1. Background and significance of the project

1.1 The transformation of the medical model is in line with the national strategy

Today's medicine is undergoing a transition from the traditional "passive medicine" model to the new paradigm of "active health/active medicine". Traditional passive medicine (disease-centric) has promoted the rapid development of acute disease diagnosis and treatment technology in the past century, but it faces many challenges in the context of chronic disease epidemic and aging population. China has put forward the national strategy of transforming from "disease treatment as the center" to "people's health as the center", and the "Healthy China 2030" Planning Outline There is a clear call for a shift in the medical model from reactive treatment to active prevention. **As a result, the concept of active health has emerged, emphasizing that the individual is the first person responsible for health, and through daily self-monitoring, lifestyle intervention, and early screening and early intervention, "**prevention before disease" is realized. Active health, as a useful supplement to passive medical care, has enhanced the awareness of preventive health care among the whole people, and has achieved remarkable results in reducing the burden on hospitals and delaying the occurrence of diseases. However, the active health model driven by individuals and market forces alone lacks unified standards and institutional support, and it is difficult to comprehensively reverse the situation of "emphasizing treatment over prevention". To this end, it is necessary to upgrade the medical paradigm as a whole at a higher level, which is exactly the direction advocated by active medicine.

As a revolutionary change to the medical model, active medicine provides a holistic view of medicine at the philosophical and institutional levels. Active medicine retains the technical advantages of passive medicine in professional diagnosis and treatment, absorbs the flexible mechanism of active health in daily prevention, and integrates more macro institutional design and value concepts, shapes "disease-free" as the ideal health benchmark, and promotes medical practice from simple treatment to focus on the harmonious coexistence of man, nature and society. This paradigm shift is in line with the country's strategic requirement of "people's health as the center", and shifts the focus of medical care to the health maintenance of the whole population and the whole life cycle. The "Healthy China Action (2019-2030)" proposes to move the threshold forward, put prevention first, and strive to make the masses not sick and less sick, which is the only way to achieve high health performance at a lower cost. Active medicine is a theoretical sublimation of this strategy: it emphasizes active monitoring and intervention at all stages of the life cycle to achieve continuous health maintenance from genetics, fetal period to old age. In terms of value orientation, active medicine advocates a higher goal of "centering on harmony between man and nature", rather than just eliminating the symptoms of disease. This concept coincides with the traditional Chinese medicine idea of "curing diseases before they occur" and the philosophy of "harmony between man and nature", and also coincides with the modern medicine to preventive medicine and whole health (One HealthThe international trend of transformation coincides.

From a national strategic point of view, the proactive medicine paradigm is proposed. On the one hand, China's aging population and the heavy burden of chronic diseases have made it difficult for traditional medical models to cope with the growing demand for health services. Healthy aging has risen to a national strategy, the "14th Five-Year Plan".The plan emphasizes the promotion of the integration of medical treatment and prevention and the integration of medical care and elderly care, and requires a shift from disease treatment as the center to health maintenance as the center of the service model. Active medicine provides a paradigm support for the Healthy China strategy by strengthening prevention in the system and integrating the advantages of traditional Chinese and Western medicine in technology. On the other hand, the rapid development of a new generation of artificial intelligence also provides a technical opportunity for the implementation of active medicine. Active medicine emphasizes multidisciplinary collaboration and institutional innovation, and needs to use artificial intelligence, big data and other technical means to achieve dynamic monitoring of personal health status and intervention and decision-making support. This is highly consistent with the development direction of China's vigorous promotion of digital health and Wisdom medical care. The Ministry of Science and Technology's "Active Health and Aging Science and Technology Response" key project is to deploy scientific research forces around these needs, focus on the demonstration and application of key technologies for active health, build a new health perception, identification, intervention and management technology system, and promote the establishment of an integrated service model for elderly care, rehabilitation, nursing and medical care. This project has theoretically been upgraded from active health to active medicine, which is in line with the requirements of the national strategic transformation and will provide forward-looking guidance for deepening the construction of a healthy China.

1.2 Artificial intelligence development trends and opportunities for active medicine

In the 20s of the 21st century, there is a new trend in the field of artificial intelligence (AI) from "perceptual intelligence" to "cognitive intelligence". AI systems represented by large-scale pre-trained models have unprecedented language understanding and knowledge reasoning capabilities, but they also expose problems such as "black box" decision-making, opacity and uncontrollability. To this end, AI research is evolving in the direction of "intelligent self-knowledge", that is, to enable AI to have the ability to reflect and explain its own cognitive processes, and to achieve explainable and controllable intelligence at a higher level. This trend brings important opportunities for the realization of active medicine: active medicine needs the support of "active AI", that is, artificial intelligence with autonomous purpose and transparent decision-making, in order to truly assume the role of active monitoring and intervention.

"Data-Information-Knowledge-Wisdom–Purpose (DIKWP)" proposed by Prof. Yucong Duan's team The AI cognitive model provides a key theoretical basis for this AI paradigm change. The DIKWP model adds the highest level of "Purpose/Purpose" on the basis of the classic DIKW (pyramid model) to form a five-layer network cognitive structure of data, information, knowledge, Wisdom, and Purpose. It realizes the full-link traceability of AI from perception to decision-making. This innovative model is considered to be the underlying path to solve the current "black box" problem of large models, laying the foundation for the realization of AI with autonomous awareness. As Professor Yucong Duan noted, "The DIKWP model builds a common cognitive language between humans and machines, allowing every step of the AI decision-making process to be traced, explained, and understood by humans. By embedding the critical layer of 'purpose' inside the model, we are not only making AI smarter, but also ensuring that it remains in service of human values and security needs." "This kind of design, which integrates human purpose into AI cognition, transforms AI from a tool that passively executes instructions to an active AI subject that can actively understand the target, actively obtain information, and take action.

The concept of "active AI" is a natural fit with the paradigm of active medicine. While traditional AI is mostly a passive response model, waiting for human input and then outputting results, active AI seeks to give AI a certain degree of autonomy: it can actively perceive the environment, collect the required data, reason about the hidden information, and intervene when necessary, driven by internal purposes. This coincides with the requirement of "active intervention" in active medicine. With the help of the DIKWP model and its derived artificial consciousness white-box evaluation system, we can ensure that active AI has the ability to have different cognitive levels and present the decision-making logic in a transparent way. For example, the DIKWP white-box evaluation standard led by the World Artificial Awareness Association has taken the lead in building a full-link evaluation system from data perception, knowledge reasoning to purpose recognition, which is used to quantify the "consciousness level" of large models 。 This evaluation system shows that by allowing AI to show each cognitive step and accept multi-dimensional assessments, the cognitive decision-making process of AI can be comprehensively analyzed, thereby setting a new benchmark for AI explainability. This project will make full use of these cutting-edge AI theoretical achievements, take the DIKWP model as the core framework, introduce the artificial consciousness white-box evaluation mechanism, and create an "explainable, controllable, and autonomous purpose" active AI system for the health management of middle-aged and elderly groups. This not only responds to the country's development requirements for a new generation of AI to be "safe and controllable", but also provides a path for active medicine: that is, through artificial intelligence empowerment, intelligent perception, semantic understanding and active intervention of individual health status.

1.3 The urgent need for health management of middle-aged and elderly people

China has entered a deeply aging society, and the health problems of middle-aged and elderly people are becoming increasingly prominent. Chronic non-communicable diseases (NCDs) are highly prevalent among the elderly, and the coexistence of multiple diseases among the elderly is common, putting medical resources under great pressure. The traditional medical model mainly treats the diseases that have occurred in the hospital environment, and it is difficult to cover the daily health management of the community and home scenarios in a timely manner. Due to the lack of active health monitoring and intervention, many elderly people often miss the opportunity for early intervention of diseases. In addition, problems such as cognitive decline and Alzheimer's disease are also growing rapidly, causing a heavy burden on individuals, families and society. At the national level, the goal of "healthy aging" is proposed, emphasizing the need to integrate health into the whole process of elderly life, strengthen preventive health care, and develop community and home-based medical care services. However, the current digital health management services for middle-aged and elderly groups are still imperfect, and there are the following pain points:

l Fragmentation of health data and lack of continuous monitoring: The medical data of the elderly are scattered in hospitals at all levels, and the health data in daily life (such as exercise, sleep, blood pressure, blood sugar, etc.) lacks systematic collection, making it difficult to detect health risks in time. It is necessary to establish a health record and continuous monitoring mechanism for the whole life cycle of individuals to realize the dynamic assessment and early warning of the health status of the elderly.

l Disconnect between prevention and intervention: At present, primary medical institutions and community health services are weak in early disease screening and health education, and elderly care services and medical services have not been effectively connected. When the elderly have sub-health signals (such as vision loss, memory loss) at home, there is often no timely professional intervention. There is an urgent need for an intelligent system that can give intervention suggestions at an early stage, and integrate medical, health care, and health care resources to actively serve the elderly.

l Insufficient integration of traditional Chinese and Western medicine resources: Chinese medicine has rich experience in health care and chronic disease management, but in the health management of the elderly, traditional Chinese medicine interventions (such as dietary therapy, acupuncture, exercises, etc.) and Western medicine methods (drugs, surgery, rehabilitation training, etc.) are often separated, and it is difficult for the elderly to obtain comprehensive solutions. How to integrate the advantages of traditional Chinese medicine and Western medicine to provide personalized and multi-dimensional health interventions for the elderly is a major challenge.

l Insufficient intelligence and age-appropriateness: The elderly have limited acceptance of high-tech products, and the current popularity of wearable devices and health apps among the elderly is low, and the design for the elderly is insufficient. It is necessary to develop more non-inductive and convenient ubiquitous health monitoring technologies (such as non-invasive sensing, smart home embedded devices, etc.), as well as interactive means close to the use habits of the elderly (such as voice assistants, large-character displays, etc.) to reduce the digital divide.

