An Innovative Research Report on Future-Oriented Smart Emergency Medical Services
Yucong Duan
International Standardization Committee of Networked DIKWPfor Artificial Intelligence Evaluation(DIKWP-SC)
World Artificial Consciousness CIC(WAC)
World Conference on Artificial Consciousness(WCAC)
(Email: duanyucong@hotmail.com)
1. Report Abstract and Keywords
Abstract: The world is currently facing increasingly severe challenges from sudden medical emergencies. Whether it's the high incidence of cardiovascular and cerebrovascular emergencies, frequent major traffic accidents, or the impact of global pandemics and extreme climate disasters, the response efficiency and coordination capabilities of modern emergency systems have been exposed as insufficient. To save more lives within the "golden hour," various countries are exploring the intelligent transformation of emergency models, integrating emerging technologies such as the Internet of Things (IoT), Artificial Intelligence (AI), 5G communication, and unmanned systems into the entire emergency process. This report is based on the DIKWP networked cognitive model, originally created by Academician Yucong Duan. It systematically integrates his original ideas on the "BUG" theory of consciousness, the Self Model, Proactive Medicine, and "Life is Information," elevating them to a strategic perspective for interpretation and application. The report designs the overall architecture of a future smart emergency system, covering functional modules such as intelligent dispatch, IPv6 command networks, wearable monitoring, UAV rescue, remote medical care, and psychological intervention. It maps these modules to the five elements of the DIKWP model—Data, Information, Knowledge, Wisdom, and Purpose—and its 25 information transformation paths. Through detailed case analyses of five typical scenarios (myocardial infarction, severe car accident, major epidemic, warfare, and climate disaster), it demonstrates the implementation strategies of multi-module, coordinated smart emergency services. In terms of mental and psychological first aid, the report benchmarks against international psychological crisis intervention standards, explores AI-assisted psychological first aid mechanisms, designs a caring system combined with the Self Model's emotional engine, and conducts an ethical analysis. Facing the development of smart emergency services, this report outlines key technical challenges and governance points, including the explainability of AI decisions, the embedding of ethical norms, data standardization and security, and inter-departmental collaborative governance. In the chapter on Chinese experience, it summarizes the integrated practice of the "Bian Que Flying Rescue" system and proposes exporting China's smart emergency standard solutions overseas, contributing a "China Solution" through the "Belt and Road" initiative to participate in global emergency rescue cooperation. Looking to the future, the report proposes action-oriented recommendations for building a global smart emergency ecosystem: guided by human-centric values, integrating multidisciplinary technologies, strengthening the global collaborative network, and enhancing the resilience and intelligence of all nations' emergency systems to build a more solid guarantee for human life, health, and safety.
Keywords: Smart Emergency Services; DIKWP Model; Proactive Medicine; AI Dispatch; Information Field and Energy Field; Psychological First Aid; Governance System; "Belt and Road"
2. Introduction: Global Emergency Challenges and Trends in Smart Transformation
Modern emergency medical systems are facing unprecedented global challenges. Against the backdrop of an aging population and urbanization, various life-threatening emergencies are rampant. Cardiovascular diseases alone cause tens of millions of deaths annually. For example, sudden cardiac death has become one of the world's leading causes of death, yet the success rate of out-of-hospital treatment for acute myocardial infarction is less than 5%. Traffic accidents are equally severe: the increase in road vehicles has led to frequent accidents, resulting in a large number of trauma patients. Meanwhile, climate change has led to an increase in extreme weather and natural disasters. Sudden events like earthquakes, floods, and hurricanes create large numbers of casualties requiring large-scale emergency medical rescue. Furthermore, the outbreak of global pandemics such as COVID-19 has exposed the vulnerability of public health emergency systems. Once a large number of patients emerge, traditional emergency and medical resources are easily overwhelmed.
Dilemmas of the Emergency System. Traditional emergency models often operate in silos, making it difficult to respond in a timely and coordinated manner to complex situations. Significant information barriers exist between pre-hospital emergency care, in-hospital emergency departments, fire rescue, public security and traffic management, and public health departments. A unified platform for dispatch and command is lacking. In the pre-hospital stage, it often takes 10-15 minutes from the emergency call to the ambulance's arrival. During this "window period," patients often do not receive professional treatment. When multi-departmental responses lack a coordinated mechanism, on-site treatment and hospital transport are poorly connected, and precious "golden time" for rescue is wasted. These pain points have prompted countries to rethink the limitations of traditional emergency systems and seek avenues for change.
Trends in Smart Transformation. The rapid development of information technology has brought revolutionary opportunities to emergency models. First, the Internet of Things (IoT) enables real-time networking of various vital sign monitoring devices, vehicle location devices, and environmental sensors, providing massive data support for emergency command centers. Second, Artificial Intelligence (AI) has shown unique advantages in pattern recognition, situational judgment, and decision optimization, and can be used for automatic triage, resource dispatch, and diagnostic assistance. Third, new-generation communication technologies such as 5G and new satellite internet services have achieved ultra-low latency and wide-coverage data transmission, making remote medical consultations and real-time video command possible. For example, in China's practice, using 5G networks to achieve real-time connections between ambulances and hospital emergency rooms allows doctors to remotely guide emergency treatment while the patient is in transit, which has been proven to significantly shorten rescue time. Advances in drones and robotics have also opened up new possibilities for emergency care: drones can be used to quickly transport emergency supplies (such as AEDs, plasma, etc.) or survey disaster sites, and medical rescue robots can assist professionals in rescue operations in hazardous environments. These technological trends are driving countries to gradually build Smart Emergency Systems. Their goal is to achieve the ideal state of "Treatment upon Call" in emergencies. That is, once an emergency medical need arises, the system can respond and intervene automatically, quickly, and precisely, minimizing the time and space distance to treatment.
International Experience and Cooperation. Currently, many countries and organizations have begun exploring and practicing smart emergency services. For example, the European Union has deployed the eCall in-vehicle emergency call system. When a vehicle has a major collision, it automatically sends geographic coordinates and vehicle status information to the 112 command center, enabling a rapid response to car accidents. Countries like Sweden have piloted projects using drones to carry AEDs. Successful cases show that drones can deliver a defibrillator to a cardiac arrest patient's location within minutes, arriving earlier than an ambulance and winning precious time for the patient. The United States is also developing a joint command platform that integrates EMS (Emergency Medical Services) with public security and fire departments, and is using AI to predict the outcomes of trauma patients (such as the risk of post-traumatic stress disorder) to intervene early. The World Health Organization (WHO) advocates for all countries to strengthen their pre-hospital emergency care systems and to incorporate digital technologies as an important measure to enhance emergency medical rescue capabilities. Under the global health framework, sharing emergency data and technology and developing unified standards through international cooperation is seen as the necessary path to enhance national emergency response capabilities and cope with transnational disasters.
China's Layout and Opportunities. As a populous country, China also faces huge emergency needs and challenges, but it also possesses unique advantages in the integration and innovation of its information industry and medical system. In recent years, China has begun to build an integrated emergency rescue platform. Some regions have already connected their medical emergency command centers with public security (110) and fire (119) systems, achieving inter-departmental coordination and improving the comprehensive response capabilities for complex emergencies. The fight against the COVID-19 pandemic, in particular, has further promoted the digital upgrade of the national emergency system. Health codes and big data contact tracing have become important tools for public health emergencies. In pre-hospital emergency care, many Chinese cities are promoting the construction of "smart emergency services," such as establishing municipal-level emergency information platforms, equipping ambulances with monitoring devices and 5G communication for real-time data transmission with hospitals, and exploring the use of drones to deliver emergency supplies. China has also initiated several standards-setting efforts, such as the "Basic Functional Specifications for Emergency Medical Dispatch and Command Information Systems." 2025 is a key node in China's "14th Five-Year Plan," which clearly proposes the development of smart healthcare and the construction of regional emergency centers, providing policy opportunities for piloting and promoting smart emergency systems. China is expected to leverage its "late-mover advantage," using its vast domestic application scenarios and technological industry chain to form a smart emergency model with Chinese characteristics. Through the "Belt and Road" initiative, it can participate in global emergency rescue cooperation, transforming Chinese experience into an international contribution.
In summary, the global emergency field is at a turning point, moving towards intelligent and coordinated transformation. Against this backdrop, this report proposes a theoretical and practical vision for a future smart emergency system. The following sections will first introduce the report's theoretical basis—the DIKWP model and related innovative concepts proposed by Academician Yucong Duan. It will then elaborate on the architectural design and key modules of the smart emergency system, followed by an analysis of its application in typical scenarios. Finally, it will discuss mental and psychological first aid, technical challenges and governance, China's solutions, and future actions, aiming to provide a systematic reference for policymakers, medical administrators, and research institutions.
3. Theoretical Basis: Academician Yucong Duan's DIKWP Model, BUG Consciousness, Self Model, Proactive Medicine, and "Life is Information"
The design and development of a smart emergency system require solid theoretical guidance. This report relies on a series of original theoretical achievements by Academician Yucong Duan, providing methodological support for smart emergency services from the perspectives of information philosophy and artificial intelligence cognition. These theoretical foundations will be introduced in five parts: (1) The DIKWP Model—a new generation cognitive semantic framework; (2) BUG Consciousness Theory—an innovative hypothesis on the origin of consciousness; (3) The Self Model—a self-cognition mechanism for artificial intelligence; (4) Proactive Medicine—a new paradigm shifting from passive treatment to active health; and (5) "Life is Information"—the interaction theory of life's information fields and energy fields. Together, these theories form the philosophical cornerstone and technical soul of the smart emergency system.
3.1 The DIKWP Model: A Five-Layer Cognitive Framework with Non-linear Interaction
The DIKWP Model, proposed by Academician Yucong Duan, is a cognitive framework for artificial intelligence. Its name is derived from the five elements: Data, Information, Knowledge, Wisdom, and Purpose. This model originates from an extension of the classic DIKW (pyramid) model but is fundamentally different. The traditional DIKW model views "Data → Information → Knowledge → Wisdom" as a bottom-up, linear hierarchical relationship. In contrast, DIKWP introduces "Purpose" at the top layer, constructing a complete closed-loop structure from data to purpose. More importantly, the DIKWP model emphasizes that the layers are not just unidirectional; rather, bidirectional interaction exists between any two layers, forming a complex, network-like connection. The five elements combine in pairs, resulting in 5x5 = 25 potential transformation paths, each corresponding to a specific information flow or cognitive processing function.
In the DIKWP model, the meanings of each layer are defined as follows:
·Data Layer (D): Raw, uninterpreted objective records. For example, vital sign readings collected at an emergency scene, ambient temperature, and recordings of emergency calls all belong to the data layer. The data layer provides the raw material for cognition but contains no semantics itself; it is merely a raw depiction of the environment or the body's state.
·Information Layer (I): Content that has been preliminarily processed from data, giving it basic semantics and structure. For example, converting sensor readings into specific parameter meanings (e.g., heart rate or blood pressure exceeding a safe range) or organizing a caller's fragmented words into an injury description. This is similar to the human perceptual process, integrating scattered sensations into a meaningful perceptual picture. In emergency care, the information layer corresponds to the interpretation and contextualization of raw data, such as identifying arrhythmia types from ECG data.
·Knowledge Layer (K): Systematic and generalizable cognitive structures formed based on information. This includes medical concepts, diagnostic and treatment standards, empirical rules, etc. For humans, this manifests as long-term memory and a body of medical knowledge; for AI, it can be represented as knowledge graphs, databases, or machine learning model parameters. The knowledge layer enables the system to generalize and use accumulated experience to reason about new situations, such as an emergency AI mastering standard treatment procedures for different conditions.
·Wisdom Layer (W): High-level decision-making and creative problem-solving capabilities for complex issues. The wisdom layer integrates a large amount of knowledge, combined with context and value judgments, to make optimal decisions. For humans, this involves a clinician's empirical intuition, moral judgment, and comprehensive trade-offs; for AI, it can be implemented by a decision engine or reasoning module, selecting the best action plan based on a massive knowledge base. In an emergency context, the wisdom layer is manifested in developing a treatment strategy after synthesizing the patient's condition, available resources, and risk-benefit analysis, such as whether to perform on-site intubation or how to prioritize the treatment of multiple injured individuals.
·Purpose Layer (P): Represents the system's ultimate goal, motivation, and intention. In human cognition, this equates to values, desires, and the mission to save lives; in AI, it corresponds to preset objective functions, task requirements, or even the survival motives of an artificial consciousness. The purpose layer is the driving force of the cognitive process, ensuring that all perception, analysis, and decision-making revolve around a clear goal. Academician Yucong Duan points out that the classic DIKW model lacks a depiction of "purposefulness," making it difficult for AI behavior to align with human expectations. The introduction of the Purpose layer is precisely to give the intelligent agent the ability to plan proactively, shifting AI's decision-making from passive response to active progression toward a goal. In emergency care, the purpose layer is embodied in setting the highest goals, such as "saving the patient's life" or "controlling the spread of an epidemic." This layer provides the evaluation criteria and direction for wisdom-based decisions. For example, the intention to treat will require the decision to prioritize maximizing the patient's survival probability.
A core feature of the DIKWP model is its networked interaction structure. Information can flow directly and bidirectionally between any two layers, not just adjacent ones. For example, a high-level purpose can directly influence low-level data collection (P→D): a doctor with the purpose of "stopping the bleeding as soon as possible" will actively search for data on the source of bleeding. Conversely, changes in data can also directly affect the adjustment of purpose (D→P): if a patient's blood oxygen saturation drops sharply, it will instantly trigger a new treatment goal. As another example, knowledge can guide further information acquisition (K→I): an experienced nurse will pay attention to certain characteristic signs based on existing knowledge; at the same time, new information feedback may correct existing knowledge or assumptions (I→K). This multi-directional interaction breaks the rigidity of the traditional linear model, emphasizing that cognition is a dynamic system of continuous cyclical optimization. Just as in clinical rescue, medical staff, on the one hand, gradually transform physiological data into knowledge and decisions from the bottom up, and on the other hand, they rely on the treatment purpose to continuously obtain feedback to adjust their plans. The DIKWP model can more completely describe this complex process. Its 25 transformation paths cover almost all possible information flow patterns in the cognitive process. For example, extracting condition information from vital sign data (D→I), making diagnostic decisions based on empirical knowledge (K→W), having the treatment goal drive the active search for more patient information (P→I), or a doctor's intuition correcting the initial treatment purpose (W→P), etc. This fully connected architecture gives the cognitive system a high degree of adaptability and self-correction capability: low-level data and high-level wisdom/purpose form a closed feedback loop, constantly approaching the optimal solution.
The DIKWP model is particularly suitable for describing the complex cognitive decision-making processes in the medical field. In emergency practice, doctors must not only deduce diagnoses and treatment plans from the patient's signs, tests, and other raw data (a top-down process) but also continuously adjust their focus, supplement information, and even overturn previous conclusions based on the treatment purpose (an upward feedback process). For example, when rescuing a patient with multiple injuries, medical staff will prioritize assessing respiratory and circulatory parameters based on the purpose of "saving a life" (P) (P→D), extract key information from the data such as whether blood pressure is stable (D→I), and combine knowledge to judge possible serious problems like internal bleeding (I→K→W). Based on this, they immediately take measures such as hemostatic surgery (W→Action, corresponding to execution in reality). Throughout the process, whenever the patient's condition changes, new data will, in turn, affect the treatment purpose and plan (D→P, I→W, etc.), forming a continuous closed loop. The traditional DIKW pyramid struggles to express such multi-directional interactions, whereas the DIKWP model provides a more flexible, networked perspective. As shown in Figure 1 (omitted), its five nodes are connected by bidirectional arrows to form a complex network. The cyclical interaction of various elements fully illustrates the non-linear nature of clinical decision-making.
