduction to the design rationale behind each term of the formula:

[0128]

[0129] Basic traffic flow level β0: Represents the basic traffic flow under conditions without other influencing factors, used to set a baseline value; sum of influencing factors It reflects the comprehensive impact of different factors (such as weather conditions and special events) on traffic flow, ensuring that the prediction model can adapt to various changes in the external environment; the periodic component γ·sin(ωt+φ)·exp(-λt) captures the periodic and trend components in traffic flow, simulating changes in daily traffic patterns; S-shaped function Simulate the impact of sudden events (such as traffic accidents) on traffic flow to ensure that the model can respond quickly to emergencies.

[0130] The following is a brief introduction to how the parameters of this formula are obtained:

[0131] β0: Derived through historical data analysis; β i : Determined based on the historical impact of influencing factors; X i (t): Collected through real-time sensors and event reports; γ,ω,φ,λ: Derived by fitting historical traffic data; δ: Set according to the impact range and intensity of the emergency; μ: Adjusted according to the response speed of the emergency; τ: Determined according to the occurrence time of the emergency.

[0132] The following is a brief introduction to the design rationale behind each term of the formula:

[0133]

[0134] Distance to average speed ratio The calculation of basic travel time based on a fixed distance and average speed provides a basic time estimate; traffic flow cumulative effect To measure the cumulative impact of traffic flow on travel time, the model must account for changes in traffic conditions over a long period; the impact of speed changes... Assess the impact of current speed changes on travel time to ensure the model can adapt to sudden speed changes.

[0135] The following is a brief introduction to how the parameters of this formula are obtained:

[0136] D: Determined through a geographic information system; V avg (t): Acquired from real-time traffic data; α: Set based on the impact of traffic flow on travel time; κ: Obtained by fitting historical traffic data; ρ: Set based on the impact of speed changes on travel time; ΔV(t): Acquired from real-time sensor data; V max : Set according to road regulations.

[0137] Suppose an emergency medical incident occurs in a city center, and the person seeking help is located near an area hosting a large event. First, the system collects and analyzes historical traffic data, real-time traffic information, weather forecasts, and event reports that may affect traffic. Combining this with a Geographic Information System (GIS), it assesses the impact of these factors on a specific road segment, providing a comprehensive basis for subsequent traffic flow prediction. Then, the system calculates the expected traffic flow Q(t) = 500 + 0.8 × weather conditions + 1.2 × impact level, obtaining γ = 200, ω = 0.1, φ = 0.5, λ = 0.05, δ = 300, μ = 0.2, τ = 10. Finally, the system uses this result along with the total distance D = 10 km along the rescue route and the average vehicle speed V... avg By combining the parameters ΔV(t) = 30 km / h and the speed change ΔV(t) = 5 km / h, the vehicle travel time can be predicted. Among them, α=0.5, κ=0.1, ρ=0.2, V max =60km / h, the final calculated T(t) = 0.4 hours; assuming a threshold of 0.3 hours, the result T(t) = 0.4 hours is greater than this threshold, indicating that the system fully considered the impact of traffic flow on travel time during decision-making, avoiding delays; the result Q(t) also reflects the significant impact of emergencies on traffic flow, making the decision more aligned with actual needs. These two factors work together to ensure that the system can make scientific and efficient decisions in complex and ever-changing emergency response environments. After calculating the estimated travel time T(t), the system dynamically adjusts the rescue route to avoid high-flow or congested areas, continuously monitors the deviation between actual traffic conditions and predictions, and updates the prediction model; it sends real-time traffic condition updates and estimated arrival times to emergency requesters and response units, generating detailed traffic condition prediction results to support efficient rescue dispatch. Through these steps, the system not only improves the speed and accuracy of rescue response but also ensures information transparency and smooth communication, significantly improving the overall emergency response effectiveness and success rate.