In view of the above needs, this project is of great practical significance to introduce the active medicine paradigm into this field based on digital health management and intervention services for middle-aged and elderly people. On the one hand, it can significantly improve the health level and quality of life of the elderly: through continuous health monitoring and early intervention, it can prevent diseases before they occur, replace "treatment of existing diseases" with "treatment of existing diseases", and help the elderly delay functional decline and reduce the occurrence of disability and dementia. This will reduce the burden on families and medical insurance, in line with Healthy China's goal of "increasing healthy life expectancy". On the other hand, this project will explore the active health service model of the integration of traditional Chinese and Western medicine, and give full play to the "preventive treatment" of traditional Chinese medicine The synergy with the precision diagnosis and treatment of Western medicine creates a new paradigm of elderly health management. This also has exemplary value for improving China's elderly health service system and cultivating a new health industry. In general, the project closely follows the national strategic needs and the frontier of technological development, and is of great significance in theory and practice: it not only promotes the medical model from passive to active, from fragmentation to integration, but also promotes the in-depth application of artificial intelligence in the field of people's livelihood and health, providing a replicable and promotable model for the construction of an "active and healthy society".

2. Project objectives and research content

2.1 Overall Objectives of the Project

This project is aimed at the Ministry of Science and Technology's "Regional Comprehensive Application Demonstration of Active Health Digital Technology".Focusing on the major needs of digital health status management and intervention services for middle-aged and elderly people, an active medical intelligence semantic fusion system based on the DIKWP model was constructed, and comprehensive application demonstrations were carried out in communities, homes, villages and other scenarios. The overall goal is to build a digital health service system that integrates traditional Chinese and Western medicine knowledge, has cognitive intelligence and active intervention capabilities, realizes continuous monitoring, semantic understanding and active intervention of the health status of middle-aged and elderly people, and provides technical support and demonstration models for healthy aging.

Specifically, the overall objectives of the project can be broken down into the following points:

l Paradigm innovation goal: A new paradigm of "active medicine" health service is proposed at a theoretical level, and based on the DIKWP model, the medical model is upgraded from passive diagnosis and treatment to a comprehensive system of active perception, active prevention and active intervention, which meets the needs of the national health strategy. Through the practice of this project, the theoretical connotation and technical framework of active medicine will be improved, and the medical paradigm change will be led.

l Technology R&D objectives: Develop an active medical intelligence semantic fusion system, including a knowledge graph of integrated traditional Chinese and Western medicine, a cognitive agent for artificial awareness white-box evaluation, an active intervention decision-making engine, and a user-oriented digital health management platform. Realize the intelligence of the whole process of cross-modal health data collection, semantic analysis, risk assessment, and intervention plan generation, and ensure that the system decision-making is explainable, traceable, safe and controllable.

l Application demonstration objectives: Deploy demonstration applications in typical scenarios such as urban communities and rural elderly care to verify common health problems in middle-aged and elderly people (such as cataract screening, fall prevention warning, cognitive impairment intervention, etc.). Form quantifiable health improvement effects and service model experience, formulate relevant standards and specifications, and promote the promotion and application of the results on a larger scale.

l Industry-driven goal: Cultivate a new form of active health service with multidisciplinary integration, and promote the integrated development of artificial intelligence and the medical and health industry. Through the implementation of the project, a number of core technologies and products with independent intellectual property rights (such as intelligent monitoring equipment, knowledge graph platform, health intervention app, etc.) will be produced, and the formulation of standards and the establishment of industrial alliances will be promoted, so as to provide momentum for the upgrading of China's health technology industry.

2.2 Main research contents

In order to achieve the above objectives, the project has developed the following main research contents and task modules:

2.2.1 Architecture design of active medical intelligence semantic fusion system

Construct the overall project architecture and develop a technical roadmap. The architecture of this system is based on the DIKWP model proposed by Professor Yucong Duan as the core soul, and is composed of five functional modules of the cognitive layer: data layer, information layer, knowledge layer, Wisdom layer and Purpose layer, supplemented by cognitive cycle and metacognitive control mechanism, to realize the independent cognition and regulation of health status by AI. The specific design includes: the perception layer is responsible for the collection and processing of ubiquitous health data, the semantic layer is responsible for the semantic labeling and integration of multi-source data, the cognitive layer carries the knowledge reasoning and scenario understanding of traditional Chinese and Western medicine, the decision-making layer conducts comprehensive evaluation and intervention plan generation, and the purpose layer sets the overall goals of the system and monitors the output and feedback of each layer. Through five-layer collaboration and two-way feedback, a circular closed loop of "data→ information→ knowledge→ Wisdom→Purpose" is formed, so that the system has the ability of continuous learning and self-optimization. The architecture design also includes secure rights management and privacy protection solutions to ensure that personal health data is collected, transmitted, stored, and used in compliance with ethical norms and security standards.

2.2.2 Knowledge graph construction and semantic integration of traditional Chinese and Western medicine

Constructing a health semantic network that integrates the knowledge of traditional Chinese medicine and Western medicine is the basis for realizing intelligent semantic understanding. The research contents include: 1) TCM knowledge graph construction: collect and sort out the knowledge in TCM classics and clinical guidelines, such as TCM theoretical system (viscera, meridians, yin and yang five elements, etc.), TCM syndromes and symptoms, TCM and prescriptions, acupuncture and moxibustion health care, etc., and formalize it into knowledge graph nodes and relationships. 2) Construction of Western medicine knowledge graph: Construct a standardized Western medicine knowledge graph by using the existing medical knowledge base (such as disease diagnosis, anatomy, physiology, biochemical indicators, western medicine and surgery, etc.) and geriatric medicine guidelines. 3) Semantic mapping and ontology integration of traditional Chinese and Western medicine: study the correspondence between the conceptual system of traditional Chinese and Western medicine, and establish cross-system ontology mapping rules. For example, the "syndrome" of traditional Chinese medicine is associated with the "disease/syndrome" of Western medicine, the mechanism of action of traditional Chinese medicine and Western medicine is corresponded, and the meridian effect of acupuncture and moxibustion is correlated with the function of the nervous system. Through semantic bridging, a unified health knowledge representation model is formed. 4) Knowledge graph fusion algorithm: develop a semantic fusion algorithm, based on the method of DIKWP semantic mathematics, to realize the association reasoning and consistency maintenance of traditional Chinese and Western medicine knowledge at the knowledge layer. Semantic mathematics is used to formally constrain knowledge, so that the fused knowledge graph remains rigorous and reliable in terms of conceptual level and logical reasoning. 5) Knowledge update and evolution: Design a dynamic update mechanism of knowledge graph, and continuously feed new medical evidence and project clinical data back to the knowledge base through machine learning and expert verification to ensure the timeliness and applicability of knowledge.

2.2.3 Research and development of cognitive agents and active intervention mechanisms

On the basis of the knowledge graph, the research and development of AI agents with human-like cognition and initiative is the core technology of the project. Main research contents: 1) Integration of artificial consciousness white-box evaluation system: The DIKWP artificial consciousness white-box evaluation standard is applied to the development of the AI model of this project, and the ability performance of the agent at all levels of data perception, information extraction, knowledge reasoning, wisdom decision-making, and purpose guidance is tracked and evaluated throughout the process. Through white-box design, the model outputs intermediate semantic explanations (such as "link inference" and "purpose analysis" steps) during the inference process, so that developers and medical experts can understand the basis of model decision-making. 2) Cognitive reasoning engine: Develop a cognitive reasoning algorithm based on semantic mathematics to map the collected multi-source health data (physiological indicators, behavioral data, environmental data, etc.) to the semantic space for inference. For example, through formal rules, it is inferred that "an elderly person has recently had unstable gait + eye discomfort + TCM 'liver and kidney deficiency' syndrome = increased risk of cataracts", or"Decreased sleep quality + mild memory loss = early signs of Alzheimer's disease". Semantic mathematics ensures that the reasoning process remains semantically consistent and reduces logical bias. 3) Active decision-making and purpose regulation: realize the ability of agents to make active decisions according to health goals. With the help of the Purpose layer of the DIKWP model, the overall goal of the system service is set to "maintain and improve personal health and delay the decline". Based on this, the agent evaluates the contribution of various intervention programs to the goal and selects the optimal plan for implementation. For example, for the detected risk, the agent can consider the user's preference and effect to decide whether to recommend TCM conditioning (diet/acupuncture) or Western medical visit/medication intervention, or a combination of both. The Purpose layer also enables the agent to be self-driven: when data is missing or inconsistent, it can proactively obtain more information (such as prompting the user to make additional measurements or ask questions); When the intervention effect is not good, the strategy should be actively adjusted to form a closed-loop management. 4) Human-computer interaction and feedback learning: Design natural human-computer interaction interfaces (voice assistants, home companion robots, etc.) so that the agent can provide suggestions and explanations in a way that is easy for the elderly to accept, and at the same time obtain user feedback. Through reinforcement learning, feedback signals are used to optimize model parameters to realize personalized learning of individuals by agents, and the more they use it, the more they "understand" the user. Focus on ethics and safety in the interaction process to ensure that the agent's recommendations are always supervised by humans and can be discontinued or modified.