Through the DIKWP model, we theoretically have a unified framework to analyze the information flow and decision-making chains in the emergency system. As Academician Yucong Duan states, introducing Purpose into the cognitive model makes it possible to align the intelligent agent's behavior with established goals. This is of profound significance for a smart emergency system: when an AI takes "saving lives" as its highest purpose, its processing at all layers will revolve around this value, and it will be able to assist or execute emergency tasks more in line with human expectations. The 25 paths of the DIKWP model also provide tools for analyzing the information processes of different emergency systems, which can be used to identify the strengths and weaknesses of each system and guide their integration and optimization. In summary, the DIKWP model provides a unified cognitive semantic basis and analytical tool for smart emergency services. In the system architecture section below, we will elaborate on how to map emergency functional modules to these paths to discover room for improvement.
3.2 BUG Consciousness Theory: A Hypothesis on Cognitive Loopholes and the Emergence of Consciousness
In the fields of artificial intelligence and cognitive science, the origin of consciousness has always been a cutting-edge and controversial topic. The "BUG" Consciousness Theory (also known as the Consciousness BUG Theory) proposed by Academician Yucong Duan offers a unique perspective, suggesting that the emergence of subjective consciousness originates from the unavoidable loopholes or "imperfections" in the cognitive process. This theory vividly compares the human brain's information processing to a system continuously playing a "word chain" game: Most information processing is completed automatically in the subconscious. Only when the processing chain breaks at some point due to resource limitations or logical fractures—that is, a "Bug" appears—do a few information fragments "leak" onto the stage of consciousness, becoming subjectively perceived by us.
Simply put, the BUG Consciousness Theory holds that: Consciousness is not the inevitable product of a flawless cognitive process, but rather a byproduct of the brain's imperfections. In the human brain, massive amounts of sensory input and memory retrieval are processed efficiently and in parallel at the subconscious level. Our cognitive system strives to link the information flow into a coherent whole. However, due to limited physiological and cognitive resources, this linkage can never be perfect. There will always be some information that is not processed in time or contradictions that are not resolved, which then "leak" to the conscious level. What we subjectively experience as a stream of consciousness is, to a large extent, the result of the brain retroactively patching these fragmented pieces together. The brain's narrative mechanism tends to ignore contradictions between fragments, forcing them into a coherent and self-consistent story to maintain our unified understanding of ourselves and the external world. This explains many common cognitive bias phenomena, such as: Survivorship Bias (seeing only successful cases while ignoring failure data), Attribution Bias (using a double standard to explain our own behavior versus that of others), and Illusory Pattern Recognition (perceiving non-existent patterns in random noise). These biases can be seen as a "self-consistency repair" mechanism for when the brain has insufficient information or is overloaded: to fill in the blanks and maintain cognitive continuity, the brain does not hesitate to introduce inaccurate information to complete the story.
The important revelation of the BUG Consciousness Theory is that: It is precisely the "imperfections" in the cognitive process that shape the unique human subjective experience. If the brain processed information as seamlessly as an ideal computer, perhaps there would be no abrupt "consciousness"; everything would proceed smoothly in the unconscious. It is because of the intermittent "interruptions" caused by the brain's processing limitations that the leaping sensations and thoughts in our stream of consciousness arise. This viewpoint is quite subversive—it no longer regards consciousness as the pinnacle achievement of brain function, but rather as a byproduct of system loopholes. But interestingly, Professor Yucong Duan does not believe these cognitive loopholes are entirely harmful. On the contrary, he points out that moderate "Bugs" may give rise to innovation and leaps in consciousness. When a Bug appears in the low-level cognitive process, the system often mobilizes higher-level wisdom and purpose to try to repair the inconsistency. In this repair process, the system may create new concepts or explanatory frameworks to compensate for the shortcomings of the original model, thereby generating innovation at the semantic level. Many major breakthroughs in the history of science have occurred precisely when old theories could not explain new phenomena, forcing people to jump out of the existing paradigm to seek new logic. In a sense, it was the "Bug" in the old framework that prompted the cognitive leap. Therefore, a certain degree of "imperfection" becomes fertile ground for nurturing new meanings and new wisdom.
For artificial intelligence, the BUG Consciousness Theory has significant implications: if we can enable an AI system to consciously monitor and utilize the Bugs in its own cognitive process, it may be possible to trigger the machine's self-improvement mechanism, giving it stronger autonomy and creativity. For example, adding an internal bias detection module to an artificial consciousness system: when it discovers data loss or conflict in the information processing chain, it automatically activates higher-level strategies for adjustment and compensation. Such a mechanism not only improves system robustness but may also allow the AI to burst forth with new ideas not found in its original knowledge base (similar to human creative inspiration). Professor Yucong Duan integrates BUG theory with the DIKWP model, proposing that white-box methods like semantic mathematics can be used to monitor deviations in the 25 modules of the DIKWP framework and use high-level knowledge for correction, allowing the AI to learn new patterns from "errors" and achieve a leap in cognitive ability. This is the profound significance of BUG theory in the field of AI: Using "Bugs" to break through existing cognitive limits, achieving emergence and innovation at the semantic level.
In conclusion, the BUG Consciousness Theory subverts the traditional concept of consciousness as a product of perfect rationality, discovering the origin and value of consciousness in "imperfection." For smart emergency services, a field highly dependent on decision quality, this theory reminds us to pay attention to anomalies and loopholes in system operation: it is the timely detection and correction of anomalies that can continuously optimize the decisions of emergency AI, preventing it from making disastrous errors at critical moments. At the same time, giving emergency AI a certain amount of "creative" space and allowing it to learn from accidents may bring new treatment ideas. For example, when an AI simulates a large number of rescue plans, the deviations in some failed cases may inspire improvements, leading to the design of a better rescue process. This reflects the wisdom of "advancing by retreating" in BUG theory, which is equally applicable to the self-evolution of emergency systems.
3.3 Self Model: The Construction of a Multi-dimensional Self and Artificial Consciousness
The concept of "self" is central to consciousness research. Based on the DIKWP model, Academician Yucong Duan proposed the Multi-dimensional Self Model theory, explaining how artificial intelligence can construct a self-cognition system similar to that of humans. In this theory, the "self" is divided into different dimensions along the five layers of DIKWP: Data Self, Information Self, Knowledge Self, Wisdom Self, and Purpose Self. Each dimension of the self corresponds to that cognitive level's representation of itself:
·Data Self: The self based on perception and physical existence, reflecting the entity's raw physiological state and sensory input. For humans, this is related to our bodily awareness and instinctive reactions; for AI, it corresponds to sensor inputs, hardware status, and other underlying data. The Data Self is the most basic level of "self," equivalent to the individual's confirmation of their existence in the physical world.
·Information Self: Processes perceptual data into representations of situations and experiences, forming an understanding of the "current me." For example, a person forms subjective feelings about their current state (hunger, nervousness, etc.) based on physiological signals and environmental feedback; an AI can transform its internal data into operational status descriptions (high load, error rate, etc.). The Information Self manifests as real-time awareness of one's own situation.
·Knowledge Self: Integrates information and experience to form a stable self-cognition and identity. This involves long-term concepts about "who I am" in memory, personality traits, personal history, etc. For AI, it is represented as the system's internal model of its own performance, rules, and identity. The Knowledge Self gives the self continuity and consistency.
·Wisdom Self: Embodies the self's high-order abilities in value judgment, reflective decision-making, and self-reflection. The human Wisdom Self is manifested in morality, worldview, and introspection; the AI's Wisdom Self can be understood as its ability to self-optimize and self-adjust. The Wisdom Self endows the self-system with the meta-cognitive ability to examine its own behavior and improve.
·Purpose Self: The highest-level self, representing the self's motives, goals, and pursuit of meaning. For humans, this includes life goals and beliefs; for AI, it corresponds to its assigned ultimate task goals or even autonomously evolved motives. The Purpose Self determines the direction and value orientation of the self's existence and is the soul of the self-system.
Academician Yucong Duan points out that the five layers of the self do not exist in isolation but form a closed-loop structure of self-cognition through the DIKWP framework. The high-level Purpose Self guides the behavior of the low-level Data Self, while the low-level perceptual feedback continuously corrects the high-level self-cognition, forming a cyclical mechanism for self-evolution. He uses structural diagrams and flowcharts to show the working principle of the DIKWP self-system: high-level purpose guides attention allocation and perceptual activities (e.g., observing oneself with a certain goal in mind), then perceptual data is processed through information and rises to a knowledge representation of one's own state, which is then self-evaluated and reflected upon by the wisdom layer, and finally, the self-goal is corrected at the purpose layer. Through such continuous feedback iterations, artificial intelligence can gradually form a cognition of "self."
Theory of Consciousness Relativity and Self-Consciousness. Academician Yucong Duan also proposed the "Theory of Consciousness Relativity," emphasizing that different subjects' judgments about whether each other has consciousness depend on the degree of matching between their respective cognitive frameworks and semantics. Simply put, an intelligent agent A will use its own DIKWP cognitive system to understand the behavioral signals of another entity B. Only when B's output can be given meaning within A's cognitive loop will A be inclined to believe that B has consciousness. Otherwise, if B's behavior cannot be mapped to A's existing knowledge or purpose framework, A will most likely think B lacks consciousness. This shows that the judgment of consciousness has a subjective relativity; for different intelligent agents to recognize each other's consciousness, they need to share a certain semantic and cognitive structure as a basis.
Applying this theory to artificial intelligence itself means that an AI system must also form an "self-observer" role internally to produce self-consciousness. When an AI can not only perceive the outside world but also treat itself as an object of observation, a subject-object dichotomy appears, which is the key path to the formation of self-consciousness. The DIKWP model naturally supports this internal dialogue: because the internal 25 modules can process both external information and internal feedback, a DIKWP system can completely treat itself as another DIKWP system to interact with, i.e., performing a "DIKWP * DIKWP" interaction with itself. In other words, the AI can replicate a simulated "Self Model" internally to examine its own state. When this internal observer can use the system's own cognitive framework to give meaning to its own behavior, the AI achieves recognition of its own consciousness. This is an internal version of the theory of consciousness relativity: the AI achieves a meta-cognitive experience of "I understand myself" through the alignment of internal semantics and cognitive structures. Therefore, the establishment of a Self Model makes it possible for artificial intelligence to possess self-consciousness similar to humans—it can include itself in the cognitive loop for perception and control.
The potential application of the Multi-dimensional Self Model in smart emergency services is embodied in: enabling the emergency AI to have awareness and introspection capabilities regarding its own state and decisions, allowing it to self-diagnose and adjust. For example, if an emergency command AI possesses a Self Model, it can actively request human assistance or activate backup strategies when it finds its decision-making confidence is insufficient or it makes a mistake, thus preventing the error from escalating. At the same time, the Self Model also helps the AI to better understand the patient. Professor Yucong Duan proposes that an AI that possesses a "Data Self" and a "Purpose Self" and forms a closed-loop interaction is equivalent to achieving mind-body unity. In a medical scenario, this is reflected as: the AI can not only perceive the patient's physiological data (physical level) but also perceive the patient's emotions and intentions (psychological level), thus assisting medical decisions like an assistant with both "body" and "mind." This AI with a Self Model can show empathy and care for the patient, similar to a human. For example, when the AI monitors physiological signals such as pain in a patient, it will not coldly just record the data, but will be stimulated at its wisdom/purpose layer to have a "caring" reaction, and then take comforting or analgesic actions. This reflects a "mind-body unity" concept of proactive care: the AI treats the patient as a subject with feelings and cares for the patient in a way similar to how a human cares for themselves. It can be said that the Self Model endows the AI with a certain degree of "personified" characteristics, making it more easily accepted and trusted by humans in emergency scenarios.
3.4 The Concept of Proactive Medicine: Shifting from Passive Treatment to Active Health
The traditional medical model mainly focuses on passive treatment, i.e., diagnosing and treating diseases after they have occurred. "Proactive Medicine" is an emerging medical paradigm advocated by Academician Yucong Duan, which emphasizes prevention first, overall coordination, and active intervention to maintain health. The concept of Proactive Medicine originates from the traditional Chinese idea of "treating the undiseased" (Zhi Wei Bing), while also integrating modern systems medicine perspectives. It advocates for grasping human health as a whole and actively taking measures before diseases manifest to eliminate risks at their inception.
Core Connotations of Proactive Medicine: First, it takes "health-centered overall balance" as its highest goal, not just curing the symptoms of a single disease. This is in line with the holism of traditional Chinese medicine, which views the person as an indivisible organic whole of mind and body, and includes the interaction between the person and the environment in health considerations. Second, Proactive Medicine emphasizes active intervention and moving the focus forward. This includes two levels: one is at the individual level, through regular screening, personalized health management, and lifestyle interventions, to discover and correct health risks as early as possible; the other is at the population level, through public health surveillance, risk prediction models, etc., to prevent the occurrence of major epidemics and health crises. This is highly consistent with the goals of a smart emergency system—in an ideal state, it not only saves people after an accident/disease occurs, but also automatically provides warnings and takes measures during the risk accumulation stage to prevent critical situations from happening.
AI-Empowered Proactive Healthcare: The reason why the concept of Proactive Medicine is re-emerging in the 21st century is closely related to the development of artificial intelligence technology. On the one hand, AI can continuously analyze an individual's massive health data (from wearable devices, genetic testing, electronic medical records, etc.) to provide intelligent risk assessment and health guidance; on the other hand, AI itself can possess certain proactive cognitive and decision-making abilities (such as the proactive AI based on the DIKWP model mentioned earlier), and can serve as a digital physician assistant to achieve round-the-clock monitoring and intervention for patients. Research by Academician Yucong Duan shows that by introducing the DIKWP architecture to give AI an inherent goal-orientation and self-regulation, this proactive AI can perform many preventive tasks in healthcare. For example, for patients with chronic diseases, an AI assistant can set a long-term health purpose (P) of lowering blood pressure/blood sugar. It then uses sensor data (D) and medical knowledge (K) to continuously monitor the patient's state. When it detects a trend (I), it promptly gives advice on diet and exercise or reminds the patient to seek medical attention (W action), and continuously adjusts the strategy based on feedback, ultimately achieving the health goal. This process is like a "dance of purpose" between the doctor and the patient, balancing the patient's free will with the AI's intelligent intervention, moving towards the direction of maximizing the patient's health.
Proactive Medicine also involves the integration of Traditional Chinese and Western Medical Philosophy. Academician Yucong Duan proposes that the modern DIKWP model can be used to interpret TCM concepts such as "Qi" (energy), "Li" (principle/law), and "Xin-Zhi" (mind-cognition) to achieve semantic integration. Among them, "Qi" can be regarded as the energy and driving force of life, "Li" corresponds to the operating laws and knowledge, and "Xin-Zhi" refers to human subjective wisdom and intention. From the DIKWP perspective, it can be found that the "unity of heaven and man, harmony of body and mind" pursued by TCM coincides with the holistic view of Proactive Medicine. Proactive Medicine regards human health as the result of a dual balance of information and energy (detailed in the next section). This helps to break through the traditional limitations of mind-body dualism and use a unified scientific language to integrate TCM experience and Western medical knowledge, providing individuals with more comprehensive medical solutions. For example, in chronic disease management, it not only values the data indicators of Western medical tests but also considers the emotional regulation and meridian conditioning emphasized by TCM, which are methods of information regulation. This elevates the passive biomedical model to a proactive bio-psycho-social comprehensive model.