[0138] This method ensures the speed, accuracy, and efficiency of emergency response through precise traffic flow prediction and dynamic route optimization. Specifically, by initializing rescue routes and accurately predicting traffic conditions, the system can avoid potential traffic obstacles in advance, significantly shortening rescue time. Combining IoT technology with real-time updates of road conditions and dynamic route optimization further improves rescue efficiency. Simultaneously, continuous monitoring of actual driving conditions and communication with the requester ensures information transparency and timely feedback, enhancing the requester's sense of security and trust. Ultimately, this achieves efficient and precise emergency dispatch, significantly improving the overall rescue effect and user experience.

[0139] 104. Using the optimal rescue route and the latest rescue progress notification, visualize the emergency rescue operation process to generate intuitive visual guidance.

[0140] In this step, visualization refers to converting complex first aid procedures into easily understandable and executable forms, such as animated demonstrations, graphical interfaces, or augmented reality displays. Intuitive visual guidance aims to clearly convey first aid steps through visual means, helping non-professionals to perform initial self-rescue or mutual rescue while waiting for professional assistance.

[0141] In this embodiment, the system visualizes the emergency rescue process using the generated optimal rescue route and the latest rescue progress notification, generating intuitive visual guidance, including text descriptions, animation demonstrations, and voice prompts, to help the caller perform effective self-rescue or mutual rescue while waiting for professional rescue, thereby improving the effectiveness and success rate of emergency rescue.

[0142] Imagine a sudden cardiac arrest occurring in a high-rise building. The person seeking help sends an emergency call request via the building's emergency call button. The system first uses the pre-determined optimal rescue route, combined with detailed building floor plan data, to visualize the rescue elevator's path, generating a dynamic route map. Second, based on this dynamic route map, the system updates and displays the rescue elevator's current location and estimated arrival time in real-time, providing a real-time progress display so the person seeking help can clearly understand the rescue's progress. Third, based on the real-time progress display, and combined with the person seeking help's specific symptoms (such as severe chest pain) and the previously generated first-aid guidance plan, the system uses augmented reality technology to demonstrate first-aid procedures (such as CPR), generating a series of first-aid demonstration sequences. Finally, using these demonstration sequences, the system generates intuitive visual guidance including text descriptions, animations, and voice prompts, helping the person seeking help and their companions to effectively perform self-rescue or mutual rescue while waiting for rescue. Through these steps, the system not only provides clear operational guidelines but also enhances the person seeking help's confidence, significantly improving the effectiveness and success rate of first aid.

[0143] Optionally, step 104 involves using the optimal rescue route and the latest rescue progress notification to visualize the emergency response process and generate intuitive visual guidance, including:

[0144] Using the optimal rescue route, the travel route of the rescue vehicle is visualized by combining map data to obtain a dynamic route map. Based on the dynamic route map, the current location and estimated arrival time of the rescue vehicle are updated and displayed in real time according to the latest rescue progress notification, generating a real-time progress display. Based on the real-time progress display, combined with the specific symptoms of the requester and the generated first aid guidance plan, the first aid operation steps are demonstrated using augmented reality technology to obtain a first aid operation demonstration sequence. Using the first aid operation demonstration sequence, intuitive visual guidance is generated.

[0145] In this step, the optimal rescue route is the driving route that achieves the shortest time and best safety conditions after optimization through traffic flow prediction and IoT technology; the dynamic route map refers to the result of visualizing the rescue vehicle's route by combining map data, used to intuitively display the rescue progress; the real-time progress display updates and displays the current location and estimated arrival time of the rescue vehicle in real time based on the latest rescue progress notification; the first aid operation demonstration sequence is the result of demonstrating the first aid operation steps using augmented reality (AR) technology, helping the requester or on-site personnel to carry out effective self-rescue or mutual rescue while waiting for professional rescue; and the intuitive visual guidance refers to integrating all the above information to generate easy-to-understand and implement visual first aid guidance.

[0146] In this embodiment, firstly, the optimal rescue route is utilized, combined with map data, to visualize the rescue vehicle's travel route and generate a dynamic route map. Secondly, based on this dynamic route map, the current location and estimated arrival time of the rescue vehicle are updated and displayed in real time according to the latest rescue progress notification, forming a real-time progress display. Thirdly, based on the real-time progress display, combined with the requester's specific symptoms and the generated first aid guidance plan, the first aid operation steps are demonstrated using augmented reality technology to obtain a first aid operation demonstration sequence. Finally, these demonstration sequences are used to generate intuitive visual guidance to ensure that the requester can clearly and accurately perform initial first aid operations.