2.2.4 Digital health management platform and application demonstration

Develop a digital platform for end users and service providers to integrate and put these technologies into practice. Research contents: 1) Ubiquitous health monitoring network: build an Internet of Things health monitoring network covering homes, communities, and medical institutions. Deploy smart terminals (such as wearable bracelets, mattress sensors, smart pill boxes, environmental sensors, etc.) in elderly homes, configure health all-in-one machines, fundus screening devices and other equipment in community elderly care centers, and access 5G communication to achieve seamless monitoring near human space. Develop health data standards and interface specifications (such as personal health passport data specifications) to ensure compatibility and integration of data from different devices. 2) Active health management app/applet: Develop age-friendly mobile apps to provide the elderly and their family members with functions such as health status browsing, early warning reminders, and intervention guidance. The app interconnects with the agent's backend to push reminders when risks are detected (e.g., "recent weight loss + poor appetite, it is recommended to strengthen nutrition"), provide explanations supported by knowledge graphs, and give suggestions for actions (such as dietary adjustments, TCM health care methods, or hospital visit guidelines). At the same time, it provides health education content and online consulting services to improve users' health literacy. 3) Doctor and nursing support system: develop a management platform for community doctors, family nurses and other professionals. Functions include viewing the health risk ranking of the elderly under its jurisdiction, receiving abnormal alerts pushed by AI, reviewing AI-recommended intervention plans and deciding whether to adopt them, remote guidance, etc. The platform integrates electronic health records and knowledge graphs to assist doctors in making global decisions. Realize two-way collaboration between medical staff and AI: AI actively screens risks, provides decision-making references, and doctors review and check, execute interventions, and feedback results for model improvement. 4) Application demonstration and evaluation: select representative pilot areas to carry out application demonstration. For example, an active health management demonstration station was established in urban communities, and a certain number of middle-aged and elderly residents were recruited to use the system, which was observed for six months to one year, and the evaluation indicators included: health behavior improvement rate, chronic disease control rate, change in acute hospitalization rate, user satisfaction, etc. Another example is in remote rural areas, where remote monitoring services are provided for the left-behind elderly through simplified equipment to assess the suitability of the system in a low-resource environment. For the two typical scenarios of cataract and Alzheimer's disease, a special intervention process was developed and the effect was verified (see Chapter 4 for details). Finally, the new model and technical specifications of digital health continuous service are extracted, and a complete application demonstration report and promotion strategy are formed.

To sum up, the research content of the project covers the whole chain of theoretical methods, key technologies and application demonstrations, including top-level architecture design, specific technical module research, and practical scenario verification. The modules are closely connected: knowledge graphs and semantic algorithms provide intelligent "knowledge brains", cognitive agents endow the system with "active thinking and decision-making" capabilities, and digital platforms transform technological achievements into service capabilities. Through phased implementation and integrated testing, a fully functional active medical digital health service system will be finally realized, providing a new health security paradigm for the middle-aged and elderly population in China.

3. Technical route and innovation points

3.1 Technical roadmap

The technical route of this project follows the path of "theoretical guidance-technical research-integrated application", and adopts the method of iterative development and verification to gradually achieve the system goal. The overall technical roadmap is shown in the figure (the project technical roadmap can be inserted here if necessary), and is mainly divided into the following parallel and cross-interactive routes:

Route 1: Build a DIKWP intelligent semantic framework. Based on the DIKWP model, a systematic semantic cognitive framework is first constructed. This includes defining the five-layer cognitive function modules and their interfaces, and establishing the mapping mechanism of the four cognitive spaces (conceptual space, cognitive space, semantic space, and consciousness space). Through simulation, it is verified that each layer module can produce logical explanations and responses to simple health cases (such as fluctuations in a single indicator). Ensure that the framework design is reasonable, and provide standards for the subsequent development of each technical module.

Route 2: Multi-source health data collection and processing. Develop various data acquisition methods and pre-processing algorithms required for the project. Including the selection and deployment of wearable monitoring devices, the construction of home sensor networks, and the development of data interfaces for medical information systems. Develop data cleaning and feature extraction processes for different data sources, and transform raw data into structured information into the data layer of the DIKWP framework. At the same time, a master index of personal health is established, and scattered data is aggregated by user to form a "personal health digital portrait".

Route 3: Knowledge Graph and Semantic Reasoning R&D. In parallel, the construction of knowledge graph of traditional Chinese and Western medicine was promoted, and the semi-automatic method (NLP text extraction + expert verification) was used to complete the initial graph. Then, based on the theory of semantic mathematics, a knowledge fusion and reasoning engine is developed to realize the semantic interpretation of personal health data at the information layer and the knowledge layer. This route requires constant interaction with route two: data processing provides new health facts, and the knowledge engine performs inference validation, which in turn guides data collection (e.g., if a new indicator is needed to be monitored).

Route 4: Cognitive agent development and evaluation. With knowledge graph and semantic reasoning as the "center", the active AI agent is developed. Modular design: perception module, inference module, decision-making module and purpose management module. The artificial consciousness white-box evaluation method was introduced, a series of cognitive tasks were designed to test the agent's ability, and the model structure was iteratively optimized. At the same time, the agent is allowed to access real user data for trial operation in a small area to find problems (such as whether the decision is robust and whether the explanation is easy to understand), and then improve the algorithm. This route requires collaboration with Route 5 in terms of human-computer interaction.

Route 5: Platform integration and application validation. Integrate the above results into the digital health management platform, including front-end apps, back-end databases, doctor-side interfaces, etc. Firstly, the interfaces of each module are jointly debugged in the experimental environment to ensure the smooth flow of data and decision-making. Subsequently, pilot users were selected to carry out a closed test: the platform pushed intervention suggestions, manually recorded feedback, and compared the changes in health indicators with or without AI intervention. Adjust platform functionality and interaction based on test results. For example, if it is found that elderly users are not very good at operating apps, add voice interaction or assistance from grassroots personnel. Finally, open application was carried out at the demonstration site, and actual effect data were collected for evaluation.

In the implementation process, the above routes are not isolated, but spirally intertwined: the improvement of the quality of the knowledge graph will enhance the accuracy of the agent's decision-making, and the new knowledge fed back by the agent will feed back to the knowledge base; The expanded scope of data collection has enabled more scenarios to be covered, and new requirements found in the application of the platform have driven improvements in sensing devices. Milestone node control will be adopted in project management, and each stage will enter the next stage after completing the milestone acceptance to ensure that the technical route is carried out as planned.

3.2 Key technological innovations

This project integrates multidisciplinary cutting-edge theories and applied innovations, and is expected to achieve breakthroughs in the following key technologies:

(1) Innovation of semantic integration of active medicine: Breaking through the limitations of the existing health management based on a single system of Western medicine or traditional Chinese medicine, this paper proposes an active medical framework for the semantic integration of traditional Chinese and Western medicine. Through DIKWP semantic mathematics, artificial intelligence is given a formal expression of concepts such as "health" and "disease-free", and the medical paradigm is upgraded at the philosophical level. This innovation elevates medical knowledge from the empirical layer to the semantic layer, and uses mathematical and computational language to uniformly represent the "syndrome" and "treatment" of traditional Chinese medicine and the "disease" and "medicine" of Western medicine to fill the gap that there is no deep integration model of traditional Chinese and Western medicine knowledge at home and abroad.

(2) DIKWP Artificial Awareness White-box Evaluation Application: For the first time, the latest DIKWP white-box evaluation method in the field of artificial intelligence is applied to the development of health service systems to build an interpretable cognitive health AI。 By dividing the AI decision-making process into five parts: data, information, knowledge, wisdom, and purpose, the whole process of the agent's decision-making mechanism can be traced, which greatly improves the credibility and security of the system. Compared with the traditional black-box medical AI model, this system can clearly understand the reasoning behind each health recommendation and establish a trust foundation for human-machine collaboration.

(3) Active AI autonomous intervention technology: A new implementation method of Active AI in the field of medical and health care is proposed. Traditional health management AI is mostly passive query or preset rule reminder by users, but this project uses purpose-driven cognitive agents to enable AI to have the ability to independently find problems and take action. This proactivity is reflected in the fact that AI no longer waits for users to ask questions, but continuously monitors the user's status and actively identifies potential health risks; AI is not limited to a single intervention, but can independently choose a combination of multiple interventions such as traditional Chinese medicine, lifestyle guidance or medical services. AI can also track the effectiveness of interventions after they have been executed, and automatically adjust the plan if necessary. This kind of self-purposeful AI application is the first of its kind in the field of health management in China, and will lead the transformation of "from passive medical care to active health".

(4) Knowledge graph and semantic reasoning of integrated traditional Chinese and Western medicine: develop medical knowledge graph technology across semantic space to realize unified reasoning of complex heterogeneous knowledge. These include: semantic extraction and alignment techniques for multi-source medical knowledge, fusion methods of fuzzy logic in traditional Chinese medicine and quantitative rules in western medicine, and hybrid reasoning engines that take into account probability and logic. The innovation lies in the introduction of the axiom of semantic consistency to ensure that the reasoning process does not deviate from the real clinical meaning. At the same time, the DIKWP hierarchical framework is used to divide reasoning into different spaces of concept, cognition, semantics, and consciousness for parallel processing, so as to improve the ability to solve complex problems of reasoning. These explorations will enrich the knowledge graph and methodology in the field of cognitive computing, and promote the evolution of medical artificial intelligence from purely data-driven to knowledge-driven.

(5) Ubiquitous health monitoring system for the elderly: At the application level, the ubiquitous health Internet of Things in the elderly care scenario is innovatively constructed. The project will integrate Wisdom building, smart home and wearable device technologies, and take the lead in creating a "healthy building embedded healthcare" solution in China. For example, we will integrate floor pressure sensing for fall monitoring, smart lighting for mood and sleep intervention, air quality sensing for environmental health assessment in nursing homes, etc., to develop a smart health environment suitable for home/nursing homes. All devices are interconnected with the AI platform in real time to realize the active health intervention of the environment-human-AI trinity. The innovation of the system lies in truly "people sit at home, health comes from heaven", and extend health care services to the daily life space of the elderly.