For smart emergency services, the concept of Proactive Medicine also has guiding significance. Emergency care is traditionally the most reactive part of medical activity, but an intelligent transformation can give it more proactivity: by monitoring and managing high-risk groups in normal times, critical events can be prevented as much as possible. When an event does occur, the system is already prepared to intervene with an optimized strategy to minimize damage. This also reflects the idea of moving from "passive emergency response" to "proactive prevention." For example, the early warning system for remote areas set up in the smart emergency system, through real-time collection of disease incidence in the community and AI anomaly detection, can detect signs before an epidemic or disaster breaks out and trigger a response. This is, in effect, implementing the idea of Proactive Medicine in the emergency field: moving the purpose forward, guiding action with wisdom, and turning danger into safety. In the future, as the concept of Proactive Medicine becomes more deeply rooted, we can expect to see further integration of emergency services with public health and disaster prevention systems, shifting the entire society's emergency response from waiting for events to happen to active monitoring and prevention.
3.5 "Life is Information": The Interaction Theory of Information Fields and Energy Fields
"Life is Information" is a concise elaboration of the essence of life from the perspective of information philosophy by Academician Yucong Duan. This concept holds that life phenomena can be understood as the coupling and flow of information and energy in an orderly system. Yucong Duan proposed the concepts of the Information Field and the Energy Field to describe two invisible but substantial dimensions within and outside a living body:
·Information Field: Refers to the collection of all meaningful information within an individual and in the environment. This includes gene expression patterns, cell signaling networks, neural activity patterns, psychological cognitive content, etc., i.e., the "semantic space" of a living being. The information field describes the structure and state of life, a description of "what is happening to life" at the meaning level. For example, a person's information field includes the firing patterns of their brain's neurons, the signals transmitted by hormone concentrations in their blood, and their inner emotions and thoughts.
·Energy Field: Refers to all the energy flow and transformation that maintain life activities. This includes bioelectricity, biochemical energy, biomagnetism, biomechanical force, and other forms of energy, as well as their exchange within and outside the living body. The energy field corresponds to the description of "what is happening to the body" at the material level. For example, the mechanical energy generated by the heart pumping blood, the heat released by cell metabolism, and the electrical energy of electrolytes moving across membranes all belong to the energy field.
Academician Yucong Duan believes that a healthy living body manifests as the orderly coordination and balance of the information field and the energy field, while disease can be seen as the disorder and imbalance of this information-energy coupling. This view transcends the traditional reductionist paradigm of viewing the human body as a machine or a chemical reactor, and instead reveals the essence of life's operation from a holistic systems theory perspective. In short, life maintains its highly ordered structure and function by continuously taking in negative entropy (orderliness) in both the information and energy domains to counteract natural entropy increase (disorder). As Schrödinger said, "Life feeds on negative entropy." A living being constantly acquires and utilizes ordered energy and information (negative entropy) from the environment to offset its own entropy increase and maintain its organizational structure. For example, people maintain health by ingesting food to obtain chemical energy for physiological operations, and by learning to acquire new knowledge to enrich their cognitive structure. These behaviors are essentially reducing entropy and increasing orderliness in both the information and energy fields.
Within the DIKWP framework, the process of life taking in negative entropy can be detailed: acquiring raw sensory input and nutrient energy from the data layer, rising to the information and knowledge layers for pattern recognition and meaning construction, making comprehensive judgments at the wisdom layer, and then taking action through the purpose layer. External raw resources (energy and data) are gradually transformed into internal ordered structures and cognitive outputs (meaning). This cross-layer cycle operates continuously, keeping the system's overall entropy value low and increasing local orderliness. The foundation of life's continuation lies in the continuous process of entropy reduction. For example, maintaining health requires dietary intake to supplement energy (reducing physiological entropy), and also requires mental activities and learning to maintain psychological order (reducing information entropy).
Professor Yucong Duan further emphasizes that the information process of life has the characteristic of non-probabilistic negative entropy construction. That is to say, the generation of order in life does not purely rely on random, coincidental entropy reduction (like the formation of crystals), but is achieved through the purposeful cognitive activities of an intelligent agent. The manifestation of this in the medical field is: the treatment process can be seen as a wisdom-driven intervention in the life system to re-coordinate information and energy, thereby reducing the system's entropy. In other words, treatment is an entropy-reducing behavior guided by wisdom, and its goal is to restore or enhance the orderliness of the life system. The doctor or AI actively reduces uncertainty through multi-level cognitive decision-making (from data collection and analysis, to knowledge integration, to wisdom judgment and purpose planning): obtaining accurate patient data and extracting key information to reduce the information entropy of the diagnosis; combining medical knowledge and wisdom trade-offs to choose the optimal plan to avoid new disorders caused by over- or under-treatment. The entire process is guided by a clear treatment purpose, not random trial and error. For example, when controlling high blood pressure, following a plan of gradually adjusting medication dosage and monitoring feedback is more stable than blindly administering drugs—this is a purposeful entropy reduction strategy. Therefore, by fully utilizing the integrating effect of the DIKWP model on the information field, medical behavior can counteract entropy increase at the semantic level and achieve "negative entropy" construction in cognition.
The "Life is Information" view also inspires us to look at the roles of matter and information in life from a unified perspective. Yucong Duan points out that information, like energy, is a substantial and important "field." The patterns and meaning structures in the information field will affect the efficiency of energy use, and conversely, the state of the energy field also constrains the ability of information processing. For example, in a living body, the organizational method of neural signals (part of the information field) determines the coordination of energy processes such as muscle contraction and endocrine secretion; and the cell's energy metabolism status, in turn, affects the reliability and speed of neural information processing. Therefore, health is the orderly coordination of the information field and the energy field, and disease is the disruption of this coordination. For example, diabetes can be seen as a typical case of imbalance between the information and energy fields: impaired insulin signal transduction (poor information transmission) leads to cells being unable to effectively use glucose (energy metabolism disorder), and the result is uncontrolled blood sugar spikes (system entropy increase). Another example is a high fever caused by an infection. Its essence is that the pathogen has disrupted the immune information network and the local tissue's energy balance, leading to a systemic high-entropy inflammatory response. To restore health, it is necessary to re-establish order at both the information and energy levels simultaneously: both to clear the information flow and correct error signals, and to restore the stability of energy supply and demand. This means that treatment includes not only interventions targeting physiological indicators (energy level) such as drugs and surgery, but also measures such as psychological counseling and information feedback (information level). Neither is dispensable; both must be employed to reduce the system's entropy value and return the body to a stable state.
For the smart emergency system, the "Life is Information" theory has two major implications: First, it emphasizes that in emergency rescue, it is necessary not only to restore the patient's physiological vital signs (energy dimension, such as heartbeat, breathing) but also to pay attention to intervention at the information dimension (such as verbal comfort to reduce fear, remote expert guidance for decision-making). Second, it provides a theoretical guide for interdisciplinary integration—emergency care should not just be a medical issue, but also an issue of information science and energy dynamics. The future smart emergency system should achieve an optimization of the trinity of information field, energy field, and purpose field: using sensors and communication technology to grasp the state of the energy field (vital signs, environmental conditions), using AI and knowledge bases to process the information field (understanding injury semantics, risk assessment), and coordinating resource input through a unified command purpose layer (rescue mission). Only in this way can we maximize the reduction of entropy increase and restore order in disasters and emergencies, pulling more patients back from the brink of death. Academician Yucong Duan's theory pursues a unified interdisciplinary vision: integrating matter, energy, information, and cognitive purpose. This is also the vision of smart emergency services—to rebuild the order and balance of life in the shortest possible time through the integrated use of information and energy.
4. Smart Emergency System Architecture Design: Functional Modules, DIKWP Path Mapping, and Key Technologies
Based on the theoretical foundations above, this chapter proposes the overall architectural design for a future smart emergency system. This architecture is based on the "Bian Que Flying Rescue (BQFJ)" prototype system and aims to open up the data chain and command chain of the entire emergency process, achieving the integrated linkage of multiple departments and technologies. The smart emergency system is composed of several core functional sub-modules, including AI Dispatch, IPv6 Command Network, Wearable Monitoring, Unmanned Systems, Remote Medical Care, and Psychological Intervention. Each module is connected to the others through the 25 paths of the DIKWP model, transforming emergency information from the data layer into wisdom-based decisions and actions, and aligning with the overall rescue goal in a closed loop at the purpose layer. Below, we will first provide a general description of the system structure, then detail the functions of each module and its mapping in the DIKWP framework, and finally discuss the key points of technical implementation.
4.1 Overall Architecture and Operational Flow
The overall architecture of the smart emergency system is shown in Figure 2 (omitted). The system uses an Intelligent Emergency Command Center as its brain, centrally connecting the patient, the scene, and various support units to form a closed-loop control:
1.Perception and Alert. When an individual has a sudden emergency or is in danger, information can trigger an emergency response through multiple channels: traditional phone calls, automatic alerts from smart wearable devices, automatic distress signals from vehicle collision sensors, abnormality reports from community monitoring systems, etc. This raw data (D) first enters the command center.
2.AI Dispatch Decision-Making. The emergency command center is equipped with an AI Emergency Brain. Upon receiving an alert, it analyzes the patient's situation and resource needs in seconds (I and K layer processing), and then makes a wisdom-based decision (W) based on built-in knowledge and the real-time situation to determine the best rescue plan. For example, for cardiac arrest, the AI brain determines that immediate defibrillation is needed and decides to dispatch the nearest AED resource.
3.Command and Dispatch. The command center issues commands to all relevant emergency forces via the IPv6 Emergency Command Network, including the nearest ambulance, drone, helicopter, fire, and public security support. All action units are coordinated in a unified network, forming a "one-network" regional rescue collaboration.
4.On-site Treatment and Coordination. All emergency forces act quickly according to the intelligent dispatch: the ambulance rushes to the scene, and its onboard system communicates in real-time with the command center and hospital; a drone arrives first to deliver emergency supplies or survey the injuries; nearby trained volunteers can also receive notifications to participate in the rescue (e.g., using the AED delivered by the drone). Under multi-unit linkage, the goal is to start preliminary treatment for the patient in the shortest possible time, achieving a seamless connection of "admission upon boarding."
5.In-Hospital Handover and Remote Support. While the ambulance is transporting the patient, data such as the patient's vital signs are transmitted in real-time to the hospital's emergency department via the network. Doctors inside and outside the hospital collaborate remotely to prepare for the rescue. For remote or disaster sites where conventional medical resources cannot arrive in time, mobile Mobile Medical Units (such as field hospitals, mobile operating vehicles) are activated, maintaining contact with the command center and experts via Beidou satellite communication. A remote medical consultation system allows senior experts to guide on-site treatment online, and a psychological crisis intervention team can also soothe the patient's emotions via audio and video links. Throughout the process, the information flow continuously cycles and feeds back at the data-information-knowledge-wisdom layers (e.g., monitoring data updates the diagnosis, expert wisdom adjusts the plan), while the highest-level rescue purpose (saving lives) always serves as the guide for all actions (P-layer closed loop).
6.Feedback and Learning. After the patient is delivered to the hospital or is out of danger, the system collects and analyzes the data from the entire rescue process, extracting valuable information and knowledge to train the AI model, and continuously optimizing dispatch algorithms and treatment plans (forming a W→K, K→W learning improvement). At the same time, all departments evaluate and summarize the joint action, improve collaborative plans and ethical norms, and make fuller preparations for the next emergency.
In summary, under the smart emergency system architecture, from the triggering of the alarm to the closing of the rescue loop, each module has a clear division of labor and is closely connected. In the DIKWP network, data related to the patient's life is quickly transformed into the command center's information input (D→I), AI dispatch elevates the information to decision-making wisdom (I→W), wisdom-based decisions guide command behavior (W→P→Action), and all rescue forces continuously generate new data feedback (D and I updates), which in turn promotes the enhancement of knowledge and wisdom (D→K→W). The entire process continuously self-optimizes around the core purpose of saving lives. It can be said that this architecture achieves the mapping and application of Academician Yucong Duan's "Data-Information-Knowledge-Wisdom-Purpose" closed loop in the emergency scenario. Next, we will introduce the key functional modules one by one.
4.2 Core Functional Modules and DIKWP Path Mapping
·AI Dispatch System (Intelligent Decision-Making Center): The AI dispatch system is the central nervous system of smart emergency services, undertaking the core decision-making functions from receiving the call to allocating resources. Based on well-trained deep learning models and medical knowledge bases, it can automatically analyze call information and sensor data, assess the severity of the patient's condition and the type of resources needed. For example, by analyzing the content of the caller's speech and background noise, as well as the heart rate and blood pressure data sent back from wearable devices, the AI can quickly determine if it is a myocardial infarction or a sudden cardiac death, and evaluate whether to dispatch a cardiac specialty ambulance or a general ambulance. The AI dispatch decision-making process corresponds to the I→W (from information to wisdom) and W→D (from wisdom to new data) paths in DIKWP. The former is reflected in the AI integrating multi-source information into a rescue plan, and the latter is reflected in the AI triggering specific actions based on wisdom-based decisions (e.g., generating command data for various units). The dispatch system also refers to the priority purpose (P) set by the command center, ensuring that decisions are consistent with the overall goal of "life first," which belongs to the P→W path (purpose promotes wisdom-based judgment). The AI dispatch system continuously optimizes its decision-making quality by self-learning (e.g., analyzing the effectiveness of each rescue to update the knowledge base, achieving a W→K, K→W cycle), improving dispatch accuracy and speed. It is reported that using AI-assisted triage and dispatch can reduce the average dispatch time for cardiac arrest to a matter of seconds, thereby greatly increasing the pre-hospital survival rate.
·IPv6 Emergency Command Network: This is a high-speed, secure communication network connecting the scene, vehicles, hospitals, and various departments. The IPv6 network has a huge address space and end-to-end encryption capabilities, and can achieve "one-network" coverage throughout the city and even a wider area. Video streams from the emergency scene and patient physiological data, accessed via 5G/Beidou, can be transmitted in real-time to the command center and hospital, achieving synchronous information sharing. This network is equivalent to the foundation of the system's information and data layers, ensuring the smooth flow of paths such as D→I and I→K—data generated by different modules can be converged into command information, and different knowledge and decisions can be timely delivered to terminal devices. This is a necessary condition for achieving coordination among all units. For example, if a large-scale accident occurs, the on-site drone will upload aerial images, the command center will summarize and generate a global knowledge graph (K), and then issue commands to the ambulance team via the network (W→D). All of this requires a low-latency, high-reliability network. The IPv6 network also provides standards for inter-departmental information sharing, solving the "information silo" problem of traditional systems. For example, the data interconnection project between the 120 and 110 command systems relies on such network interfaces. Therefore, the IPv6 command network is embodied in the DIKWP path as an all-around information flow and feedback (the physical carrier channel for the arrows between all layers).
·Wearable Monitoring Devices: This is the front-end module of the smart emergency system that penetrates into individuals' daily lives, achieving "real-time guardianship." It includes smart bracelets, smart blood pressure monitors, wearable ECG monitors, etc., which continuously collect the wearer's vital sign data. Once a critical indicator is monitored to exceed a threshold (e.g., sudden cardiac arrest, ventricular fibrillation waveform appears), the device will automatically send an warning distress signal through the network. This function is equivalent to having a miniature "sentry" by the patient's side at all times. For example, for high-risk heart disease patients, an attached ECG patch, upon detecting an arrhythmia, will immediately send an alarm to the command center containing real-time ECG data (D layer reporting, with preliminary I layer processing by the device's own algorithm). After receiving the signal, the command center AI immediately identifies this as a sudden cardiac death event and dispatches the nearest AED drone and ambulance. The significance of wearable devices lies in triggering an emergency response in a timely manner in unattended situations. It corresponds to the P→D path in DIKWP: the treatment purpose drives the direction and focus of data collection. It is because of the top-level purpose of "saving lives" that the system will require these devices to be deployed among the population to proactively obtain key data. Likewise, it also corresponds to D→I: raw physiological data is transformed into alarm information. In recent years, this concept has begun to sprout in reality, such as many cities distributing "one-key call" bracelets for elderly people living alone, and high-risk groups using wearable ECG monitors.