[0147] Suppose a hiker is injured in a remote mountainous area. The system first uses the established optimal rescue route, combined with detailed mountain map data, to visualize the rescue helicopter's flight path, generating a dynamic route map. Second, based on this dynamic route map, the system updates and displays the rescue helicopter's current location and estimated arrival time in real time, providing a real-time progress display so the person in need of help can clearly understand the rescue's progress. Third, based on the real-time progress display, and combined with the hiker's specific symptoms (such as fractures) and previously generated first-aid guidelines, the system uses augmented reality technology to demonstrate first-aid procedures (such as immobilizing injured limbs), generating a series of first-aid demonstration sequences. Finally, using these demonstration sequences, the system generates intuitive visual guidance including text descriptions, animations, and voice prompts to help the hiker and their companions effectively perform self-rescue or mutual aid while awaiting rescue. Through these steps, the system not only provides clear operational guidelines but also enhances the hiker's confidence, significantly improving the effectiveness and success rate of first aid.

[0148] Through the systematic process of steps 101 to 104, this solution achieves intelligent processing of the entire process from receiving an emergency request to completing a rescue, significantly improving the speed, accuracy, and efficiency of emergency response. This method not only overcomes many limitations of existing emergency response systems but also significantly improves the overall quality of emergency care and user experience through intelligent, personalized, and dynamic processing methods, providing strong technical support for saving lives.

[0149] Figure 2 This application provides a schematic diagram of the structure of a human-computer interaction system in an emergency rescue scenario, as shown in the embodiment of the present application. Figure 2 As shown, the device includes:

[0150] The parsing module 21 is used to analyze and process the multimedia information of the emergency caller using multimodal sentiment analysis technology and geographic information system to obtain an emergency situation perception model.

[0151] Selection module 22 is used to select possible emergency response strategies based on the emergency situation perception model using a deep reinforcement learning algorithm, and to generate an emergency guidance plan by combining medical knowledge graph technology to analyze the relationship between the requester's symptoms and possible causes.

[0152] The adjustment module 23 is used to optimize the rescue route in real time based on the emergency rescue guidance plan by using a traffic flow prediction algorithm, and dynamically adjust the driving route of the rescue vehicle by combining road condition information collected by Internet of Things technology, so as to generate the optimal rescue route and the latest rescue progress notification.

[0153] The processing module 24 is used to visualize the emergency rescue operation process using the optimal rescue route and the latest rescue progress notification, and generate intuitive visual guidance.

[0154] Figure 2 The aforementioned human-computer interaction system for emergency rescue scenarios can perform... Figure 1 The implementation principle and technical effects of the human-computer interaction method in an emergency rescue scenario described in the above embodiment will not be repeated here. The specific methods by which each module and unit of the human-computer interaction system in an emergency rescue scenario performs operations have been described in detail in the embodiments related to this method, and will not be elaborated upon here.

[0155] In one possible design, Figure 2 The human-computer interaction system for an emergency rescue scenario shown in the embodiment can be implemented as a computing device, such as... Figure 3 As shown, the computing device may include a storage component 31 and a processing component 32;

[0156] The storage component 31 stores one or more computer instructions, wherein the one or more computer instructions are invoked and executed by the processing component 32.

[0157] The processing component 32 is used to: parse and process the multimedia information of the received emergency requester using multimodal sentiment analysis technology and geographic information system to obtain an emergency situation awareness model; based on the emergency situation awareness model, use deep reinforcement learning algorithm to select possible emergency response strategies, and combine medical knowledge graph technology to analyze the relationship between the requester's symptoms and possible causes to generate an emergency guidance plan; based on the emergency guidance plan, use traffic flow prediction algorithm to optimize the rescue route in real time, and combine road condition information collected by Internet of Things technology to dynamically adjust the driving route of the rescue vehicle to generate the optimal rescue route and the latest rescue progress notification; and use the optimal rescue route and the latest rescue progress notification to visualize the emergency operation process and generate intuitive visual guidance.