(6) Data standards and service model innovation: The project will develop personal health data normative standards (similar to health passport standards) and semantic interoperability standards to fill the gap in the field of active health standards. At the same time, we explore a new health supply model of "model as a service", and continuously provide multi-party collaborative services for individuals through third-party digital platforms. This model connects the service capabilities of different institutions (hospitals, pharmacies, health management companies, and elderly care institutions) through the platform, and AI matches and supplies them according to the model decision-making results, creating a new ecology of digital health services. This model of multi-party collaboration and on-demand supply is of innovative significance for solving the current problem of fragmentation and mismatch between supply and demand between medical and elderly care services.

In summary, this project has significant innovations in theory, technology and application. Whether it is the improvement of the academic theory of the active medical paradigm, the implementation of artificial consciousness white-box technology in the field of health, and the various new technologies and new models of cross-border integration, they are all cutting-edge explorations. The successful implementation of the project is expected to form original achievements with independent intellectual property rights, and establish China's international leading position in the field of active health/active medicine.

4. Application Scenarios

The research results of this project will focus on the application of digital health management and intervention for middle-aged and elderly people, and select three types of scenarios for demonstration: community, home and rural areas. Among them, each scenario selects common and representative health problems of the elderly to simulate the service process under the paradigm of "active medicine" in the future. Typical application scenarios are described below:

4.1 Active screening and intervention scenarios for cataract in the community

Scenario Overview: An urban community elderly care service center introduces an active health management system of this project to provide regular health monitoring and intervention services for elderly residents in the jurisdiction. Cataract is a common blinding eye disease in the elderly, and many elderly people in the traditional mode delay diagnosis and treatment due to lack of symptoms. With the support of the active medicine system, the community carried out active cataract screening and intervention.

Participating Characters: Uncle Zhang, a 70-year-old man, retired at home; Dr Lee, a community general practitioner; AI health management assistant "Xiaoyuan" (the incarnation of the project's agent, which provides services through voice interaction devices).

Scenario flow:

l Daily monitoring and risk detection: Uncle Zhang wears smart glasses-type devices every day, which can monitor his visual clarity, reading distance and other indicators; The TV set-top box at home is equipped with vision AI software to analyze his gaze and expression when watching TV. The system collects this data continuously and analyzes it weekly by the AI assistant "Xiaoyuan". In a certain month, AI found that Uncle Zhang's reading distance had become significantly closer recently, and he frequently squinted when watching TV, combined with the slow downward trend of his vision examination within half a year, it was judged that he had early signs of cataract and an increased risk score. The AI sent this information to Dr. Li through the community platform, and took the initiative to remind Uncle Zhang on the smart speaker at home: "Recently, it has been detected that your visual difficulty may increase, and it is recommended to undergo cataract screening." After receiving the voice reminder, Uncle Zhang agreed to be arranged by the system.

l Door-to-door screening and semantic analysis: The next day, the community health service station sent a mobile screening vehicle to Uncle Zhang's community. The screening vehicle is equipped with a portable digital slit lamp and fundus camera, and is connected to the AI platform. During the examination, the AI assistant guides Uncle Zhang to complete the eye chart test and eye photography, and uploads the data in real time. The AI first invokes the knowledge graph of Western medicine for image analysis to identify the degree of lens opacity. At the same time, combined with the knowledge of traditional Chinese medicine (such as asking Uncle Zhang if he has symptoms of dry eyes, liver and kidney insufficiency), the semantic judgment of the combination of traditional Chinese and Western medicine is carried out. The AI reasoning results showed: "Very mild opacity behind the lens of the left eye, and the visual acuity decreased from 0.6 to 0.4, which is an early senile cataract, and traditional Chinese medicine belongs to the category of 'dizziness', and the possibility of liver and kidney deficiency is high." The system fed back the conclusions and evidence to Dr. Li and Uncle Zhang in the form of natural language and diagrams. After Dr. Li verified the AI analysis, he confirmed that Uncle Zhang had early cataracts.

l Active intervention plan formulation: In response to this situation, the AI assistant generates a personalized comprehensive intervention plan in the background: in Western medicine, it is recommended to undergo elective surgery within six months and use nutritional lens drugs before surgery to slow down the progression; In terms of traditional Chinese medicine, it is recommended to carry out eye exercises and acupressure every day (focusing on pressing the eyes, temples, etc.), and oral Chinese medicine Mingmu decoction to regulate the liver and kidneys; In terms of daily life, it is recommended to wear UV protection glasses outdoors and increase the intake of foods rich in vitamin A. The AI informed Uncle Zhang of the key points of the plan through the voice of "Xiao Yuan", and explained: "Your cataract is still very early, and it is expected to delay the development through the intervention of integrated traditional Chinese and Western medicine. We recommend ......(specific scheme)". At the same time, the plan was pushed to Dr. Li, and after being approved by Dr. Li, it was entered into the system as a formal health intervention plan.

l Intervention execution and tracking: In the following months, the AI assistant "Xiao Yuan" reminded Uncle Zhang to drop eye drops according to the doctor's instructions, instructed him to perform eye acupressure massage, and pushed weekly recipe suggestions (such as carrot scrambled eggs and other eye-friendly foods). Uncle Zhang feedback "completed" after each execution or the family members record the completion on the App, and the system tracks compliance accordingly. If Uncle Zhang forgets it one day, the system will patiently remind him; If it is not completed many times, the community doctor will be notified for follow-up care. Three months later, the AI arranged a re-examination of visual acuity and updated the risk assessment, and found that the visual acuity was basically stable at 0.4, and the intervention was initially effective. At this time, the AI offered to postpone the surgery plan appropriately and continue to observe. At the same time, the system will send the phased results to the WeChat of Uncle Zhang's children, so that the family can understand the health status of the elderly.

l Multi-party linkage and dynamic adjustment: During the continuous service process, community doctors regularly review the AI white-box evaluation report, which details what conclusions the system has drawn based on what data and knowledge. For example, the report shows that "this month, the probability of cataract determination based on a 0.1 decrease in visual acuity, a 20-centimeter reduction in reading distance, and a 10% increase in the probability of cataract judgment based on knowledge base rule R45" enables doctors to fully grasp the AI logic. If Dr. Li finds that Uncle Zhang has other eye diseases (such as high risk of glaucoma), he can adjust the focus of intervention through the system, and the AI will quickly learn the doctor's new instructions and incorporate them into subsequent decision-making. In the whole process, community services, home care and AI assistants form a human-machine collaborative network - AI is responsible for continuous monitoring, intelligent analysis and active reminders, doctors are responsible for diagnosis and confirmation and key decision-making, and the elderly are responsible for cooperating actions and feedback feelings, and all links are closely connected. After a year of demonstration, Uncle Zhang's cataract has not progressed significantly, and his quality of life is good. At the community level, due to the active screening intervention, the incidence of cataract in the middle and late stages of the elderly in the jurisdiction was reduced, and unnecessary loss of function was avoided. This scene shows the value of the paradigm of active medicine in the prevention and treatment of blindness in the community: diseases are not treated until blindness is lost, but are caught and controlled at the beginning of the process.

4.2 Home Alzheimer's disease intelligent monitoring and intervention scenarios

Scene Overview: In the home of an empty nest elderly couple, their wife, Aunt Wang (68), has recently become forgetful. The project's intelligent health management system continues to serve them as a "digital family doctor". For elderly cognitive impairments such as Alzheimer's disease (AD), traditional passive medical care often waits until patients have obvious memory loss before starting intervention, missing the early window. Active medicine systems are expected to enable smart home monitoring, early identification, and cognitive intervention to slow down the disease process.

Roles: Aunt Wang, husband Uncle Li; AI cognitive health assistant "Xiaozhi"; Community GP/Psychiatrist.

Scenario flow:

Daily behavior monitoring: Smart home sensors and voice assistant "Xiaozhi" are installed in Aunt Wang's home. The system observes her cognitive behavior every day in a multimodal way: for example, the kitchen sensor records her cooking steps, the smart medicine box records whether she takes her medication on time, the voice assistant chats with her from time to time, and the memory and calculation skills are assessed through dialogue. After a while, the AI noticed several anomalies: Aunt Wang forgot to take her high blood pressure medication twice in a week; The kettle was dry boiled several times, showing that she forgot to turn off the heat; During the conversation test with "Ash", she sometimes couldn't answer the previous day's activities. These fragmented signals were comprehensively analyzed through the knowledge graph: AI identified this red flag as mild cognitive impairment (MCI) from Western medical knowledge, and corresponded to the "forgetfulness (brain and kidney insufficiency)" syndrome from the perspective of traditional Chinese medicine. The system derives a preliminary risk assessment: Aunt Wang may be in the early stage of AD. The AI then triggered the intervention process, notifying Uncle Li and his children to pay attention, and on the other hand, reporting the situation to the community doctor and recommending a professional cognitive assessment.

Professional assessment and diagnosis: After receiving the notification, the community doctor came to conduct a simple mental state examination (MMSE, etc.) for Aunt Wang, and the result score was slightly lower than normal. The doctor entered the examination results into the system, and the system, combined with the previous family monitoring data, characterized Aunt Wang's status as "suspected MCI, possible early Alzheimer's disease". Diagnosis by a psychiatrist and development of an intervention plan by a psychiatrist is recommended. Soon, under the coordination of the community, Aunt Wang went to the memory clinic for an MRI examination and neuropsychological evaluation, ruled out other causes, and was diagnosed with early AD.