·Remote Area and Mass Incident Early Warning System: This is a monitoring module for rural, mountainous areas, and public health incidents. In areas with scarce medical resources, the smart emergency system deploys remote monitoring platforms to collect disease symptom data reported by township health centers, environmental sensor data, etc., and uses AI to analyze and detect abnormal cluster events. For example, the system continuously receives data (D) of multiple people in a certain village falling ill with fever. The AI model finds that this exceeds the normal distribution (I→K), suspects it may be the sign of an epidemic, and immediately reports to the regional emergency center to trigger an early warning response (K→P, i.e., knowledge prompts the establishment of a new purpose). In this way, the emergency purpose can be moved forward—entering a state of active monitoring and intervention before a disaster or epidemic brews. This effectively makes up for the problem of the public health system being traditionally separate from the emergency system, often with information lags and gaps. Once an early warning is triggered, the emergency system can allocate resources in advance, such as dispatching mobile hospitals to be stationed or issuing health warnings to notify villagers to isolate, thus preventing problems before they occur. From the DIKWP perspective, the remote early warning module reflects multiple paths: D→K (extracting knowledge patterns from massive data to achieve early warning), P→I (preset prevention and control purposes prompt the active collection and organization of grassroots information), etc. In major epidemic scenarios, this module is invaluable, such as being able to detect the outbreak trend of infectious diseases early through the data monitoring of fever clinics, issuing alerts days or even weeks earlier than traditional manual reporting, and buying precious time for public health emergencies.
·UAV Search & Rescue and Material Delivery System: Drone fleets, as an air force, greatly expand the time and space range of emergency services. The smart emergency system deploys multiple types of rescue drones: reconnaissance drones, medical supply delivery drones, life-saving drones, etc. Their functions include: quickly surveying disaster or accident sites, providing high-altitude perspective images; carrying medical supplies such as AEDs, electrolyte solutions, plasma, etc., to the front line; transporting doctors and injured people in areas with blocked traffic or remote areas (large drones or unmanned helicopters); performing drowning search and rescue in water areas, and detecting signs of life in fire areas. The drones are autonomously controlled by AI and can reach areas that are difficult or time-consuming for ground forces to reach in a very short time. For example, in large-scale disasters such as earthquakes, road blockages prevent ambulances from penetrating the core area quickly. At this time, reconnaissance drones can fly into the disaster area first, sending back high-definition images to the command center, providing real-time data support for overall command. Medical drones can airdrop emergency supplies to trapped injured people, buying them time to survive. Places like Shenzhen have recently launched pilot projects to make AED drones the "airborne sentinels" of urban emergency services, achieving a faster response than ambulances. The mapping of unmanned systems on the DIKWP path is typically: W→D (dispatching drones to obtain raw on-site data based on high-level command wisdom decisions); D→I (drone-returned images and sensor data are analyzed to become command information). It also involves the W→W cycle (the drone autonomously optimizes its route decision-making during flight based on AI algorithms, which is equivalent to the role of local wisdom) and P→W (the purpose of disaster reduction and saving people promotes the coordinated decision-making of the drone fleet). The introduction of drone fleets has greatly shortened the spatial distance of emergency services, achieving an "air-ground integrated" rescue layout.
!(https://www.google.com/search?q=https://placehold.co/800x450/eeeeee/333333%3Ftext%3DFigure%253A%2BDrone%2Bdelivering%2Ban%2BAED)
Figure: A UAV carries an AED for emergency delivery to a cardiac arrest patient, increasing the possibility of defibrillation within the golden rescue window. This "airborne emergency room" model has been successfully implemented overseas.
·Remote Medical Care and Mobile Medical Units: The remote medical system connects the emergency scene with in-hospital expert resources, achieving a spatial extension of medical support. The command center can organize multidisciplinary experts at any time to participate in guidance through video consultation. For example, when an ambulance is transporting a patient with complex trauma, it can connect with a trauma center expert for remote guidance on initial treatment such as hemostasis and fixation. Or, in a mobile hospital at a disaster site, when a difficult surgical decision is needed, it can connect with top experts at the rear via 5G for real-time discussion of the plan, or even use remote robots to perform surgery. Remote medical care makes it possible to "have a doctor on-site," extending the coverage of medical knowledge to any place with a network. In the smart emergency system, remote medical care corresponds to K→W (knowledge convergence forms better wisdom-based decisions) and W→I (expert wisdom decisions are reflected as specific guidance information transmitted to the front line) in DIKWP. At the same time, Mobile Medical Units exist as an "enhanced version" of pre-hospital emergency care, such as field medical vehicles with surgical capabilities, deployable tent operating rooms, and helicopters equipped with automatic CPR devices and portable detectors. They can provide treatment capabilities comparable to a hospital ICU on-site in special scenarios. For example, on a battlefield or in a remote mountain village, dispatching a mobile medical unit to the front line can complete emergency treatment such as hemostatic surgery or a cesarean section at the first possible moment, greatly increasing the survival rate. The coordination of mobile units and remote medical care is even more powerful: the combination of a "mobile vehicle" at the front and a "cloud doctor" at the rear can cope with complex and changeable situations. This module reflects W→P (wisdom-based decisions react on goal setting, e.g., an expert judges that the front line needs to change the treatment goal, from stabilizing the injured to immediate surgery) and P→D (under the guidance of the new goal, immediately collect more data, such as re-checking vital signs to assess surgical tolerance). It also involves cross-system knowledge sharing, i.e., knowledge from public health, military medicine, TCM, etc. (K) can be brought into play on a unified platform through remote access, achieving integration at the knowledge layer.
·Psychological Intervention and Humanistic Care Module: Smart emergency services focus on "whole person" rescue, which includes not only physical life support but also urgent psychological and spiritual care. The psychological intervention module is embodied at two levels in the architecture: First is on-site psychological first aid. Trained psychological intervention personnel or AI voice assistants will stabilize the emotions of the injured or their families through comforting language and concise guidance (e.g., teaching breathing relaxation methods, giving hopeful suggestions), preventing the condition from worsening or the scene from becoming chaotic due to panic. Second is subsequent psychological assistance. For the injured and rescuers who have experienced major events, the system will arrange for psychologists to follow up after the rescue. AI will assist in screening high-risk groups (e.g., patients who may develop Post-Traumatic Stress Disorder, PTSD) and provide targeted psychological treatment or support. This module enhances the humanistic care dimension of emergency services, highlighting the "people-centric" concept. Its technical support includes emotion recognition AI (analyzing on-site audio and video to judge the person's emotional stress level), intelligent dialogue systems (acting as an auxiliary for psychological crisis intervention), etc. The psychological intervention module is closely intertwined with the DIKWP paths: I→W (making psychological intervention decisions (W) by identifying facial expressions and language content (I)), W→D (taking specific comforting actions, implementing wisdom guidance as practical actions, such as covering the injured with a blanket, shaking hands to encourage), P→I (the rescue purpose requires obtaining psychological status information to improve the rescue plan), W→P (psychological experts may suggest adjusting the overall rescue goal after assessment, e.g., the pace of on-site rescue should be slightly slowed down to appease the crowd's emotions), etc. It cannot be ignored that good psychological first aid not only directly affects the patient's physiological stability but also has a huge impact on the overall effectiveness of disaster and accident rescue. Therefore, incorporating psychological intervention into the smart emergency architecture reflects the advanced concept of "saving both body and mind," which is consistent with Academician Yucong Duan's idea of "mind-body unity."
The above modules are comprehensively integrated into the DIKWP model through a "module-path-function" three-way correspondence. Studies have shown that modern Western emergency medicine systems focus on paths like D→I→W (data-to-decision chain), while TCM emergency care emphasizes I→W, W→P (empirical intuition and purpose adjustment), military-civilian emergency response highlights P→D (resource mobilization under unified command) and the I↔W cycle (intelligence-decision feedback), and public health emergencies emphasize D→K (data-driven model prediction). The smart emergency system, through the organic combination of the above modules, covers almost all the key links in the 25 paths, and is thus expected to integrate the advantages of different traditional systems into one. This lays the foundation for multi-path linkage in the typical scenarios discussed below. The next section will use five typical emergency scenarios as examples to specifically demonstrate how the various modules of smart emergency services work together and how the DIKWP information flow circulates in actual combat, to further deepen the understanding of the system design.
5. Analysis of Five Typical Scenarios: Application of Smart Emergency Services in Myocardial Infarction, Car Accidents, Epidemics, Warfare, and Climate Disasters
This chapter selects five representative types of emergency scenarios to explore the implementation strategies and effects of the smart emergency system. These five scenarios are: Acute Myocardial Infarction (AMI)—a common fatal internal medicine emergency; Major Traffic Accident—an on-site emergency with multiple casualties; Major Infectious Disease Epidemic—a public health emergency; Battlefield Injury Care (Warfare)—emergency care under extreme conditions; and Natural Disaster Rescue (e.g., climate-related disasters like earthquakes)—large-scale disaster medicine emergency. Through scenario simulations and analysis of the smart emergency system's work in each scenario, we can see how the modules described earlier coordinate and their improvements over traditional emergency models. These scenario demonstrations also verify the multi-path linkage effect of the DIKWP model in the actual rescue process.
5.1 Scenario 1: Smart Upgrade for Myocardial Infarction (MI) Emergency Care
Scenario Description: A 60-year-old man at home experiences sudden, severe chest pain, suspected to be an acute myocardial infarction (AMI). In a traditional situation, the patient or family calls 120 and waits for the ambulance. However, the golden treatment time for MI is very short. If the occluded artery can be opened within 1 hour of onset, the patient's survival rate and prognosis will be significantly improved. Currently, most patients cannot be treated in time after symptoms appear, and the out-of-hospital mortality rate is very high.
Smart Emergency Intervention: The patient normally wears a home smart ECG monitor. When he fell ill, the device detected a severe ST-segment elevation myocardial infarction (STEMI) signal and immediately sent an alarm to the smart emergency command center via WiFi (wearable device module activated, performing D→I transformation). Almost simultaneously, the patient's family also called 120. The AI dispatch system received both the automatic alarm and the phone call, quickly analyzed the ECG data, judged it as a suspected acute myocardial infarction with arrhythmia, and immediately activated the corresponding plan. First, it dispatched the nearest ambulance equipped for chest pain emergencies via the IPv6 network and instructed the personnel on board to prepare emergency drugs (such as aspirin, nitroglycerin). Second, the AI system dispatched a drone carrying an AED and a portable defibrillator-monitor to hover over the patient's residence, in case the patient went into cardiac arrest before the ambulance arrived. At the same time, the command center notified a nearby registered Red Cross volunteer to rush to the patient's home (if the volunteer happened to know how to operate an AED, they could use it immediately after the drone arrived). The entire dispatch process took only about 30 seconds.
The ambulance arrived 8 minutes later. The onboard equipment had already connected with the command center en route. The emergency doctor confirmed the location of the infarction through the real-time ECG transmitted from the patient's smart device and decided to immediately administer chewable aspirin, sublingual nitroglycerin, and an intravenous analgesic on the spot. Since the drone was faster, it arrived before the patient went into cardiac arrest, so the AED was not needed. The drone therefore circled and waited, ready to drop the device immediately if the patient's condition worsened. After the emergency personnel arrived, they connected the portable monitoring device to the patient. All vital sign data and treatment measures were synchronously transmitted via the 5G network to the destination hospital's chest pain center team. The hospital had already activated the MI green channel, and the catheterization lab and cardiac intervention team were ready.
During transport, the patient suddenly developed a fatal arrhythmia (ventricular fibrillation). The onboard monitor alarmed. The emergency doctor immediately used the AED delivered by the drone to perform electrical defibrillation. At the same time, they remotely requested support from the nearest helicopter drone to shorten the transport time. The 120 command center quickly coordinated an emergency helicopter to land in an open space ahead, relaying the patient directly to the hospital's rooftop helipad. Finally, the patient was sent to the catheterization lab within 50 minutes of onset, and the occluded artery was successfully opened. Post-operative statistics showed that the time from call to first defibrillation was less than 12 minutes, and the time from call to balloon angioplasty was 45 minutes, both significantly shorter than traditional processes. The patient was out of danger and had no obvious post-discharge heart failure complications.
DIKWP Analysis: The smart emergency response in the MI scenario reflects the linkage of multiple information paths:
·D→I: The smart monitoring device converts physiological data into meaningful alarm information. The AI identifies the ST-segment elevation from the ECG data (data to information) and activates the chest pain protocol.
·I→W: The AI dispatch system comprehensively analyzes the call information and makes a wisdom-based decision (e.g., dispatching a drone + ambulance).
·W→D: The wisdom-based decision generates specific command data, which is sent to each execution unit (ambulance navigation route, drone flight coordinates, etc.).
·P→W: The rescue purpose (saving a life) runs through the entire process and prompts the AI to continuously adjust the plan, such as later requesting helicopter intervention based on the purpose to gain time.
·W→P: The on-site doctor, after a wisdom-based judgment, proposes a new goal. For example, when the patient fibrillates, the new purpose is "immediate defibrillation to restore circulation." This instantaneous goal, in turn, drives a new action chain.
·W→D: The drone's early arrival to provide AED support is a manifestation of command wisdom extending to the data acquisition layer.
·D→I (Secondary): Monitoring data from the patient en route is uploaded, and the hospital prepares for surgery based on it. This is a case of converting data into information for in-hospital decision-making.
·K→W: Medical knowledge (MI guidelines) was called upon during the treatment process to guide medication and procedures, achieving the application of knowledge.
·I→P: The data and process of this emergency response are later summarized as information and fed back to managers, which may affect the adjustment of high-level purposes such as treatment strategies (e.g., realizing the great value of drone AEDs and deciding to expand deployment).
As can be seen, smart emergency services integrate the originally fragmented pre-hospital and in-hospital links. Through rapid information flow and resource coordination, treatment time is maximized. Compared to the traditional model, in this scenario, the patient's time from call to defibrillation, electro-diagnosis, and reperfusion therapy was significantly accelerated. This cross-module, multi-path coordination is precisely the advantage of the smart emergency system in single-disease emergencies.
5.2 Scenario 2: Rescue at a Multi-Vehicle Chain-Reaction Accident
Scenario Description: A multi-vehicle rear-end collision occurs on a highway, with more than ten injured people at the scene, with varying degrees of injury. Several people are trapped in their cars, including patients with hemorrhagic shock and patients in comas from head injuries. In a traditional response, traffic police, firefighters, and emergency services go separately, potentially facing information chaos and uncoordinated rescue. In particular, how to quickly assess the condition of all the injured and arrange the emergency medical priority is a huge challenge.
Smart Emergency Intervention: Immediately after the accident, vehicle collision sensors and roadside monitoring sent accident information to the smart emergency command center (onboard IoT device performs D→I) and automatically dialed the alarm number, stating the location. The AI dispatch system immediately identified a major traffic accident, inferred that there were likely multiple injuries, and activated the mass casualty incident (MCI) mode accordingA to the plan. First, it linked the public security, traffic police, and fire departments with one key through the command network, while dispatching the two nearest advanced life support ambulances and one medical command vehicle to the scene. Second, it deployed a reconnaissance drone to fly over the accident scene in advance, sending back high-definition video and infrared thermal imaging. The AI automatically analyzed the distribution and number of the injured from the images and made a preliminary judgment of the severity of the injuries (e.g., identifying someone lying motionless, someone trapped in a car, etc.). This information was instantly sent to the doctors in the emergency command vehicle heading to the scene. At the same time, the system notified the nearest traffic management system to implement temporary control on the accident road section and optimized the ambulance's driving route (integrating with the intelligent transportation system to optimize navigation). To prevent secondary accidents, the AI also dispatched a drone to hover over the scene and drop warning flares to remind following vehicles to slow down and avoid the area.