[0158] The processing component 32 may include one or more processors to execute computer instructions to complete all or part of the steps in the above-described method. Alternatively, the processing component may be implemented as one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the above-described method.

[0159] Storage component 31 is configured to store various types of data to support operations at the terminal. The storage component can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0160] Of course, computing devices may also include other components, such as input / output interfaces, display components, communication components, etc.

[0161] Input / output interfaces provide interfaces between processing components and peripheral interface modules, which can be output devices, input devices, etc.

[0162] The communication components are configured to facilitate wired or wireless communication between computing devices and other devices.

[0163] The computing device can be a physical device or an elastic computing host provided by a cloud computing platform. In this case, the computing device can refer to a cloud server, and the aforementioned processing components, storage components, etc., can be basic server resources rented or purchased from the cloud computing platform.

[0164] This application also provides a computer storage medium storing a computer program, which, when executed by a computer, can perform the above-described functions. Figure 1 The embodiment shown illustrates a human-computer interaction method in an emergency rescue scenario.

[0165] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0166] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0167] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0168] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A human-computer interaction method for emergency rescue scenarios, characterized in that, include: Using multimodal sentiment analysis technology and geographic information system, the multimedia information received from emergency callers is analyzed and processed to obtain an emergency situation perception model; Based on the emergency situation awareness model, a deep reinforcement learning algorithm is used to select and process possible emergency response strategies. Combined with medical knowledge graph technology, the relationship between the requester's symptoms and possible causes is analyzed to generate an emergency guidance plan. Based on the aforementioned emergency rescue guidance scheme, the rescue route is optimized in real time using a traffic flow prediction algorithm. Combined with road condition information collected by Internet of Things technology, the driving route of the rescue vehicle is dynamically adjusted to generate the optimal rescue route and the latest rescue progress notification. By utilizing the optimal rescue route and the latest rescue progress notifications, the emergency rescue operation process is visualized to generate intuitive visual guidance. The analysis process includes: using natural language processing technology to analyze the voice or text information of the emergency caller, identifying emotion-related keywords and expressions to obtain emotion state data; based on the emotion state data and environmental factors, accurately locating the caller's position to generate location information; performing comprehensive analysis based on the emotion state data and location information to assess the correlation between emotion and geographical location, obtaining emotion-location correlation analysis results; and using the emotion-location correlation analysis results for fusion processing to obtain emotion-location features. The selection process includes: based on the emotional location features, using a deep reinforcement learning algorithm to evaluate multiple first aid response strategies to obtain the optimal response strategy; The dynamic adjustment process includes: using the optimal rescue route, continuously monitoring the actual driving situation, maintaining communication with the requester, and generating the latest rescue progress notification.

2. The method according to claim 1, characterized in that, The process involves using the emergency situation awareness model, employing deep reinforcement learning algorithms to select and process possible emergency response strategies, and combining medical knowledge graph technology to analyze the relationship between the requester's symptoms and possible causes to generate an emergency guidance plan, including: Using the aforementioned emergency situation perception model, the emotional state and location information of the requester are extracted and processed to obtain emotional location features; Based on the optimal response strategy and combined with medical knowledge graph technology, the requester's symptoms and possible causes are analyzed for correlation to obtain the most relevant disease type and its urgency. Using the most relevant disease types and their urgency levels, a set of first aid guidelines is generated.

3. The method according to claim 2, characterized in that, Based on the optimal response strategy and combined with medical knowledge graph technology, the requester's symptoms and possible causes are analyzed for correlation to obtain the most relevant disease types and their urgency, including: Using the symptom identification information contained in the optimal response strategy, the requester's symptom presentation is preliminarily classified to obtain the symptom classification result; Based on the symptom classification results, and combined with the disease and symptom relationship database in the medical knowledge graph, the association between the requester's symptoms and known diseases is matched to generate a list of potential diseases. Based on the list of potential diseases, medical knowledge graph technology is used to further analyze the typical symptoms, common complications and pathogenesis of each disease, assess the probability and urgency of each disease, and obtain a disease probability and urgency assessment report. Using the aforementioned disease probability and urgency assessment report, the most relevant disease types and their urgency levels are obtained.