Active intervention plan: The AI assistant "Xiaozhi" formulated a personalized cognitive intervention plan for Aunt Wang based on multidisciplinary knowledge: in terms of medicine, combined with the cholinesterase inhibitor prescribed by the doctor, the system reminds her to take the medicine on time; In traditional Chinese medicine, nootropic Chinese medicine prescriptions (such as gastrodia, calamus, etc.) are distributed by the community traditional Chinese medicine center, and they are decocted daily; In terms of cognitive training, the system arranges certain mental exercises (such as recall games, trivia quizzes) every day and chats with her; In terms of lifestyle, it is recommended to increase exercise (walking with you every night and counting steps), social activities (participating in community dance activities every week), etc. In particular, AI has designed a series of thoughtful measures to assist memory: for example, the smart screen combined with the refrigerator magnet lists the to-do items of the day every day, and repeatedly reminds important items with voice reminders; For Aunt Wang, who often forgets things, the AI provides an instant query function - she only needs to ask "Ash, have I taken any medicine?" The AI can answer based on the sensor recording. Through these measures, AI has played the role of "digital caregiver" and "cognitive trainer" to some extent.

Multi-source feedback and dynamic adjustment: During the intervention, the system continued to collect feedback data from Aunt Wang in all aspects: the smart bracelet monitored whether her exercise volume met the standard; The voice assistant assesses whether her correct rate of answering questions has improved; Uncle Li can also report her daily memory status through the app. The AI summarizes the effect of the intervention once a month and reports to doctors and families: "This month, Aunt Wang walked an average of 5,000 steps a day, an increase of 20% from the previous month; 95% compliance rate of drugs and traditional Chinese medicines; There was a slight improvement in cognitive test scores, and the overall trend of memory decline slowed. "If some interventions are found to be not working well, the AI will suggest adjustments. For example, if the side effects of the drug are obvious, you can ask your doctor to adjust the dose; If she doesn't exercise enough, the AI will encourage her to go out more often or pair up with volunteers to accompany her to exercise.

Humanistic care and long-term companionship: In addition to technical interventions, the system pays great attention to humanistic care. AI "Xiaozhi" talks with Aunt Wang every day, chatting about the opera she is interested in or interesting stories about her children to relieve her emotions; Observe her expression through the camera, and if depression is detected, comfort and notify the family in time. As a caregiver, Uncle Li also received guidance through the system (such as how to deal with some of his wife's memory loss behaviors) and received respite service referrals in a timely manner to reduce his stress. In this way, the AI system not only coldly issues tasks, but also acts as a "close friend" in the lives of elderly couples and an assistant for home care.

Scene Effect: After systematic active intervention, Aunt Wang's cognitive function decline rate slowed down significantly, and she was still able to take care of her daily life basically after two years. Compared with no intervention, delaying the course of the disease buys valuable time for both the family and society. This scenario verifies the value of proactive medicine in chronic disease management: through early intervention, comprehensive measures, and full accompaniment, there are new ways to deal with such a thorny problem as Alzheimer's disease. In particular, AI guards the home environment 24 hours a day, making up for the gap that traditional medical services are difficult to cover family life in a timely manner, and reflects the continuity and depth of active health services.

4.3 Rural health care service scenarios

Scenario Overview: In remote rural areas, medical resources are scarce, and the children of the elderly often go out to work, and the health management of the left-behind elderly has become a problem. This project selects a rural elderly care center in a county as a pilot to deploy a rural health care active health service system. The goal is to provide proactive health services integrating prevention, medical treatment and rehabilitation for the elderly in rural areas through digital technology and AI support at the grassroots level, and narrow the gap between urban and rural health services.

Scenario elements: The pilot elderly care center has admitted dozens of elderly or disabled people, who are cared for by a small number of caregivers, and there is a lack of professional medical personnel stationed on weekdays. The project team equipped the center with telemedicine equipment and AI health management terminals, and established a collaboration mechanism with the county hospital and the county hospital of traditional Chinese medicine.

Service Process:

l Health filing and risk screening: The system establishes an electronic health record for each elderly resident, including basic information, past medical history, and TCM physical assessment results. Then, AI is used to conduct initial risk screening: for example, based on the medical history and lifestyle habits of the elderly, the risk of common chronic diseases and geriatric syndromes such as falls, malnutrition, and depression is assessed. The results showed that, for example, there were 5 elderly people at high risk of diabetes and 3 elderly people with a tendency to cognitive decline. AI generates a "health risk list" and priorities for each elderly person and hands them over to the nursing center administrator.

l Individualized active intervention: According to the risk list of each elderly, a personalized intervention plan is systematically formulated. For example, Uncle Zhang, who is at high risk of diabetes, arranges daily dietary AI monitoring (food intake and blood sugar records) and exercise guidance to prevent diabetes in advance; Granny Li, who had cognitive decline, referred to the above-mentioned AD intervention program for memory training and traditional Chinese medicine conditioning; Aunt Chen, who has limited mobility, focuses on the arrangement of fall prevention measures (install smart anti-fall sensors in the room and corridor, once she gets up and leaves the bed at night, the system automatically lights up the night light and monitors the gait, and if there is a fall, the alarm will be immediately called). The integration of traditional Chinese and Western medicine is one of the highlights of the program: for example, many elderly people have chronic stomach diseases, and they are systematically arranged to do TCM diet therapy (yam, millet porridge, etc.) every day and take probiotics and other health products recommended by Western medicine to jointly improve gastrointestinal function.

l Collaboration between telemedicine and AI: When the elderly suddenly feel unwell or need regular medical assessment, the center caregiver can initiate remote consultation through the system. For example, if an elderly person with high blood pressure has high blood pressure and dizziness symptoms for several days under AI monitoring, the AI reminds him that he needs to adjust his treatment. Caregivers connect with doctors at county hospitals through remote diagnosis and treatment devices, and the doctors review the recent blood pressure curves and medication records compiled by the AI, and listen to the AI's explanation of the white-box analysis (for example, the AI points out that "the recent sudden drop in temperature + the patient's fatigue causes blood pressure fluctuations"). Doctors adjust the antihypertensive drug regimen accordingly. The AI incorporated the new protocol into the intervention plan, and then monitored the elderly's blood pressure more closely. In addition, if the elderly have depression, the AI will first provide psychological counseling through heart-to-heart talks, but if there is no improvement, it is recommended to arrange remote psychological counseling and have professional doctors intervene. This reflects a process of "AI first + manual relay": AI discovers problems in a timely manner, provides initial responses, and introduces human experts for in-depth processing when necessary, after which AI continues to implement expert opinions for follow-up management.

l Group health management and feedback: The system summarizes the health data of the elderly in the center every day and generates a group health report. Managers of elderly care centers can intuitively see how many abnormal events there are on the day (such as fever of XX, fall prevention alarm triggering, etc.), and the trend of overall indicators (such as average blood pressure and sleep quality). Through this data, managers can adjust the daily management of nursing homes, such as work and rest, meal arrangements, etc. For example, if the report shows that the number of people suffering from insomnia has increased recently, consider adjusting your evening activities or creating a quiet environment. The knowledge map of the system can also provide suggestions according to regional characteristics: the elderly with arthritis may increase due to the humid rainy season in the south, and the nursing staff should be prompted to wear knee pads and decoction to remove dampness for them in advance. This kind of data-driven refined management has greatly improved the quality of rural elderly care services. The county government health department can also use these data to assess the health status of the grassroots and provide a basis for policy formulation.

Application significance: The demonstration of rural health care scenarios proves that active medical digital technology can break through geographical restrictions, deliver high-quality health services to the elderly in rural areas, and realize the "integration of urban and rural health services".。 Through AI's active monitoring and intervention, supplemented by telemedicine support, the elderly can enjoy continuous care even in rural areas where there is a shortage of doctors. In terms of families, the system can regularly push the parents' health status to their children on the go to alleviate their worries. In terms of society, the emergency hospitalization rate and complication rate of the elderly in the demonstration area have decreased significantly, which has improved the quality of life in their later years. This scenario explores a new path for health care in underdeveloped areas: taking active health technology as the starting point, integrating limited resources, and leveraging the innovation of health service models. As some experts have pointed out, the "active health service system" relies on technology to achieve continuous and dynamic health information collection and all-round intervention, and the rural practice of this project is a vivid illustration of this.

Through the simulation and demonstration of the above three typical application scenarios, we have verified the feasibility and effectiveness of the technical system of this project in practice. Whether it is the precise prevention and control of a single disease (cataract), the comprehensive management of chronic diseases (Alzheimer's disease), or the overall improvement of elderly care services (rural health care), the intelligent semantic fusion system of active medicine has shown the advantages of actively discovering problems, providing services, and optimizing effects. Next, these scenario experiences will guide us to further improve the system's functionality and ease of use, laying the foundation for larger-scale adoption.

5. Research team and organizational structure

5.1 Project Leader and Core Team

Project Leader: Prof. Yucong Duan – Professor, School of Computer Science and Technology, Active Medicine Committee of the World Artificial Consciousness Association, Doctoral Supervisor, Academician of the International Academy of Advanced Technology and Engineering, Corresponding Academician of the National Academy of Artificial Intelligence, President of the World Artificial Consciousness Association. Professor Duan has been engaged in the research of cognitive computing and artificial intelligence for a long time, and has proposed the landmark DIKWP artificial intelligence cognitive model, which has made outstanding contributions in the fields of artificial consciousness and knowledge engineering. As the first inventor, he has been authorized 114 domestic and foreign invention patents (including 15 PCT international patents), covering cutting-edge directions such as large model training, artificial awareness, cognitive operating system, and AI governance. Professor Duan also has in-depth research in the intersection of artificial intelligence and healthcare, advocates the concept of "active medicine" and chairs a number of related projects, and is the initiator and academic leader of this project. As the PI of the project, he will lead the overall research program design, key technology research and teamwork to ensure that the project implementation is scientific and innovative.