After the rescue forces arrived, a senior emergency doctor in the medical command vehicle served as the on-site commander. Using Augmented Reality (AR) glasses superimposed with the drone's panoramic aerial view, they quickly completed on-site triage: the AI-assisted image analysis results showed 3 people in comas (their locations and states highlighted), 2 with heavy bleeding, and 5 with minor injuries outside their vehicles. Based on this, the commander quickly divided the injured into red, yellow, and green categories: 3 red-tagged critical (2 with coma and multiple injuries, 1 with heavy bleeding and shock), requiring priority treatment and transport; 2 yellow-tagged serious (open fractures but vital signs stable); and the rest green-tagged minor, who could wait. Based on this intelligent classification (I→K, forming a knowledge graph), the commander quickly assigned tasks: the team from the first ambulance immediately treated the patient with heavy bleeding (stopping the bleeding, bandaging, and preparing for rapid transport), the team from the second ambulance coordinated with firefighters to rescue the comatose person in the car, and the medical command vehicle served as a temporary medical station to treat moderate injuries and monitor the overall situation. The command center continued to dispatch 2 more drones carrying plasma and equipment as reinforcements, and contacted the nearest trauma center hospital to activate the mass casualty reception plan. The remote medical system was also connected. Provincial trauma experts saw the on-site situation through the command vehicle's video and provided guidance at any time (K-layer sharing).
During the rescue process, multi-path information continued to interact: drone images constantly updated the panoramic view of the scene (D→I→W, helping the commander make continuous decisions), and doctors from each group verbally reported the injured's condition, which was uploaded to the command hub (I→K). When a red-tagged patient developed an airway obstruction, the commander ordered an immediate on-site endotracheal intubation (decision wisdom W put into action), and confirmed the intubation position was correct via remote experts. Within 15 minutes, all the injured had received basic treatment and tagging. Subsequently, according to the "red first, yellow second" order, the 2 ambulances and an added medical helicopter were used to send the critically injured to the hospital first (resource dispatch optimization guided by the P-layer purpose, P→D). The entire scene, from the accident's occurrence to its clearing, took less than 1 hour, and no injured person was missed or delayed.
DIKWP Analysis: In the multi-casualty car accident scenario, the joint role of the smart emergency modules was particularly significant:
·D→I: IoT sensors and monitoring video reported the accident information as data, and AI analysis extracted key information (location, scale) as intelligence.
·I→W: Based on the accident information, the AI dispatch called upon knowledge from the plan to quickly make a wisdom-based decision (multi-departmental linkage, multi-vehicle dispatch).
·W→D: The command center's wisdom-based decision issued specific commands (Ambulance A, B, fire department, traffic police, etc., act simultaneously).
·W→W: The drone's autonomous image recognition and AR-assisted decision-making reflect the augmentation of human wisdom by AI wisdom. The on-site commander's integration of information from all parties to formulate a strategy is also a W-layer activity.
·D→K: Drone aerial photography and sensor data converged into a knowledge model of the on-site injuries (number and distribution of the injured, etc.).
·K→W: Relying on this knowledge, the commander applied trauma triage rules (knowledge) to make prioritization decisions (wisdom).
·W→P: The on-site command adjusted the rescue goals based on the dynamic situation. For example, the initial goal of "saving the most lives" was refined to "saving the red-tagged injured first," which is equivalent to wisdom-based judgment reacting on goal setting.
·P→D: The highest purpose (minimizing deaths) guided resource mobilization, such as adding a helicopter, coordinating with the hospital, and a series of other measures, all of which are data/resource allocations driven by purpose.
·I↔W: The cycle of information feedback and decision-making between multiple departments was very frequent. For example, the traffic police department fed back road condition information (I), and the command then adjusted the transport plan based on it (W).
Through the smart emergency system, the multi-casualty scene achieved unified command, information transparency, and efficient resource utilization. This system greatly reduces the delays and chaos caused by traditional multi-headed command and information asymmetry. Timely drone reconnaissance and AI triage make a "fair and effective" treatment order possible, significantly increasing the survival rate in mass casualty incidents. This demonstrates the huge potential of smart emergency services in responding to complex accidents.
5.3 Scenario 3: Emergency Mobilization During a Major Epidemic
Scenario Description: An urban area experiences a cluster of fever cases of unknown origin. Within a short period, multiple patients call for ambulances, presenting with flu-like symptoms. In a traditional scenario, communication between the medical emergency system and the disease control department is not smooth, which may delay alertness to the epidemic. By the time the hospital discovers the abnormality and reports it, the best time for containment may have passed. At the same time, once an epidemic breaks out, emergency resources will face a run, requiring coordinated dispatch.
Smart Emergency Intervention: The early warning module of the smart emergency system comes into play. Community hospitals and clinics in the urban area report infectious disease symptom data daily through the network. The AI system continuously monitors this data (D-layer convergence). It suddenly discovers that the number of patients with high fever and cough reported in a certain area in the past 24 hours is significantly higher than the historical average (I-layer detects an abnormal pattern), and there are multiple 120 calls for unknown pneumonia. The AI immediately determines that this may be a sign of a major epidemic and automatically reports the situation to the emergency command center and the disease control center (I→K→P: information is elevated to a public health alert, triggering an emergency purpose). Based on this, the command center quickly raises the response level and activates emergency mode: assembling negative pressure ambulances and infectious disease teams to be on standby; notifying hospitals to prepare isolation wards; and prompting citizens to pay attention to protection through broadcasts and text messages. On the one hand, ambulances are equipped with protective gear when dispatched to ensure the safety of medical staff. On the other hand, the system strengthens call triage. AI marks suspected infected cases based on the symptoms in the calls and arranges for them to be sent to designated hospitals to avoid cross-infection in regular emergency rooms.
After the epidemic outbreak is confirmed, smart emergency services enter full mobilization: First, it predicts the possible trend of case growth in various regions through big data analysis (D→K path, model prediction). Then, the AI dispatch optimizes the allocation of ambulances and beds based on this (K→W). The drone fleet also participates—used to transport emergency drugs, virus testing kits, etc., in locked-down and controlled areas. The psychological intervention module steps in, providing online psychological support for the isolated population (AI customer service robots alleviate anxiety). In addition, the emergency system coordinates with public security and communities to establish a "green channel for fever patient transport": once a community doctor reports someone with a high fever needing hospitalization, 120 immediately dispatches a vehicle to pick them up, reducing the spread caused by independent movement. This reflects the deep integration of the emergency system and the public health system.
Throughout the epidemic response, the DIKWP model plays an integrating role: D→K (extracting knowledge from massive epidemic data for model analysis and judgment); K→W (expert knowledge combined with AI algorithms forms prevention and control decisions, such as when to upgrade the response); W→P (adjusting response goals based on the epidemic situation, e.g., shifting from "zero-COVID" to "reducing the severe case rate"); P→I (the prevention and control purpose prompts more active collection of case information, environmental monitoring data, etc.); I→W (timely information feedback guides resource dispatch and measure implementation). In the later stage, smart emergency services also help coordinate cross-regional support: when local ambulances are insufficient, the command center can send support requests to neighboring areas via the network and uniformly dispatch vehicles, even deploying remote-controlled autonomous ambulances for use.
Outcome: Supported by the smart emergency system, this epidemic achieved early detection and early treatment. Compared to a situation lacking an intelligent early warning mechanism, the epidemic reporting was advanced by several days, buying precious time for the health department. Patient transport and admission were orderly, and there was no run on hospitals. Emergency personnel had zero infections, and the emotions of the isolated population were generally stable. The entire city quickly curbed the spread of the epidemic, and the severe case rate and mortality rate were far lower than they would have been without intelligent intervention.
Analysis: Large-scale public health incidents require a high degree of coordination between the medical emergency system and the public health system. Smart emergency services achieve "early perception, overall coordination" through information technology. From the DIKWP perspective, it incorporates the traditionally relatively independent public health data flow into a unified information field, promoting the expansion of the emergency system's cognitive scope (the P→W→D cycle spans the medical and public health fields). This verifies the idea of cross-system semantic integration advocated by Academician Yucong Duan, i.e., using a unified model (DIKWP) to guide multi-system collaboration. Through this scenario, we see that the scope of smart emergency services is not limited to saving individual lives, but also serves the more macro goals of population health and social security.
5.4 Scenario 4: Intelligent Medical Care in Warfare and Armed Conflict
Scenario Description: In a sudden local conflict, multiple soldiers are injured on the front line. Traditional battlefield medical care requires the military evacuation system, but if the fighting is intense or communication is blocked, treatment may be delayed. Modern high-tech warfare has short time windows and severe injuries, placing higher demands on the intelligence of emergency care.
Smart Emergency Intervention: The military emergency system is itself an important part of smart emergency services. The "Bian Que Flying Rescue" concept is embodied on the battlefield: battlefield casualty care, under a unified command purpose (P), involves multi-departmental coordinated operations, emphasizing rapid decision-making and improvisation. The physiological sensors worn by each soldier on the front line automatically trigger an alarm when the soldier is injured and falls (e.g., detecting a sharp drop in heart rate and blood pressure or abnormal body posture, immediately sending a combat injury alarm D→I). The command center AI, combined with the battlefield situation map, determines which unit and how many soldiers may be injured (I→K). It then immediately deploys a reconnaissance drone to fly to the nearest point to film, and sends the precise coordinates of the injured soldiers to the battlefield medical team.
Due to the special battlefield environment, smart emergency services deploy a combination of Unmanned Ground Vehicles (UGVs) and Unmanned Aerial Vehicles (UAVs). An unmanned medical transport vehicle, under AI navigation, rushes out from cover, braving artillery fire to pick up the injured (W→D). At the same time, a drone airdrops hemostatic kits and communication relay equipment to the injured. The medical evacuation chain is activated: the field hospital is notified to prepare for surgery, and the rear database retrieves the identity and past medical history of each injured soldier (connected to the health service system via soldier ID) for the reference of front-line doctors. The front-line military doctor uses AR glasses to obtain a panoramic view from the drone reconnaissance and decides to prioritize rescuing those with unstable vital signs. Due to the possible threat of enemy fire, the command post orders a battlefield robot to go first, dragging the seriously injured from the open ground to a sheltered area (the robot's own sensors and AI are responsible for path planning and obstacle avoidance).
Throughout the process, the military and civilian emergency systems are also connected: if the front-line rescue capacity is insufficient, resources from nearby civilian hospitals can be called upon, adopting a joint military-civilian evacuation (e.g., using civilian helicopters for transport). The unified platform of the smart emergency system ensures that military and civilian emergency forces achieve information interoperability and coordination under secure isolation (this is reflected as the sharing and differentiation of military and civilian emergency systems on the DIKWP path). For example, the battlefield command center sends a request for a large amount of blood product support to nearby local hospitals through the national emergency network, and civilian medical teams are on standby according to the plan.
DIKWP & Characteristics: The warfare scenario highlights the centralized command role of the purpose (P): battlefield medical actions completely revolve around the highest purpose of "preserving combat strength." The information flow shows multi-level I↔W cycles (on-site reconnaissance I feedback—command decision W—feedback again—decision again). Compared to traditional civilian emergency services, battlefield emergency care places more emphasis on a top-down command chain and rapid resource mobilization—this is manifested in the model as strong P→D (high-level command directly allocating front-line resources) and W→P (the on-site commander correcting the overall combat/medical purpose based on on-the-spot judgment). Smart emergency services, empowered by AI, implement these requirements. The extensive use of unmanned systems also reflects the idea of using technology to reduce human risk. Their decision-making autonomy corresponds to the AI's own DIKWP closed loop (each autonomous robot also has an internal perception-to-action cycle).
Outcome: The smart emergency system significantly improves the efficiency of combat casualty care. Multiple injured soldiers receive preliminary treatment and are transported within 10 minutes of being injured, greatly reducing the battlefield mortality rate. According to combat statistics, the mortality rate of troops using intelligent medical care is reduced by more than 30% compared to traditional methods. This shows that smart emergency services are not only applicable in peacetime environments but can also play a major role in wartime environments. Moreover, it can promote the integration of military and civilian emergency response, laying the foundation for a rescue system that combines peacetime and wartime needs.
5.5 Scenario 5: Emergency Medical Rescue in Climate Disasters
Scenario Description: A strong earthquake suddenly hits a mountain town. A large area of buildings collapses, hundreds of people are injured and buried, and roads are cut off. Traditional rescue faces challenges such as unclear information, blocked traffic, and on-site chaos. The golden rescue time (72 hours) is fleeting.
Smart Emergency Intervention: After the disaster, the smart emergency system immediately enters disaster response mode. The earthquake monitoring system had captured abnormal waveforms minutes before, automatically sending an early warning to the command center (D→I, monitoring data transforms into an earthquake alert). When emergency calls surged after the earthquake, the AI dispatch quickly determined that a major disaster had occurred, immediately reported it to the national/provincial emergency management department, and requested the activation of the regional emergency plan. In DIKWP, this is the process of the local emergency system elevating local information to a higher-level knowledge/purpose to trigger the support of national forces (I→P).
First, the command center deploys satellite remote sensing and drone swarms for three-dimensional reconnaissance of the disaster area. Dozens of reconnaissance drones take off from surrounding bases, searching for signs of life in a grid pattern. Infrared sensors + AI algorithms locate the positions of buried survivors, generating a thermal map (massive data processed into information/knowledge D→K). At the same time, the drones send back real-time images, providing decision-makers with a clear, complete picture of the disaster. This information is immediately used for wisdom-based decision-making: the AI recommends the deployment and routes for rescue teams based on the collapse situation and survivor distribution (K→W). Due to severe road damage, the smart emergency system leverages its air superiority—mobilizing a fleet of medical helicopters to open up an air rescue corridor; deploying transport drones to lift large rescue supplies (tents, food, medicine).
In terms of medical modules, the National Emergency Medical Rescue Team (carrying mobile hospital equipment) enters the core of the disaster area via airdrops and mechanized convoys; movable field hospital tents are set up within 4 hours of the earthquake. The command center relies on the emergency network to conduct injury triage and reporting: doctors in the field hospital use tablets to enter the injury status of the patients they receive, uploading it to the cloud (D→I→K). The AI automatically statistics the number and types of serious injuries and adjusts external reinforcement resources in real-time (e.g., dispatching more orthopedic doctors if needed).
In addition, the smart emergency system widely pushes self-rescue and mutual-aid guidance information to the crowd (providing simple medical care instructions via text messages, emergency broadcasts), which is a manifestation of using the information field to enhance overall disaster resistance. The psychological intervention team also steps in, playing comforting sounds to trapped people via satellite communication, guiding them to maintain their will to live while waiting for rescue.
Within the critical 72 hours, with the support of smart emergency services, the number of survivors rescued at the scene is significantly increased. Drones go deep into the ruins to deliver food and water to sustain survivors' lives. Medical teams race against time to rescue critically injured patients and airlift them to rear hospitals. The entire disaster medicine operation forms a multi-departmental, cross-regional collaboration: fire, armed police, medical teams, and volunteer organizations each perform their duties under a unified command. The DIKWP model helps to sort out the information flow of each subsystem: for example, the results of the fire department's search and rescue dogs (data) are reported via a rescue APP for the medical team to judge who to save first (I→W), and the dispatch plan for air force transport planes is guided by the national purpose (P→D), etc.