4. The method according to claim 1, characterized in that, Based on the aforementioned emergency rescue guidance scheme, the rescue route is optimized in real time using a traffic flow prediction algorithm. Combined with road condition information collected via IoT technology, the route of the rescue vehicle is dynamically adjusted to generate the optimal rescue route and the latest rescue progress notification, including: Using the urgency and location information provided in the emergency guidance plan, the initially planned rescue route is initialized to obtain the initial rescue route; Based on the initial rescue route, the current and expected traffic conditions are analyzed and processed using a traffic flow prediction algorithm to predict factors that may affect the rescue time and generate traffic condition prediction results. Based on the traffic condition prediction results and combined with real-time road condition information collected by Internet of Things technology, the driving route of the rescue vehicle is dynamically adjusted to obtain the optimal rescue route.

5. The method according to claim 4, characterized in that, The step of dynamically adjusting the route of the rescue vehicle based on the traffic condition prediction results and real-time road condition information collected by Internet of Things (IoT) technology to obtain the optimal rescue route includes: Using the traffic condition prediction results, factors that may affect rescue time are identified and processed to obtain potential traffic influencing factors; Based on the aforementioned potential traffic influencing factors, and combined with real-time road condition information collected using IoT technology, a comprehensive analysis and processing is performed to generate real-time road condition updates. Based on the real-time traffic updates, the driving routes of the rescue vehicles are replanned to avoid unfavorable road conditions, resulting in adjusted driving routes. Using the adjusted driving route, the system is continuously optimized until the shortest time and best safety conditions are achieved, thus generating the optimal rescue route.

6. The method according to claim 1, characterized in that, The process of visualizing the emergency rescue operation using the optimal rescue route and the latest rescue progress notifications to generate intuitive visual guidance includes: Using the optimal rescue route and combining it with map data, the travel route of the rescue vehicle is visualized to obtain a dynamic route map; Based on the dynamic route map, the current location and estimated arrival time of the rescue vehicles are updated and displayed in real time according to the latest rescue progress notification, generating a real-time progress display; Based on the real-time progress display, combined with the requester's specific symptoms and the generated first aid guidance plan, the first aid operation steps are demonstrated using augmented reality technology to obtain a first aid operation demonstration sequence. The aforementioned first aid procedure demonstration sequence is used to generate intuitive visual instructions.

7. A human-computer interaction system for emergency rescue scenarios, used to execute the human-computer interaction method for emergency rescue scenarios as described in any one of claims 1 to 6, characterized in that, include: The parsing module is used to analyze and process the multimedia information received from emergency callers using multimodal sentiment analysis technology and geographic information systems to obtain an emergency situation perception model. The selection module is used to select possible emergency response strategies based on the emergency situation perception model using a deep reinforcement learning algorithm, and to generate an emergency guidance plan by combining medical knowledge graph technology to analyze the relationship between the requester's symptoms and possible causes. The adjustment module is used to optimize the rescue route in real time based on the emergency rescue guidance plan by using a traffic flow prediction algorithm, and dynamically adjust the driving route of the rescue vehicle by combining road condition information collected by Internet of Things technology, so as to generate the optimal rescue route and the latest rescue progress notification. The processing module is used to visualize the emergency rescue operation process using the optimal rescue route and the latest rescue progress notification, and generate intuitive visual guidance.

8. A computing device, characterized in that, It includes a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are invoked and executed by the processing component to implement a human-computer interaction method in an emergency rescue scenario as described in any one of claims 1 to 6.

9. A computer storage medium, characterized in that, The device stores a computer program, which, when executed by a computer, implements a human-computer interaction method in an emergency rescue scenario as described in any one of claims 1 to 6.

Images