Core team members and division of labor: The project team is led by the Active Medicine Committee of the World Association of Artificial Consciousness, and a multidisciplinary consortium is formed by a number of advantageous scientific research institutes and enterprises in China, covering experts in artificial intelligence, computer software, medicine (Western medicine + Chinese medicine), public health, electronic engineering and other fields. The main core members include:

l Artificial Intelligence: Prof. Yucong Duan (Active Medicine Committee of the World Association of Artificial Consciousness), member of the DIKWP International Standards Committee for Artificial Intelligence, engaged in semantic intelligence and knowledge graph research, responsible for the development of knowledge graph and semantic mathematical algorithms for this project; Dr. Y (CTO of an AI company) has experience in cognitive intelligence product development, and is responsible for AI platform engineering implementation and system integration.

l Medical direction (Western medicine): Professor Z (geriatric expert of a medical university), good at geriatric and chronic disease management, responsible for the review of medical knowledge map content and the formulation of clinical intervention programs; Q Chief Physician (Department of Geriatrics, Tertiary Hospital), served as the clinical consultant of the project, guiding the rationality of the intervention of integrated Chinese and Western medicine in the scene and participating in the evaluation of the demonstration effect.

l Medical direction (Chinese medicine): Professor W (a Chinese medicine university), proficient in Chinese medicine knowledge mining and informatization, responsible for the construction of Chinese medicine knowledge graph and the mapping of Chinese medicine semantics and Western medicine ontology; Chief TCM Physician (TCM Hospital) is responsible for integrating TCM experience in treating pre-existing diseases into the intervention program and guiding the evaluation of TCM intervention effects.

l Public Health and Elderly Care Direction: L Researcher (Public Health Institute), specializing in health management and policy, responsible for the design of community and rural demonstration site implementation plans, data collection and social benefit assessment; Director M (elderly care service organization), rich experience in the management of elderly care institutions, responsible for the implementation of the project in the nursing home scene and user demand feedback.

l Engineering and equipment direction: Engineer D (Internet of Things company), good at wearable and sensor development, responsible for health monitoring equipment selection and software and hardware integration; Engineer H (software company) is responsible for the development of age-friendly application interfaces and human-computer interaction systems.

Among the team members, there are more than 10 experts with senior professional titles and many young backbones, forming a talent echelon combining the old, middle and young. Prof. Yucong Duan leads the AI and overall integration, medical experts ensure professional accuracy, engineering staff are responsible for system implementation, and all exhibits work closely together. The team will also recruit postdoctoral fellows and graduate students to participate in sub-project research and strengthen its strength.

Consultants and external experts: The project hires international experts from the World Association for Artificial Awareness and the DIKWP International Standards Committee as consultants, such as AI ethics and medical informatics experts from the United States and Europe, to provide guidance on the project's key decisions and technical roadmaps to ensure that the research is in line with international frontiers. At the same time, relevant experts from the Department of Science and Technology of the State Administration of Traditional Chinese Medicine and the Department of Aging and Health of the National Health Commission are invited to provide regular guidance to conform to the development direction and policy norms of the industry.

5.2 Composition and organization of multidisciplinary consortiums

The project is organized by the active medicine committee of the World Association of Artificial Consciousness as the lead unit, and is jointly implemented by several collaborating units. The units of the consortium have a clear division of labor and are closely coordinated through effective mechanisms:

Lead unit: Active Medicine Committee of the World Association for Artificial Awareness – The Project Management Office is located in the lead unit and is responsible for project coordination, progress monitoring and fund management. The lead unit undertakes core technology research and development (DIKWP model application, AI platform development) and integration tasks, and is also the leading unit for intellectual property management and standard formulation.

Collaborator 1: A medical university in Beijing – responsible for the medical knowledge map and clinical program of Western medicine. The unit has strong geriatric research strength, providing clinical data support and hospital demonstration sites to ensure the evidence-based nature of the program.

Collaborator 2: A University of Traditional Chinese Medicine in Shanghai – responsible for the research of TCM knowledge graph and semantic integration of TCM. The TCM Big Data Research Center of the unit will provide a huge amount of TCM literature and case resources for the project, and set up TCM intervention demonstration sites in local TCM hospitals.

Collaborator 3: A well-known domestic AI company (such as DeepSeek) – The company has experience in AI productization and computing resources. Responsible for optimizing research algorithms into industrial-grade software, providing cloud service support, and assisting in the deployment of application platforms. Provide guarantees in terms of industrialization and data security.

Collaborator 4: Local elderly care service institutions (subordinate to the civil affairs/health departments in the pilot areas) – responsible for organizing community and nursing home demonstration applications, coordinating the recruitment, on-site implementation and data collection of the elderly. They will also suggest improvements to the system from a user experience perspective.

In addition, there will be several participating units as needed, such as a sensing equipment manufacturer (responsible for hardware provision), a telecom operator (responsible for communication and remote platform support), etc. These partners share responsibilities and resources by signing cooperation agreements.

Organizational structure and management mechanism: The project establishes a project implementation committee composed of the leaders of the leading unit and the project leaders of each unit, and holds regular meetings (quarterly) to report progress and coordinate issues. The committee consists of special working groups: (1) Theory and Algorithm Group (lead: Active Medicine Committee of the World Association of Artificial Consciousness, members: AI companies and university researchers), (2) Medical Knowledge and Application Group (lead: Medical University, University of Chinese Medicine), (3) Integration Demonstration Group (lead: AI enterprises, elderly care institutions). Each group has 1 team leader and 1 deputy team leader, who are responsible for the daily work of the group.

The implementation of the project will implement milestone management and responsibility to people. For each research task, the time nodes and deliverables are refined, such as completing the first version of the knowledge graph in the first year, completing the agent prototype in the second year, etc., and assigning responsibilities to specific units and responsible persons. The PMO collects the progress of each sub-task every month and prepares a monthly briefing to inform the whole. For cross-unit tasks, a joint research model is adopted, for example, the knowledge graph is jointly built by two teams of traditional Chinese medicine + western medicine, and the data is regularly connected and synchronized under the guidance of the PMO.

Communication mechanism: Use a combination of online and offline communication tools. Hold a regular video meeting every two weeks to solve technical problems; A plenary offline seminar will be held every six months to summarize the stages. Build a knowledge management platform within the project, share R&D documents, data sets and stage results, and promote information flow. Pay attention to the enthusiasm of young scientific researchers, and encourage graduate students from all units to establish academic exchange WeChat groups, joint experiments, etc., to learn from each other and cooperate.

Quality and Risk Control: The project has set up a quality assurance and ethics team, which is composed of personnel appointed by each main research unit, and Professor Yucong Duan is the team leader. The team formulates technical quality standards, data quality standards, and regularly checks the quality of the results of each module. In particular, for medical data and human experiments, we strictly abide by ethical requirements, formulate informed consent processes and data privacy protection plans, and supervise the implementation by the ethics team. Once a problem is found, it will be warned and rectified in time. For possible risks, such as the technical route not meeting expectations, the effect of small-scale demonstration is not obvious, etc., the project team has formulated alternative plans or adjustment plans in advance, and reserved a certain amount of mobile funds and manpower to cope with unexpected challenges.

In short, the project team has a strong lineup, reasonable structure, and sound organization and management mechanism. Under the leadership of Professor Yucong Duan and the full cooperation of all participating units, we are confident that we will complete the research tasks on time and with high quality and achieve the expected goals.

6. Implement the schedule

The project is planned to be implemented for a period of 5 years (from the second half of 2025 to the first half of 2030), and will be divided into three phases to gradually advance considering the difficulty of the research and the requirements of demonstration applications, with the key points and milestones of each stage as follows:

6.1 Phase 1 (Project Beginning, July 2025 – December 2026): Theoretical and basic platform development

Goal: To lay a solid foundation theoretical and technical framework and develop a prototype system. Key milestones include:

l M1: Theoretical framework and overall scheme determination (2025 Q4) – Complete the overall architecture design of the active medical intelligence semantic fusion system, and clarify the interface specifications of each subsystem. Submit the White Paper on the Overall Technical Scheme and Theoretical Framework of the Project.

l M2: Preliminary Construction of Knowledge Graph (2026 Q2) – The prototype of the knowledge graph of TCM and Western medicine has been completed respectively, covering no less than 1,000 TCM concept nodes and 1,000 Western medicine concept nodes, and achieving initial integration (more than 100 pairs of TCM and Western medicine concept mapping). Released the "Health Knowledge Graph of Traditional Chinese and Western Medicine v1.0" and formed a draft data standard.

l M3: Cognitive Agent Prototype (2026 Q3) – The prototype of the AI cognitive engine based on the DIKWP model has been developed, which can answer 100 questions in a white box test question bank in a laboratory environment. Through the verification of the 100-question version of the "Knowing Quotient" white-box evaluation, the performance of each cognitive layer meets the standard. Output the "AI Cognitive Agent Prototype Evaluation Report".

l M4: Platform prototype and small-scale trial (2026 Q4) – Completed the basic framework and app prototype of the health management platform to realize data connection with some monitoring devices. 20 elderly volunteers from the school or partner hospitals were selected for closed testing to verify the functional closed-loop. Collect user feedback and form a "Prototype System Test Report".

Phase Summary Meeting (December 2026) – Evaluate the progress of theoretical research and prototype results, and check for deviations against the plan. Through the review of the expert meeting, it is determined to move to the next stage.