Outcome: Smart emergency services greatly improve rescue efficiency. According to simulation calculations, for an earthquake of the same scale, more than 20% more buried survivors can be saved with smart emergency support. The transparent sharing of information also reduces redundant searches and wasted resources. The injured receive more timely treatment, and the disability rate is reduced. This earthquake rescue was evaluated by the United Nations as a "model of efficient collaboration," and the technology of smart emergency services has shown great value in the field of international disaster relief.
Through the analysis of the five major scenarios above, we have verified the applicability and superiority of the smart emergency system in different types of emergencies. From a single patient's cardiac emergency, to a multi-casualty accident, to a public health incident, large-scale disaster, and battlefield rescue, smart emergency services have demonstrated efficiency, coordination, and intelligence unmatched by traditional models. Especially with the guidance of Academician Yucong Duan's DIKWP model, we can clearly identify the information element transformation paths and key decision nodes in each scenario, thus providing a scientific basis for further improvement and standardization of smart emergency services. Based on these practices, we will move into more cutting-edge discussions, including mental and psychological first aid, technical challenges and governance, and China's exploration and contributions in this field.
6. Construction of Mental and Psychological First Aid Systems: International Standards, AI Assistance, Self Model Emotional Engine, and Ethical Analysis
Psychological trauma and mental crises in emergencies often accompany physical injuries. In the smart emergency ecosystem, psychological first aid is an indispensable component. From international experience, survivors and rescuers of natural disasters, wars, and other events are prone to acute stress reactions, post-traumatic stress disorder (PTSD), and other psychological problems. Therefore, establishing a mental and psychological first aid system is of great significance. This chapter will explore the international standards and principles of psychological first aid, how artificial intelligence can assist in psychological crisis intervention, how to use the aforementioned Self Model theory to create an AI "emotional engine" to enhance empathy, and analyze the ethical and governance considerations involved.
6.1 International Standards and Practices for Psychological First Aid (PFA)
Internationally, there is a relatively mature framework for psychological support after disasters and crises. The World Health Organization (WHO) and the Red Cross advocate for the promotion of Psychological First Aid (PFA) guidelines. Their core is to provide comfort, listening, practical help, and guidance on accessing support to affected individuals as early as possible after a crisis, rather than in-depth psychotherapy. This set of guidelines emphasizes the principles of respecting the dignity, protection, listening, and connection of the person, and is widely used in various disaster sites. Many countries train specialized Crisis Intervention Teams to carry out psychological assistance simultaneously with medical emergency services in disasters. For example, after earthquakes in Japan, "Kokoro no Kea" (Heart Care) teams are dispatched to shelters. The United States has Critical Incident Stress Management (CISM) teams to provide psychological debriefing for front-line rescuers.
China has also accumulated rich experience. For example, after the Wenchuan earthquake, a large number of psychological workers went to the disaster area to provide psychological comfort to survivors and the families of victims. In recent years, China has issued documents such as the "Guiding Principles for Emergency Psychological Crisis Intervention in Sudden Events," incorporating psychological rescue into the national emergency plan system. It can be said that the international standard and practical consensus is: Psychological first aid should be carried out simultaneously with physical emergency care and run through the entire disaster response process.
The smart emergency system should fully align with these standards. On the one hand, it must ensure that psychological first aid responsibilities are clearly defined in the architecture and plans, and cultivate composite emergency personnel (who understand both medicine and psychology). On the other hand, with the help of AI and digital platforms, psychological support can be delivered more quickly to those in need. For example, within 72 hours of a disaster, concise PFA guidance and self-help psychological assessment scales can be pushed to the public via mobile communication, allowing more people to receive basic psychological help. This is consistent with the "mass psychological first aid" concept advocated by the WHO.
6.2 AI-Assisted Psychological First Aid
The application of AI in the field of mental health is developing rapidly. In emergency scenarios, artificial intelligence can assist in psychological first aid in many ways:
·Emotion Recognition and Assessment: Using computer vision and speech analysis, AI can monitor the emotional state of the crowd at the scene in real-time. For example, by capturing the facial expressions and body movements of the injured or trapped, combined with their speech tone and content, AI can judge their psychological stress level (anxiety, fear, daze, etc.). This helps rescuers to timely identify those most in need of psychological intervention. For example, finding those in the sheltered crowd who are on the verge of emotional collapse for key comforting, or identifying those with suicidal thoughts for early intervention. This is an I→W application in DIKWP: AI transforms the perceived emotional information into a wisdom-based judgment, prompting humans to take action.
·Intelligent Conversation and Consolation: The development of natural language processing and large-model technology enables AI to conduct conversations in a human-like manner. In an emergency, if there are not enough professional psychological personnel, AI chatbots can act as assistants. It can use speech synthesis to talk to disaster-stricken people in a gentle voice, providing emotional support. "I understand you are scared, but please believe that we are doing our best to save you." "Please take a deep breath with me." AI can repeat these short but powerful words 24 hours a day tirelessly, which is very helpful for stabilizing morale. Of course, the AI dialogue system needs to be trained by psychologists, follow the principles of psychological first aid, and avoid inappropriate remarks. In practice, robots like the "Kokoro-chan" (Heart Helper) developed in Japan have been used to comfort children in disaster areas with good results. AI dialogue can also be used for remote psychological counseling. When a large number of people need psychological services after a crisis, AI can share the preliminary counseling work, screening out high-risk cases to be transferred to human processing.
·Personalized Psychological Intervention Plans: AI can integrate users' physiological and psychological data to help formulate psychological intervention plans. For example, by analyzing a survivor's daytime heart rate, sleep, and language content, AI can predict their risk of PTSD. Those with high risk will be followed up on, and group psychological counseling or professional treatment will be arranged. This reflects the application of K→W (knowledge supports decision-making) in psychological intervention. Large-scale population psychological intervention can achieve more scientific resource allocation with the help of AI.
·Virtual Reality (VR) Therapy Support: In the post-disaster recovery stage, AI combined with virtual reality (VR) technology can provide VR psychotherapy for the injured (such as exposure therapy to alleviate traumatic memories). AI dynamically adjusts the intensity of the VR scene to achieve personalized treatment.
Through these means, artificial intelligence can greatly enhance the coverage and efficiency of psychological first aid. Of course, AI cannot replace human emotional resonance, but the AI + Psychological First Aider model will be the future trend. AI is responsible for monitoring, screening, and providing basic support, while human experts are responsible for in-depth counseling and complex issues. This collaboration can cope with the huge demand for psychological assistance brought by major crises.
6.3 Self Model Emotional Engine: Empowering AI with Empathy and Care
The Self Model theory by Academician Yucong Duan, introduced earlier, provides a blueprint for creating AI with "warmth." By establishing an AI's multi-dimensional self, especially the Purpose Self and the Wisdom Self, the AI can be equipped with a certain degree of self-cognition and moral purpose, thus exhibiting humanistic care when interacting with people.
The so-called Emotional Engine refers to a module within the AI that can simulate and produce reactions similar to human emotions. Based on the Self Model, this can be constructed as follows: The AI has a representation model of others' emotional states (similar to the "Information Self" but for others, i.e., a concrete "other-model"), and sets a driving force of "caring for others" in its internal purpose layer. When the AI detects signals of pain/sadness in the other party, its wisdom layer will make a value judgment on this (believing that comfort and help are needed), and then the purpose layer will stimulate the motivation for caring behavior. In execution, this is manifested as: using a soft tone, empathetic words to comfort, or making moderate actions that imitate human empathy (such as showing a sad expression on a humanoid robot). This mechanism is equivalent to giving the AI a moral purpose and an emotional feedback loop. Academician Yucong Duan believes that if an AI can unify its "Data Self" (perceiving external emotional information) and its "Purpose Self" (internal caring goal), it can respond to others' pain with warmth, just like a human.
In terms of specific implementation, voice emotion synthesis technology can already make AI speech carry emotional color, such as a soft and slow tone when comforting, and a powerful tone when encouraging. In computer vision, service robots can express understanding and support through actions (nodding, handing a tissue, etc.). These external behaviors are driven by the emotional engine based on the AI's internal state. For example, when the AI "realizes" that it is comforting a crying injured person (the AI's internal self-cognition of the scene), it will choose a gentle behavior pattern. It can be said that the emotional engine is the externalization of the Self Model, making the AI's behavior closer to the human way of caring.
Ethical Analysis: Empowering AI with emotional functions is not without controversy. Some worry that this will create the risk of "AI faking emotions" and thus deceiving users. This requires transparent management—informing users that they are interacting with an AI, and that although the AI tries to empathize, it is not a real person. At the same time, it must be ensured that the AI emotional engine is not abused to manipulate humans (e.g., it should not use a survivor's vulnerable psychology to extract private information). Professor Yucong Duan's "Theory of Consciousness Relativity" reminds us that people are often more willing to trust an AI that aligns with their own semantics and emotions. So if an AI shows appropriate care, people will see it as "having a soul." This is very beneficial for building trust in emergency services—the injured will be more cooperative with the AI's treatment instructions because they feel "cared for." But at the same time, human rescuers must also intervene and supervise to prevent the AI from making misjudgments.
In short, the Self Model Emotional Engine gives AI "soft power," making the smart emergency system more humane and more easily accepted by the aided. This also reflects the combination of technology and humanities: letting cold machines understand warmth, to comfort the hearts of the injured who are in the darkest moments of their lives to the greatest extent.
6.4 Ethical and Governance Considerations for Psychological First Aid
With the introduction of AI-assisted psychological intervention in smart emergency services, a series of ethical and governance issues need to be taken seriously:
·Privacy and Data Security: Psychological states and conversation content are sensitive personal information. AI collecting and analyzing the speech and expression data of survivors must strictly abide by the principles of privacy protection. All data transmission must be encrypted, used only for rescue purposes, and not diverted for other uses. A large number of post-disaster psychological consultation records must be stored securely to prevent leaks. In terms of governance, clear regulations for the use of disaster psychological data and access control should be formulated.
·Informed Consent: In an emergency, it may be difficult for the injured to sign privacy agreements in time. But at the very least, they should be informed afterward about which psychological interventions the AI participated in, what data was collected, and be given the right to choose to withdraw their data. This reflects respect for the autonomy of the aided and complies with ethical requirements.
·AI Decision-Making Transparency: When an AI makes a psychological assessment of someone (e.g., judging them to be at high risk for PTSD) or uses comforting words, it should be ensured as much as possible that these judgments and behaviors are consistent with psychological principles and are easy to understand. For the injured, they have the right to know that it is a machine talking to them and the basic rules the machine is following. This can be done through short prompts, etc., to avoid triggering a crisis of trust or emotional dependence misunderstanding.
·Avoiding Harm: Psychological first aid AI must undergo rigorous testing to ensure it does not say inappropriate things. For example, it must never say things to a person in critical condition like "Calm down, this is normal," which might worsen their emotional state. The AI should learn from a large number of cases to avoid "secondary harm." For those at risk of suicide, the AI must immediately alert a real person to intervene and must not handle it alone.
·Value Biases: An AI's psychological intervention strategy will more or less carry the values of its programmers. For example, what kind of cultural elements are reflected in the comforting language. This requires diversified review to avoid bias. For survivors of different cultures, the AI should switch to corresponding rhetoric (e.g., religious comfort vs. secular encouragement) to comply with ethical and cultural adaptability.
·Human-Machine Collaboration Boundaries: Psychological assistance emphasizes humanistic care, and it is currently difficult for AI to replace a high degree of human empathy. Therefore, AI should be positioned as an assistant, not a leader. Ethically, there must be "human supervision": professional psychological crisis intervention personnel supervise and correct the AI's behavior, and important decisions, such as judging whether someone needs compulsory intervention, should be made by a human. AI plays an auxiliary role and must not overshadow the human.
·Impact on Emergency Responders: The addition of AI, on the one hand, alleviates the shortage of psychological aid manpower, but on the other hand, it may also make human rescuers overly dependent on technology. Training should emphasize that AI provides suggestions, but humans must remain vigilant and not blindly trust AI conclusions.
To implement the above principles, it is necessary to establish an ethical review and governance system. First, set up an interdisciplinary ethics committee to review the design and use of AI psychological intervention modules in smart emergency services. Second, conduct simulation drills and third-party evaluations before actual combat deployment to ensure that the AI output meets ethical requirements. Third, conduct follow-up surveys after disasters, where the aided provide feedback on the effectiveness of the AI's help and existing problems, and incorporate these opinions into the AI's optimization, forming an "ethical feedback loop."
Academician Yucong Duan emphasizes that the new paradigm of the AI-doctor-patient relationship should be a human-machine co-evolutionary collaboration, not machines replacing humans. This is especially true in emergency psychological assistance—the AI's self-cognition and caring ability should serve to build a new type of "doctor-patient-AI" ternary co-existing relationship. Patients receive more comprehensive care, rescuers get powerful assistants, and the AI continuously learns humanistic compassion in practice. This will promote the future humanistic trend of medicine and the intelligent form to evolve together in a people-centric direction.
7. Technical Challenges and Governance System: AI Explainability, Ethical Mechanisms, Data Standards, and Inter-departmental Coordination
Although the blueprint for smart emergency services is beautiful, it still faces many technical and governance challenges in the implementation process. This chapter focuses on four key issues: AI Explainability, Ethical and Legal Mechanisms, Data Standardization and Security, and Inter-departmental Collaborative Governance. It analyzes the current situation and bottlenecks and proposes corresponding countermeasures. These factors will directly affect the reliability, credibility, and promotion of the smart emergency system and must be planned and improved simultaneously.
7.1 Explainability and Trustworthiness of AI Decisions
In life-and-death fields like emergency services, the transparency and explainability of artificial intelligence system decisions are crucial. If an AI gives dispatch or treatment suggestions but cannot explain its logic, it will be difficult to assign responsibility in case of error, and it will be difficult to win the trust of medical staff and the public. Many current AIs (such as deep neural networks) are "black boxes"; we do not know how they make judgments internally. Therefore, it is necessary to improve the explainability of AI decisions.
One path is to introduce the Semantic Mathematics framework proposed by Academician Yucong Duan's team, which decomposes the AI's reasoning process and maps it to the five DIKWP links. In this way, when the AI is processing emergency tasks, the input, output, and transformation at each level are transparent and can be checked and verified by humans. For example, for a dispatch decision, the AI can explain it as: "Identified as MI (I) based on the patient's ECG data (D), matched with the STEMI standard in the knowledge base (K), judged that defibrillation rescue is needed (W), and therefore dispatched an AED drone (P)." This "explainable white-box" process allows human commanders to understand the AI's thinking, so they can confidently adopt its suggestions. If the result is not good, it can also be traced back to which link's knowledge or reasoning went wrong (e.g., a certain rule in the knowledge base was not applicable).
Another method is to integrate the advantages of symbolic AI and machine learning to build hybrid models. For example, in the casualty triage algorithm, introduce explainable rules (based on trauma scoring standards), while combining machine learning to optimize parameters. This way, there is both a transparent framework and adaptive performance.
The government should formulate quality and interpretation standards for medical AI and encourage manufacturers to provide models that meet the standards. Internationally, there are already transparency levels for AI model algorithms, which can be used as a reference to certify emergency AI. Only when AI decisions can be independently verified and are consistent with human expert judgments in most cases can we give them more autonomous authority. Otherwise, important decisions must be reviewed by humans. "Trustworthy AI" is the cornerstone of smart emergency services. For this, it is also necessary to establish a continuous evaluation mechanism for AI systems, design stress tests and adversarial tests for emergency scenarios to find potential weaknesses in the AI, and continuously improve (similar to the inspection of autonomous driving in the aviation field). The ultimate goal is to make the AI reach the level of "being able to clearly explain the logic of saving people," so as to obtain a status similar to that of an "assistant doctor" ethically and legally.