6.2 Phase 2 (Medium-term Study, January 2027 – December 2028): Improvement and integration of key technologies

Goal: Overcome key technical problems, complete system integration and mid-term achievement verification. Milestones:

l M5: Semantic Fusion Knowledge Graph Improvement (2027 Q2) – Improve the knowledge graph to cover the main health areas of the elderly. At least: knowledge of traditional Chinese and Western medicine for common chronic diseases (hypertension, diabetes, cognitive impairment, etc.), general medicine and health care knowledge, a total of 5,000+ nodes and 20,000+ relationships. A unified ontology is constructed, and the semantic consistency verification is passed. The dataset of "Fusion Knowledge Graph v2.0" was released and opened for trial in the industry.

l M6: Intelligent Semantic Reasoning and Intervention Decision Module (2027 Q4) – Completed the beta version of the Semantic Reasoning Engine and Decision Engine. Realize the automatic reasoning and recommendation generation of the system for at least 10 typical health scenarios (including rules for cataract, Alzheimer's disease and other scenarios). The engine is tested by simulated data and expert knowledge base, and its reasoning results are > 85% consistent with expert conclusions.

l M7: Platform and device integration and debugging (2028 Q1) – Integrate no less than 15 kinds of wearable devices, home sensors, and portable physical examination devices to complete platform data access and integration. The interface of each module is successfully debugged, and the system realizes the closed-loop operation of data flow from collection→ analysis→ intervention → feedback.

l M8: Trial operation of demonstration sites (2028 Q2-Q3) – Establish pilot demonstrations in 2 urban communities and 1 rural elderly care center, and recruit no less than 100 elderly users to participate in the trial operation for a period of 6 months. Monitor system stability and service effectiveness: If there is no major failure, there is a trend of improvement in user health indicators. Collect real-world data for model optimization. Form the "Mid-term Demonstration Operation Report".

l M9: Interim Results Publication (2028 Q4) – Writing papers and patents based on interim results. It is planned to publish at least 5 SCI/EI papers (covering topics such as semantic mathematics and the application of white-box evaluation in the field of health), and apply for no less than 5 invention patents (involving knowledge graph fusion, active intervention algorithms, smart devices, etc.). At the same time, a project technical exchange meeting will be held to show the mid-term results to the industry.

Phased evaluation (late 2028) – Authorities and domain experts are invited to conduct a mid-term assessment of the progress of the project. If individual indicators do not meet expectations, a rectification plan and resource allocation plan are proposed. After the evaluation is passed, it will move on to the final demonstration stage.

6.3 Phase 3 (Integrated Demonstration, January 2029 – June 2030): Comprehensive Application Demonstration and Achievement Condensation

Goal: Complete large-scale demonstration applications and condense mature products and standard achievements. Key Milestones:

l M10: Expand demonstration applications (2029 Q1-Q4) – Expand the system to more regions to carry out demonstrations, with the goal of covering no less than 5 demonstration areas: including 2 in various types of communities in the eastern, central and western regions, and 2 in rural areas, with a total of more than 1,000 elderly people being demonstrated. Monitor the service process data and improve the service model. Typical cases and evaluation reports are formed in each demonstration area.

l M11: Improve system functions and ease of use (2029 Q2) – Based on large-scale application feedback, the final optimization is made for system ease of use and adaptability. For example, improve the compatibility of speech recognition with other dialects, adjust the font size and color matching of the App interface to meet the habits of the elderly, etc. Ensure that the final delivery system reaches TRL (Technology Maturity Level) level 7 or higher (prototypes are validated in a real environment).

l M12: Standard specification development (2029 Q3) – Summarize and refine the technical specifications and data standards of the project. Formulate one set of standard documents such as "Personal Health Semantic Data Specification" and "Active Medical Intervention Service Process Specification", and strive to be published by national societies or group standards committees. At the same time, the DIKWP white-box evaluation indicators are integrated into the AI medical product evaluation criteria (achieved by participating in the activities of the standards committee).

l M13: Preparation for Acceptance of Results (2030 Q1) – Summarize project technical documents, data, paper monographs, patents, demonstration reports, etc., and complete the project summary report and technical report writing. Invite third-party institutions to conduct independent evaluation of the demonstration effect, including the improvement of health indicators, user satisfaction surveys, economic and social benefit analysis, etc.

l M14: Project Acceptance and Promotion (2030 Q2) – Accept the project acceptance organized by the Ministry of Science and Technology to demonstrate the on-site operation and key results of the system. The expected acceptance indicators fully meet the requirements of the task statement (see Chapter 8 Expected Outcomes for details). At the same time, a follow-up promotion plan will be formulated, and cooperation with government departments will be used to select areas that are willing to replicate the model for training and promotion across the country.

Finishing touches (second half of 2030) – Final finishing touches based on acceptance expert input. He has published 1-2 comprehensive research papers and 1 monograph (covering active medicine theory, DIKWP model application, demonstration cases, etc.) in top journals to improve academic and social influence. Organize the project achievements and apply for scientific and technological awards at all levels (if there are outstanding achievements, apply for provincial and ministerial scientific and technological progress awards).

The whole schedule reflects the characteristics of foundation first, application later, and gradual progress, and each stage focuses on different and connects with each other to ensure that the set goals are steadily achieved within five years. In the process of implementation, we will also flexibly adjust according to the objective situation, if some tasks are completed ahead of schedule, we will enter the next stage as soon as possible, and if we encounter technical difficulties, we will focus on tackling key problems to ensure that the overall progress is not affected. Through scientific and rigorous plan management, we strive to complete the project mission on time and with good quality.

7. Budget and resource allocation

7.1 Budgeting Principles and Estimates

The budget of this project follows the measures for the management of national science and technology plan funds, and is reasonably allocated according to the research content and the task volume of the participating units. The budget is based on the principle of "goal-oriented, seeking truth from facts, and taking into account overall planning" to ensure that the use of funds not only meets the needs of project research and development, but also meets the standards of fund expenditure.

Estimated total funding: The total planned funding of the project is 40 million yuan, of which 30 million yuan is applied for science and technology funds from the central government, and the rest is matched by local finance, leading units and social funds (at least 90 million yuan of matching funds to achieve a matching ratio of not less than 3:1, of which local finance is at least 1:1 matching with the central finance). This scale refers to the investment of similar "active health technology integration demonstration" projects (such as a key special project with a total budget of 36.74 million yuan), and comprehensively considers the multidisciplinary intersection and wide demonstration scope of this project, and appropriately increases the investment to ensure the depth of research and application.

Budget by category: According to the category of project activities, the main expenditure items are divided as follows (unit: 10,000 yuan):

(1) Personnel costs: 6 million yuan. It is used to pay for the labor expenses of researchers and research assistants hired by the project, as well as consulting fees and expert consultant fees. Considering the long project cycle and the wide range of personnel involved, personnel incentives should be appropriately guaranteed. Comply with the scope and standards of expenditure stipulated by the state.

(2) Equipment cost: 8 million yuan. It is mainly used to purchase or upgrade the instrument equipment and software system required for R&D. For example, high-performance servers and storage devices required to build a knowledge graph, wearable devices configured at demonstration sites, smart monitoring devices (blood pressure monitors, fundus screeners, etc.), and necessary software licenses. Strictly distinguish the depreciation of new equipment and self-owned equipment, and include those that need to be purchased after demonstration into the purchase list, and the rest are solved by leasing or sharing to save costs.

(3) Material consumables: 2 million yuan. Including low-value consumable materials used in the development process, experimental consumables (sensor components, electronic modules, traditional Chinese medicine samples, etc.), printing of publicity and training materials, etc.

(4) Testing and laboratory processing fee: 1.5 million yuan. It is used to entrust a third party to conduct special inspection tests (such as wearable device reliability testing, data security evaluation), and to prototype specific components (such as customized smart home components for aging).

(5) Travel and conference expenses: $1.8 million. Travel and transportation expenses for researchers in the project team to go out for research, participate in academic conferences, and visit demonstration sites; Organize various conference expenses such as project seminars, user training, expert consultation, etc. Because the project involves multi-regional demonstration, it requires a certain frequency of on-site exchanges, the budget is relatively abundant but the control standards are strictly observed, and the eight requirements are strictly observed.

(6) International cooperation and exchange expenses: 1,000,000 yuan. Expenses for inviting foreign consultants to China for guidance, project members going abroad to participate in international conferences and study tours, as well as expenses for joining international standards organizations and hosting international seminars. This project has international cooperation content, and this fund is used for academic exchanges, standard docking, etc. according to regulations, and is submitted for approval.

(7) Soft subject research and intellectual property affairs fee: 1,000,000 yuan. Including project-related patent application fees, PCT application fees, software work registration fees, standard-setting fees (such as standard proposal fees, publicity fees), and commissioned consultation and evaluation fees. In order to ensure the transformation of expected results and the formulation of standards, this part of the funds is indispensable.

(8) Management fees and indirect costs: calculated according to the proportion stipulated by the state, about 6 million yuan. Including the project management personnel expenses of the lead unit and the participating units, performance expenditure, water and electricity, rent depreciation, management and operation consumption and other cost compensation. The withdrawal of indirect expenses of each unit shall be carried out in strict accordance with the management measures for special funds, and shall not crowd out direct scientific research funds.

(9) Preparation fee: 700,000 yuan. Approximately 1.75 per cent of the total budget is used for adjustments of unforeseen expenditures. The project period is long, and the flexible funds are appropriately reserved to deal with exchange rate changes, price increases or unexpected expenditures.

The allocation of the above-mentioned budget items strives to cover the needs of the project and is in line with the scope of expenditure on science and technology. Each subject will be further broken down into topics and units (reflected in the detailed budget) to ensure that the use of funds matches the amount of tasks. In the implementation of the budget, if it is really necessary to adjust, it will be submitted for approval in accordance with the procedures.

7.2 Funding Sources and Matching Implementation

Central financial funds: Apply for a special fund of 30 million yuan from the Ministry of Science and Technology, which is planned to be supported by the national key research and development plan "regional comprehensive application demonstration of active health digital technology". The funds will be mainly used to support the research activities themselves, including hard investments such as equipment purchase, R&D personnel labor, intellectual property rights and cooperation and exchanges.