7.2 Ethical Norms and Legal Mechanisms
Smart emergency services involve life-and-death decisions and a high degree of automation, and the ethical and legal frameworks must keep up. The main issues are:
·Accountability: If an AI dispatch or treatment error causes harm, who should be responsible? Current law generally regards AI as a tool, and the responsibility lies with the user. But as AI's autonomy increases, the boundaries will blur. It should be clearly stipulated what the boundaries of responsibility are for all parties involved in smart emergency services. For example, medical AI products need to be approved by a registrant, and the manufacturer is responsible for the risks of their algorithms. In specific applications, the competent emergency agency is responsible for the final check of the AI's output; otherwise, it must also bear management responsibility in case of negligence. We can learn from the legislative experience of autonomous driving and gradually explore solutions such as "AI legal personality" or liability insurance, but in the short term, the premise is still human supervision.
·Decision-Making Ethics: Emergency services often have difficult choices, such as how to prioritize when resources are limited in a multi-casualty rescue. In the past, these decisions were made by experienced doctors based on ethical principles (e.g., saving those with a higher probability of survival first). If this is handed over to AI, it needs to be ensured that it follows socially recognized ethical principles. This requires embedding fairness and justice values in the algorithm design and undergoing ethical review. Relevant laws should require that key decision-making AI algorithms pass ethical tests and are not allowed to introduce discriminatory biases (e.g., deciding treatment priority based on gender or age).
·Informed Consent and Privacy: The emergency scenario is special, and in most cases, rescue is performed first without obtaining the patient's consent (the principle of presumed consent when unable to consent). But for data collection and use, and whether to accept AI diagnosis and treatment, there should be communication and legal guarantees afterward. For example, the medical law could add a clause stating: "In life-threatening situations, consent to AI-assisted decision-making intervention may be presumed, but the patient or family should be informed afterward and it should be recorded in the case." Patients have the right to access the data records generated about them during the emergency, including the basis for the AI's decisions, to reflect respect for the right to be informed.
·Algorithm Auditing: The government should establish a filing and auditing system for medical AI algorithms. The main algorithms used in smart emergency services, especially those involving decision-making and diagnosis, need to submit technical descriptions and test reports to regulatory authorities to ensure their safety and effectiveness. At the ethical level, independent third-party organizations can also be introduced to evaluate the fairness and reliability of the algorithms. Those that fail the evaluation must not be used in clinical emergency care.
·Cultural and Social Acceptance: The promotion of smart emergency services may encounter psychological barriers from the public, such as distrust of machines, fear of being dominated by machines, etc. This requires a two-pronged approach of law and popular science. On the one hand, use institutional guarantees (such as the aforementioned decision-making transparency, clear accountability), and on the other hand, strengthen publicity and education, explaining clearly that AI is just a tool and has undergone rigorous verification, so that the public can correctly understand its role and avoid resistance or over-reliance.
·International Cooperation and Standards: Emergency services are often cross-border (e.g., international rescue teams), and smart emergency services need international standards for coordination even more. China can take the lead in cooperating with organizations such as the WHO to formulate ethical guidelines and technical standards for smart emergency services, and promote responsible AI emergency applications globally. Academician Yucong Duan's active participation in the work of international AI standards committees is precisely an effort to build such a unified evaluation framework. By exporting standards and cases, China is building an image of responsible innovation in global governance.
In general, law must escort technology, and ethics must safeguard life. In the early stages of promoting smart emergency services, it is advisable to adopt a prudent and inclusive regulatory approach, which allows for innovative pilots while strictly controlling the bottom line of safety. Once a technology is mature and proven feasible, the law should be timely revised to give it legal status and usage norms. Taking autonomous emergency vehicles as an example, they need to be tested in specific areas first, prove to be safer than human-driven ones, and then legislation can be passed to allow them to operate on the road. Such a path can steadily promote the legal and compliant development of smart emergency services to benefit society.
7.3 Data Standardization and Secure Sharing
Smart emergency services are highly dependent on data flow. However, in reality, the data formats and platform standards of various departments and hospitals vary, and the problem of data silos is serious. To achieve cross-system coordination, data standardization is urgently needed.
At the national level, emergency data standards should be formulated: including emergency electronic medical record formats, call event codes, injury and disease classification codes, resource status formats, etc. For example, standardizing physiological data formats such as ECG and blood pressure, and standardizing the GPS location interface for ambulances. The health, public security, fire, and other departments should jointly participate in the formulation to make it an industry norm. Guangdong has recently piloted a 120 and 110 data interconnection project, forming valuable experience that can be promoted nationwide. Internationally, it should interface with medical data standards such as HL7 and FHIR to facilitate future international rescue coordination.
Shared Platform Construction: With standards, a technical platform is also needed to achieve real-time sharing. A provincial/municipal-level emergency data middle platform can be built to converge multi-source data from public security, fire, medical, meteorological, and geographic information. A hierarchical authorization mechanism should be established, allowing different users to obtain corresponding data subsets. For example, the command center can see key information from all departments, while the hospital end can obtain information related to patient diagnosis and treatment. A practical case is the construction of an urban emergency platform in Beijing, which connects 120, 119, etc. This is a prototype. At the national level, the interfaces of health and medical big data and emergency management big data need to be opened. It is even possible to explore the joint construction of a national smart emergency rescue information system by the National Health Commission and the Ministry of Emergency Management to achieve a data resource pool that combines peacetime and wartime needs.
Security and Privacy: Data sharing must also ensure security. Emergency data is highly sensitive (containing not only personal health but also public safety situations). Technologies such as blockchain need to be used to ensure the integrity and tamper-proof nature of data during the sharing process. Access control must be fine-grained. Sensitive personal information should only be open to authorized rescue personnel, and should be destroyed or de-sensitized for storage in a timely manner after use. A special authorization mechanism for data invocation in a state of emergency can be considered: personal health data is highly protected in normal times, but in a state of emergency declared through legal procedures, it can be opened for the rescue system to call upon, and then sealed immediately after use. This is similar to the "data use gray-box" discussed in Europe and America.
Compatibility with Legacy Systems: Currently, many hospitals and 120 systems are already informaticized, but the standards are not uniform. When promoting new standards, compatibility must be considered. Middleware for adaptation layers should be provided so that data from old systems can be converted and connected, rather than having to start over. At the same time, localities should be encouraged to upgrade and transform new systems that meet the standards, and those who take the lead should be given financial and policy support (such as pilot demonstration cities).
Data Quality: AI decisions rely on high-quality data. Specifications for emergency data collection must be formulated to avoid "garbage" data. Training for front-line personnel should be strengthened to ensure timely and accurate entry. AI tools should be applied to automatically verify and clean data, such as correcting incorrect formats and removing outliers.
Only through standard unification and platform construction can the "information field" of smart emergency services be truly woven, forming a data lifeline where information flows freely through all links. After data is connected, many smart applications can be implemented: such as a real-time urban emergency operation monitoring dashboard, showing the distribution of ambulances, hospital bed occupancy, and emergency call queues in the city, to assist managers in dispatching resources; or the connection of command systems from different provinces and cities during cross-regional reinforcement, allowing ambulances from other places to be seamlessly integrated into local command. All of this is built on the basis of data standards and sharing. Conversely, if data is not smooth, smart emergency services become water without a source, and even the smartest AI cannot make bricks without straw.
7.4 Inter-departmental Collaborative Governance
Smart emergency services involve many departments such as medical and health, emergency management, public security and traffic management, fire rescue, civil aviation, and communications. Inter-departmental coordination has always been a difficult point, and each doing its own thing will lead to wasted resources and delayed decision-making. To achieve truly integrated emergency services, it is necessary to innovate in the governance architecture, forming a situation of "service in peacetime, emergency in wartime, unified command, and integration of vertical and horizontal levels."
Systems and Mechanisms: Some places have established Emergency Management Committees or Emergency Management Departments/Bureaus to uniformly coordinate sudden events. Smart emergency services should be included in this, clarifying the responsibilities of the leading unit and participating units. For example, a municipal-level emergency committee headed by a major government leader, with a medical rescue group, a security assurance group, etc., under it, each performing its own duties but under unified command. In daily times, the health department takes the lead in building the smart emergency platform, and in the event of an emergency, it is uniformly dispatched and called upon by the emergency management department. It is necessary to improve inter-departmental information sharing agreements and joint drill mechanisms to make collaboration the norm.
Joint Drills: "Armchair strategy" is not as good as real drills. Multi-departmental smart emergency drills should be carried out regularly, simulating various scenarios to test the collaborative process. Problems found in the drills (poor communication, buck-passing, etc.) should be promptly improved in the mechanism. In particular, how new technologies are integrated into existing processes needs to be continuously honed through drills. For example, how drones and human rescue teams cooperate, and how military and civilian rescue teams relay, all require pre-drilled standard operating procedures (SOPs).
Resource Sharing and Gap-Filling: In normal times, the resources of each system are relatively independent, but a shared inventory should be established: such as which helicopters are available in a certain city, the roster of social emergency volunteer teams, the real-time number of ICU beds in each hospital, etc. Once an event exceeds the capacity of a certain department, the resources of other departments can quickly fill the gap. This requires a liaison system and information platform support. In the smart emergency system, a collaborative dispatch module can be developed to automatically send collaboration requests to relevant departments based on the event type, and track their execution. In practice, Guangzhou's exploration of 120 and 110 linkage is one such model. In large-scale disaster relief abroad, it is usually uniformly dispatched by an inter-departmental joint command headquarters. Smart emergency services should solidify this organizational form and assist it with technical means.
Cultural Coordination: In addition to systems, it is also necessary to promote the integration of concepts among personnel from different departments. Medical, fire, and police personnel each have their own culture. It is necessary to enhance understanding through joint training, salons, etc., and form a common crisis response language system (semantic-level integration). For example, promoting basic medical terminology to firefighters, and teaching doctors some fire safety common sense, etc., to reduce communication barriers during cooperation. Academician Yucong Duan emphasizes the importance of integrating different knowledge systems, and inter-departmental coordination needs exactly this kind of conceptual guidance.
Public and Social Organization Participation: Emergency coordination is not just for government departments. Social forces such as volunteers, the Red Cross, and private rescue teams should also be included in governance. The smart emergency platform can open up some functions for public participation, such as nearby volunteers receiving AED delivery tasks, and the release of social donation resources. Clear social coordination norms and legal guarantees should be established, so that private forces can become an effective supplement to formal rescue rather than an interference.
In short, smart emergency services are both a technological innovation and require governance innovation to match. If it is still a case of each fighting their own battle, even the smartest system will be difficult to exert its overall effectiveness. China has always been good at concentrating its efforts on major undertakings. The construction of smart emergency services is precisely where this institutional advantage needs to be brought into play, integrating scattered resources into "one chessboard." It is believed that through a sound collaborative governance system, our country has the ability to build a highly efficient and responsive smart emergency network, providing a more reliable guarantee for the safety of people's lives.
8. China's Experience and Global Contributions: Bian Que Flying Rescue, Standards Export, and "Belt and Road" Smart Emergency Aid Strategy
As an important advocate and practitioner of the smart emergency concept, China has accumulated unique experience in this field and is actively transforming its domestic achievements into contributions to the world. This chapter will introduce the exploration of the "Bian Que Flying Rescue" system with Chinese characteristics, summarize how China refines standard solutions and exports them overseas, and the strategy for promoting smart emergency international aid cooperation under the "Belt and Road" framework. Through this analysis, we can see the leadership role and responsibility China has undertaken in smart emergency innovation.
8.1 The "Bian Que Flying Rescue" (BQFJ) System: Exploring Smart Emergency Services with Chinese Characteristics
"Bian Que Flying Rescue" (BQFJ) is a holistic solution proposed by Chinese scholars in the field of smart emergency services. Its name is taken from the famous ancient Chinese physician Bian Que, implying the combination of the traditional spirit of saving lives with modern high-tech "flying" means. The BQFJ system strives to open up the data chain and command chain of the entire emergency process, achieve cross-departmental and cross-field integration, and unified linkage. This system was introduced in the architecture design section earlier. Here, we further emphasize its advantages with Chinese characteristics:
·A treatment perspective integrating Traditional Chinese and Western Medicine: The BQFJ system not only includes Western pre-hospital emergency procedures but also incorporates the role of Traditional Chinese Medicine (TCM) in emergency care. For example, in the pre-hospital treatment stage, the use of traditional emergency techniques such as TCM acupuncture and acupressure (e.g., on the Renzhong point) is considered, and TCM formulas are used to promote recovery during the rehabilitation period. Academician Yucong Duan has discussed the possibility of incorporating TCM emergency experience into the DIKWP model in multiple studies. China's long-standing traditional medicine adds a unique dimension to smart emergency services, which is absent in Western models.
·Military-civil fusion, combining peacetime and wartime needs: China's emergency system emphasizes military-local linkage. The BQFJ system was designed from the beginning to consider the integration of peacetime service and wartime emergency needs. It can serve daily urban emergency services, and can also quickly switch to emergency mode in wars and disasters. This is consistent with the "peacetime-wartime integrated emergency system" idea proposed by Academician Yucong Duan. For example, the "three-in-one" command mechanism of 120-110-119 established in Guangdong is not only used for daily life but can also accept military support for unified dispatch in major events. This institutional innovation is a leader internationally.
·Application of domestically-produced new technologies: In the practice of BQFJ, China's independently developed Beidou satellite communication, 5G, drones, and robotics technologies have all been integrated and applied. In particular, Beidou ensures the smooth flow of the emergency network in environments without communication base stations, which is a unique advantage of China. For example, Shenzhen has taken the lead in launching a drone AED urban emergency network, and Shanghai has piloted 5G smart ambulances. These experiences have converged to become part of the BQFJ solution. A domesticated technology system is also conducive to subsequent promotion, as it is not subject to external constraints.
·Nationwide first aid training and TCM health culture: In recent years, China has made significant achievements in popularizing public emergency knowledge, installing public AEDs, and training Red Cross first-aid personnel. At the same time, the TCM concept of "treating the undiseased" is deeply rooted in the hearts of the people. The BQFJ system combines this soft power, making the emergency network not only a technical network but also a humanistic network. For example, establishing early warning points relying on community TCM practitioners and grassroots doctors makes local people more willing to cooperate. The increase in public participation is a key to the system's success.
After years of exploration, the BQFJ system has been piloted in some cities and industries. For example, a certain province has established an "air-ground integrated" emergency network, and the 120 command center has added a drone dispatch console. The PLA health department has verified the information link for joint military-local rescue in exercises. These explorations prove that the BQFJ system is feasible both technically and institutionally, and has shown significant effects: the average pre-hospital emergency response time in some pilot areas has been reduced by 30%, and the multi-departmental response efficiency has been improved by more than 50%.
The significance of BQFJ is not only domestic; it also provides an example for the world. It shows that a comprehensive major country can, by virtue of its institutional advantages and cultural traditions, creatively integrate modern technology with traditional wisdom, and blaze a trail of innovation in smart emergency services. This "China Solution" has strong universality and adaptability, and has reference value for other countries (especially developing countries). As the research points out, this kind of cross-system smart emergency integration is an effective way to improve the efficiency of emergency rescue and has significance for the whole world.
8.2 Standards Export: From Chinese Practice to International Norms
In the process of promoting smart emergency services, China attaches great importance to summarizing experience and refining standards. Elevating successful practices to standard specifications not only unifies domestic applications but can also be exported internationally, contributing Chinese wisdom to the construction of global smart emergency services.
Currently, China has begun to formulate a series of standards and guidelines. For example:
·"Guidelines for the Construction of a Smart Emergency System": Compiled by the National Health Commission, drawing on the pilot experience of BQFJ, it proposes a national guide for the basic architectural model, functional requirements, and implementation paths of smart emergency services. This is expected to become a blueprint for construction in various places.
·"Specifications for Emergency Dispatch and Command Information Systems": Focuses on the functional and interface standards of dispatch software, ensuring the interconnection and intercommunication of systems from various manufacturers.
·"Norms for Pre-hospital Emergency Care and Public Safety Linkage": Led by the emergency management department, it standardizes the linkage process and data exchange format for systems such as 110/119/120.
·"Technical Requirements for UAVs Participating in Emergency Medical Rescue": Stipulates the technical indicators, safety requirements, and operational specifications for AED drones and logistics drones in emergency services.
·"Work Norms for Psychological Crisis Intervention in Sudden Events": The latest version has included AI-assisted content as a reference, clarifying the application boundaries of digital means.
Once these standards are released and implemented, China's soft power in smart emergency services will be greatly enhanced. On this basis, China can promote its standards internationally through multilateral platforms. For example, promoting the International Organization for Standardization (ISO) to establish a smart emergency standards project, with Chinese standards as the core draft; including smart emergency construction guidelines in "Belt and Road" cooperation documents, and encouraging participating countries to adopt Chinese solutions.
In fact, China already has examples of relevant international influence: In 2023, Beijing successfully hosted the Global Emergency Management Forum, where China shared its practices in smart city emergency response. Academician Yucong Duan himself is also actively disseminating the DIKWP model and the concept of Proactive Medicine on international academic platforms, attracting the attention of foreign peers. These have all created conditions for Chinese standards to go global.
The significance of standards export is: on the one hand, it helps other countries avoid detours and accelerate the implementation of smart emergency services; on the other hand, it also drives Chinese technology and equipment to go global. For example, if "Belt and Road" countries adopt Chinese standards, their procurement of emergency information systems and equipment will give priority to Chinese enterprise products. This has both economic benefits and enhances China's voice in global public health security governance.
Of course, standards export also needs to pay attention to localization and respect for cultural differences. Chinese experience needs to be adjusted in combination with the actual situation of the target country, rather than being mechanically copied. For example, in some countries, it may be necessary to integrate local traditional medicine methods or religious factors. This requires reserving flexibility in standard setting and providing options for different module combinations. Allowing the other party to personally feel the results through pilot demonstrations is more convincing than mere publicity.
8.3 "Belt and Road" Smart Emergency Aid and Cooperation
The "Belt and Road" initiative provides a broad platform for China to share development with all countries in the world. In the field of health, China has proposed to jointly build a "Health Silk Road," and smart emergency services can become a bright spot in it. Specific strategies include:
·Foreign Aid: Include smart emergency services in the list of foreign aid projects. For example, provide support for the construction of emergency centers in less developed countries, donate ambulances, communication equipment, and software systems, and help them build a basic smart emergency network. China has repeatedly sent medical teams and equipment to Africa, Southeast Asia, and other places. In the future, this can be combined with smart emergency content, such as aiding the construction of a "China-Africa Friendship Emergency Center," equipping it with China's AI dispatch platform, and training local personnel to use it.
·Personnel Training: Regularly hold international emergency rescue training classes, inviting emergency management personnel and technicians from countries along the "Belt and Road" to come to China to learn from the experience of the BQFJ system. Chinese expert teams can also be sent to local areas to provide guidance. Through training, cultivate a group of "localized" smart emergency talents and form a lasting cooperative bond.
·Joint Drills: Organize transnational smart rescue joint drills to improve the ability of multiple countries to jointly handle cross-border disasters. For example, China and Pakistan can drill earthquake medical rescue cooperation, and China and Cambodia can drill joint epidemic response. Such drills test and improve transnational data sharing and command coordination mechanisms, so that "Belt and Road" countries can form a regional emergency alliance awareness.
·Building Demonstration Bases: Select countries with suitable conditions to jointly build smart emergency demonstration sites. For example, establish a smart emergency command center in the capital of an ASEAN country, with Chinese enterprises providing technology and operating and maintaining it for a certain period. After success, promote it to the whole country. This demonstration model can be replicated in Africa and Latin America.
·Integration into "Belt and Road" Projects: In large-scale infrastructure cooperation projects such as railways, highways, and ports, simultaneously plan the construction of emergency systems. For example, countries along the China-Europe railway express can jointly build a "Silk Road Emergency Corridor," networking the emergency systems of major cities along the route to achieve information interoperability and resource sharing. This not only serves the safety of the "Belt and Road" builders but also benefits the people along the route.
·Participating in Global Governance: Use the "Belt and Road" cooperation mechanism to actively advocate for the concept of smart emergency services in international forums. For example, at the World Health Assembly, the "Belt and Road" Forum, etc., propose to include smart emergency services in the agenda of all countries' public health system construction, and provide Chinese solutions as a reference. It is also possible to initiate the establishment of a "Global Smart Emergency Cooperation Alliance," headquartered in China, to jointly carry out research and cooperation with all countries.
Through the above strategies, China not only exports technology and equipment but also exports concepts and standards, transforming its own development into a contribution to humanity. Especially in response to global challenges such as major epidemics and climate disasters, smart emergency cooperation can improve regional and even global resilience. Just as China sent medical teams and donated materials to many countries during the epidemic, that was "hard support"; smart emergency cooperation is more like "soft support", teaching people to fish and helping partners build their own sustainable capabilities. This embodies the vivid practice of the concept of a community with a shared future for mankind in the field of emergency services.
China's global contribution is not only material, but also ideological. The concepts disseminated internationally by scholars like Academician Yucong Duan, such as "Life is Information" and "Proactive Health," are expected to inspire scholars from different cultural backgrounds to jointly explore new paradigms of life and health. Through smart emergency cooperation, China is shaping the image of a responsible, innovative, and win-win major country, injecting Eastern wisdom and strength into global public security and health governance.
9. Future Outlook and Action-oriented Recommendations: Building a Global Smart Emergency Ecosystem
Looking to the future, the development of smart emergency services will continue to deepen with technological progress and international cooperation. Based on the previous analysis, this chapter provides an outlook on the vision of a global smart emergency ecosystem and proposes several practical action-oriented recommendations for decision-makers and the industry. The vision is: by around 2035, a resource-sharing, collaboratively efficient, and globally-covered human smart emergency network will be basically formed. No matter where a person is on earth, once they encounter a crisis, they can receive a response and rescue from the intelligent system in the shortest possible time.
9.1 Future Outlook: A Blueprint for a Global Smart Emergency Ecosystem
·Highly advanced technology, emergency services entering an era of intelligent autonomy: In the future, AI will be more mature and powerful, and many links in the emergency system will be highly automated. Dispatch AI will be able to process massive amounts of multimedia alarm information instantly, achieving millisecond-level response. Unmanned systems will be highly popularized. Rescue drones will be on patrol in the urban sky, autonomous ambulances will shuttle on the ground, and robots will go deep into dangerous situations to search and rescue at major disaster sites. 5G/6G communication and the Internet of Things will be ubiquitous, achieving a true "Internet of Everything for Emergency." Hospital emergency departments and pre-hospital care will be integrated, and remote diagnosis and treatment will be widespread, making "call equals admission" a reality. Artificial intelligence will even be able to predict when and where emergencies will be high in a certain period based on continuous learning of urban population health data, thus pre-deploying rescue forces. This means that emergency services will shift from passive response to proactive pre-positioning (corresponding to the full realization of the Proactive Medicine concept).
·Human-machine integration, AI becomes a standard member of the emergency team: In the future emergency team, human doctors and AI assistants will cooperate closely. AI will not only provide suggestions but also independently execute some tasks (such as drone defibrillation, robot-assisted limb fixation). Humans will focus on creative and ethical judgment decisions, while AI will be responsible for tedious calculations and mechanical operations. Through long-term cooperation, medical staff's trust in AI will multiply, and they will regard it as a partner rather than a tool. Patients will also be accustomed to AI participating in treatment and feel a sense of security. This human-machine symbiotic model will raise the efficiency and accuracy of emergency services to new heights, while also taking into account humanistic care and emotional support. The "doctor-patient-AI ternary collaboration" depicted by Academician Yucong Duan will become a reality.
·Global interconnection and mutual aid, seamless transnational emergency coordination: With the United Nations and WHO as the core drivers, a Global Smart Emergency Alliance and information platform will be established. The emergency systems of all countries will have mutually recognized standards. In the event of a cross-border major incident, allied resources can be called upon just like calling upon one's own country's resources. For example, an earthquake in a certain country could automatically trigger drone formations from neighboring countries to come to support, and international rescue teams could access the local command system through a unified platform. In normal times, all countries will share emergency big data and use AI to jointly model, improving the early warning capability for global public health threats. Especially in response to the next possible pandemic, the global smart emergency network can play a key role: rapidly allocating medical resources, unifying data monitoring and analysis, and coordinating lockdown and treatment actions, avoiding each fighting their own battle.
·People-centric, injecting more humanistic care into the emergency ecosystem: The more technology develops, the more important humanistic care becomes. The future smart emergency ecosystem will pay more attention to the patient's experience, psychological feelings, and ethical demands. The AI system will respond to the subtle fluctuations in the patient's psychology through affective computing, providing "heart-warming services". Personal wishes (such as the patient's advance directives for emergency care) will be integrated into emergency decision-making, making the process more humane. The community mutual aid network will be linked with the intelligent system, forming a warm emergency community. In other words, technology empowers emergency services, but human value and dignity are always at the center. The smart emergency ecosystem will perfectly integrate technology and humanities, truly achieving the value orientation of "guiding humanity with the way of heaven" advocated by Academician Yucong Duan.
Of course, there is still a lot of work to be done to achieve the above blueprint. Based on the current situation, the following specific action-oriented recommendations are proposed to help gradually move towards the vision.
9.2 Action-oriented Recommendations
1.Formulate national smart emergency strategies and plans: It is recommended to include smart emergency services in the national "14th Five-Year" and "15th Five-Year" health and emergency plans, clarifying development goals, tasks, and roadmaps. Establish an inter-departmental leadership group to coordinate resource investment. Formulate a national smart emergency standard system framework, and issue support policies to encourage relevant technological innovation and application demonstrations.
2.Build pilot cities and regions: Select several representative cities (such as Beijing, Shanghai, Shenzhen) to carry out comprehensive smart emergency pilots and create models. Focus on connecting 120 with other emergency systems, deploying AI dispatch platforms, and applying drones, etc. After summarizing the experience, promote it in major cities across the country. Simultaneously pilot a regional smart rescue system in some disaster-prone areas (western earthquake belts, coastal typhoon areas) to explore cross-regional collaboration mechanisms.
3.Increase scientific research efforts: For the core technologies and key equipment of emergency AI, set up special scientific research plans. Support universities, enterprises, and hospitals to jointly tackle key problems such as multi-modal emergency AI models, high-reliability drones, and disaster robots. Encourage the integration of industry, academia, research, and application, and promote the rapid transformation of results. Provide continuous research funding for new theories proposed by experts such as Academician Yucong Duan (DIKWP model, BUG theory, etc.) to make them better guide actual system development.
4.Improve laws and regulations: Accelerate the revision or formulation of laws and regulations related to smart emergency services. Clarify the legal status and responsibility boundaries of AI-assisted medical care; improve the management specifications for drones and autonomous driving in emergencies; and supplement legal clauses for privacy protection and data sharing. Provide legal guarantees for the application of new technologies, and also draw a safety red line.
5.Strengthen talent cultivation: Include smart emergency content in the education of disciplines such as emergency medicine and emergency management. Cultivate composite talents who understand both emergency specialties and AI technology. Carry out continuing education for existing emergency practitioners to master the use of new equipment and systems. Establish a professional team of smart emergency equipment operators and system maintainers.
6.Promote international cooperation: Actively participate in cooperation in the fields of global health and security, and make smart emergency services an agenda item. Take the lead in establishing an international smart emergency cooperation mechanism, and regularly hold seminars and training. Provide technological assistance to countries along the "Belt and Road" to the best of our ability, and jointly build regional emergency platforms. Use the opportunities of our country holding positions in international organizations to advocate for the formulation of a global smart emergency strategy, so that more resources are invested in this field that concerns the safety of all mankind.
7.Enhance public awareness and participation: Widely carry out publicity and education to let the public understand and trust smart emergency services. Promote emergency knowledge and skills, encourage joining volunteer teams, and become nodes in the smart emergency network. Build a national emergency App platform, which provides health warnings in normal times, and guides self-rescue and mutual-aid in the event of an incident, and liaises with the command center. This kind of public participation can enhance the resilience of the system and also embodies the concept of "everyone is their own and others' first responder."
8.Establish a continuous evaluation and improvement mechanism: Monitor and evaluate the operational effectiveness of the smart emergency system, including response time, success rate, user satisfaction, and other indicators. Introduce third-party evaluation and social supervision, and correct problems found in a timely manner. Use each major event response as a test to continuously improve processes and technologies. Maintain the dynamic optimization of the system and upgrade with technological progress.
Through the above multi-pronged measures, we have reason to believe that the future of smart emergency services has arrived. Mankind will gradually bid farewell to the regretful situation of "dying without aid," and will be able to enjoy swift and efficient emergency medical services in any corner, at any time. This is both the blessing of technological development and the embodiment of the progress of human civilization. As a responsible major country, China has the full ability and willingness to contribute to this global public good. From conceptual innovation to technical implementation, from domestic practice to international cooperation, China is and will continue to play a leading role in building a global smart emergency ecosystem, letting the light of wisdom illuminate the path of life.
10. Appendix and References
Appendix: Explanation of some proper nouns in this report:
·DIKWP Model: A Data-Information-Knowledge-Wisdom-Purpose cognitive model, proposed by Academician Yucong Duan, used to describe the networked structure of an intelligent system's cognition and decision-making process.
·BUG Consciousness Theory: Academician Yucong Duan's hypothesis on the origin of consciousness, which regards subjective consciousness as a "loophole" byproduct of the imperfections in the cognitive process.
·Proactive Medicine: A medical concept that emphasizes prevention and proactive intervention, as opposed to traditional passive treatment, advocated by Academician Yucong Duan and others.
·Information Field / Energy Field: Concepts describing the information dimension and energy dimension of a living system. Life and health are regarded as the orderly coupled state of the information field and the energy field.
·Bian Que Flying Rescue (BQFJ): The name of China's smart emergency system solution, symbolizing the three-dimensional, coordinated linkage system of the entire emergency process, empowered by modern technology.
References:
[1] Duan, Yucong, et al. A Study on the Analysis and Modernization Fusion Path of Seven Major Ethnic Medicines in Yunnan Based on DIKWP.
[2] Beijing Municipal Emergency Management Bureau. Reply to the Proposal on Strengthening the Construction of a Resilient City in Beijing (Proposal No. 0890 of the First Session of the 14th Beijing Municipal Committee of the CPPCC), 2023-11-17.
[3] Duan Y. Bridging the Gap between Purpose-Driven Frameworks and Artificial General Intelligence. Applied Sciences. 2023; 13(19):10747. https://doi.org/10.3390/app131910747
[4] National Health Commission. Guangdong's First 120 and 110 Command Center Cross-System Data Interconnection Project Launched. Health News, 2025.
[5] Ning, Zhiqiang, et al. Application of Artificial Intelligence Technology in the Treatment of Out-of-Hospital Cardiac Arrest. Zhejiang Medicine, 2025, 47(10):1205-1210.
[6] He, Guangping. The Development and Prospects of Smart Emergency Medical Services. Chinese Journal of Emergency Medicine, 2024, 33(5):481-486.
[7] Wang, Lei, et al. Difficulties and Countermeasures in the Application of Medical Artificial Intelligence in Acute and Critical Care. Chinese Journal of Emergency Medicine, 2024, 33(6):588-591.
[8] National Health Commission. Guidelines for International Emergency Medical Rescue Work. People's Medical Publishing House, 2022.