Local and unit support: According to the requirements of the guidelines, local finance, units and social funds shall be matched in a ratio of not less than 3:1. In order to implement the supporting facilities, the project has been promised by the local government and supporting units in the application stage: the Hainan Provincial Department of Science and Technology plans to support the construction of infrastructure for demonstration applications of the project (such as the transformation of demonstration sites and the configuration of community equipment); The active medical committee of the World Association of Artificial Consciousness, the lead unit, promises to invest no less than 5 million yuan in the self-raising of personnel and equipment (such as the allocation of on-campus equipment for open use, the provision of office and experimental venues, etc.); Participating enterprises (AI companies, equipment manufacturers) will invest about 5 million yuan in R&D funds and equivalent technical service support (such as stationing engineers, providing some equipment and cloud services for free, etc.). In addition, the local health commission and the civil affairs department will also invest a certain amount of funds in the pilot communities and pension institutions to improve the supporting conditions, and the estimated equivalent funds are not less than 3 million yuan.

In this way, the central financial funds are 30 million: the matching funds are about 23 million yuan, and the ratio is close to 1:0.77, which is less than 3:1. In order to meet the requirements of 3:1 matching, the project will actively strive for more social capital investment. It is proposed to sign a cooperation agreement with potential future users of the relevant results of the joint project (such as pension service groups, health insurance companies, etc.), which will provide demonstration or follow-up promotion financial support. If a large-scale pension institution is introduced to participate in the co-construction in the demonstration area, its investment can reach millions of yuan. In the implementation of the project, the results of the stage can also be used to apply for additional investment from local special funds, such as applying for the improvement of the local digital economy industry fund support system platform.

Fund management and use: Project funds will be managed in a unified manner and accounted for separately. The financial department of the lead unit shall uniformly allocate the central financial funds to the accounts of the participating units, and supervise the special use of funds. The financial affairs of each unit shall control expenditures according to budget items, and shall not be arbitrarily crowded out and adjusted. The supporting funds of the project will also be included in the overall budget management, and the implementation will be checked regularly. Establish a fund monitoring mechanism: The project management office collects the progress of the implementation of each unit's funds every six months, and timely warns and asks for explanations for abnormal implementation rates. Strictly implement the government procurement and bidding system, and openly compare and select large-scale equipment purchases, service outsourcing and other organizations to ensure the efficient use of funds.

Cost-effectiveness considerations: Although the overall budget is high, the project covers the whole chain of basic research, technology development, and pilot construction, and the demonstration is oriented to multiple regions, so the input is proportional to the output. Once the results of this project are realized, the socio-economic benefits will be significant (see Chapter 8) that far outweigh the investment. Through reasonable allocation and management of funds, the project will ensure that each fund is spent on the cutting edge, and strive to leverage the major changes in medical and health services with less investment, so as to maximize the benefits of the use of national scientific research funds.

8. Expected results

Upon completion of the project, significant results and results are expected in the following areas:

8.1 Standards, Specifications and Theoretical Achievements

(1) Improvement of the theoretical system of active medicine: A systematic theoretical framework of the paradigm of "active medicine" is proposed, and a unique medical philosophy perspective is formed by combining the DIKWP model. It is expected to publish more than 10 academic papers, including several active medical theory and application research papers in top journals of artificial intelligence and core journals of medical management. One monograph "Active Medicine: A Paradigm Revolution in Health Services in the AI Era" was formed to summarize the theoretical innovation of the project. This theoretical achievement will lay the foundation for the influence of active medicine in academia and industry, and lead a new direction of related research.

(2) Data and semantic standards: The project will develop a set of group standards for personal health passport data specifications. This standard defines the format, content and sharing interface of personal health data throughout the life cycle, filling the gap in the field of active health data standards in China. In addition, one set of DIKWP semantic health knowledge representation specifications was developed as a guide for health knowledge graph construction and AI inference, and was strived to be included in the international/domestic standard system (through the channels of the World Association of Artificial Consciousness or the China Health Information Standards Committee). The introduction of these standards will help to achieve cross-platform and cross-institutional health data exchange, as well as the evaluation of AI health systems based on evidence, setting a benchmark for the industry.

(3) Expansion of DIKWP AI evaluation specifications: Enrich and expand the application scope of the existing DIKWP white-box evaluation standards by using project experience, and add indicators for initiative and human-machine collaboration in medical and health AI evaluation. For example, the concept of "Health AI Quotient Index" is proposed to quantify the cognitive level of health management AI. Promote the evaluation system to become part of the evaluation index of AI medical products. This move will enhance China's voice in the field of AI evaluation standards.

8.2 Technology platform and application results

(1) Active Medical Intelligence Semantic Fusion Platform: The project will deliver a fully functional active health semantic space intervention platform. The platform is composed of software and hardware, including: integrated traditional Chinese and Western medicine knowledge graph database, DIKWP cognitive reasoning engine, artificial awareness white-box evaluation toolkit, user interaction front-end (App, applet, voice speaker interface), etc. The platform can be deployed in the cloud or on local servers, and its service capabilities cover the whole process of personal health data management, risk assessment and early warning, intervention plan generation, and service resource docking. The platform is expected to reach the technical maturity level of TRL-7 (working in a real-world verification environment) when it passes the acceptance, and can continue to be optimized and upgraded to commercial products in the future. The realization of the platform will become the first active medical digital health service system in China, which is of pioneering significance. Its architecture and module design will apply for 5-8 software copyrights and 8-10 invention patents to ensure that the core technology is independent and controllable.

(2) Application of smart terminals and equipment: Combined with the application of the project, it is expected to develop or improve a variety of smart devices for health monitoring and intervention. For example: intelligent medication management box (with AI reminder and camera monitoring function), home cognitive training machine (built-in cognitive game and AI assistant interaction), traditional Chinese medicine physiotherapy intelligent equipment (such as acupuncture and massage robot with AI guidance), etc. At least 3 or more innovative product prototypes have been launched and verified in demonstrations, which are expected to be industrialized in the future. These devices will enrich China's elderly Wisdom health product line and improve the level of home health care technology.

(3) Application demonstration results: 3-5 demonstration areas of active health service system will be built in pilot communities/elderly care institutions. During the demonstration period, it is expected that the main health indicators of the elderly will be improved (such as a 20% increase in the average amount of physical activity in one year, a 15% increase in the compliance rate of chronic disease indicators such as hypertension and diabetes, and a slowdown in the rate of cognitive decline); Optimization of the utilization of medical services (reduction in hospital emergency rates and avoidable hospitalizations by about 10%); 90% of the participants ≥ satisfied. Formation of experience reports and operating models for each demonstration area. These demonstration results can directly provide reference for local governments to promote, and after the project is completed, similar service models can be promoted in the province and even the whole country to accelerate the implementation of the construction of a healthy China.

8.3 Social Benefits and Demonstration and Promotion

(1) Health benefits: The results of the project directly serve to improve the health level of middle-aged and elderly groups. Through the demonstration and promotion of the project, if active health management is extended to a larger population, it is expected to significantly reduce the incidence of major diseases and disability in the elderly population and prolong healthy life expectancy. For example, through early screening and early treatment, the blindness rate of cataracts is reduced, and the delay in the entry of Alzheimer's disease patients into the moderate to severe stage is extended by an average of X years. The health benefits are also reflected in reducing the burden of family care and improving the psychological well-being of the elderly, which is of great social significance.

(2) Economic benefits: On the one hand, proactive prevention can reduce high medical expenditures and save medical insurance costs. On the other hand, the new technologies and new products spawned by this project have broad market prospects. Based on the active health service platform, commercial services can be derived (such as membership management for chronic disease populations, information solutions for elderly care institutions, etc.). It is estimated that within 5 years after the project, the relevant output value can be driven by hundreds of millions of yuan. If it is calculated according to the annual fee of 1,000 yuan for each elderly person, the output value of serving 1 million elderly people is 1 billion yuan. In addition, achievements such as health data and knowledge graphs can also be used in insurance risk control, pharmaceutical research and development, and other fields to create more potential value. The project will also promote the increase of jobs in elderly care and health management, and promote the development of the health service industry.

(3) Industry demonstration effect: This project will become the "application of active health digital technology" in China The benchmark. The results can provide a basis for the government to formulate proactive health policies (such as incorporating personal health data into the standards of residents' electronic health cards, etc.), and can also provide a basis for enterprises to develop smart health products (such as using the project knowledge graph API to improve the intelligence of their own health applications). If the standards and platforms of the project output are promoted in the industry, it will promote the formation of a new industrial ecology with active health as the core, and bring together ICT enterprises, medical institutions, and the elderly care industry to participate. From a global perspective, the concept and practice of active health/active medicine proposed by China is expected to be promoted as a model with Chinese characteristics and contribute to the "Chinese plan" for global healthy aging. This project communicates and publishes high-level papers and standards at international conferences, and enhances China's international discourse and influence in the intersection of artificial intelligence + medicine.

To sum up, the expected results of the project are rich and far-reaching: not only theoretical breakthroughs, technological innovation, but also application achievements and standards, which have both academic value and application benefits. Through the implementation of this project, China will seize the opportunity in the emerging field of "active medicine", not only to meet domestic demand, but also to lead the development trend of global integration of science and technology and medical care.

9. Supporting the foundation and cooperation mechanism

9.1 Basis of Preliminary Research

The participants of the project have accumulated a solid foundation for the smooth implementation of the project.

The team led by Professor Yucong Duan has published a series of DIKWP models and active health research results: published a number of in-depth analysis articles on the blog and research reports of Science Web, and put forward the DIKWP semantic mathematical ideas and artificial consciousness white-box evaluation standards, which have had an international impact. The team also led the establishment of the International Standards Committee for Artificial Intelligence DIKWP Assessment (DIKWP-SC) and hosted the World Conference on Artificial Awareness (WCAC), which built a platform for promoting the concept of DIKWP and the standardization of active medicine. In the early stage, Professor Duan's team successfully developed white-box AI prototypes such as the "DeepSeek" large model, and released the world's first LLM awareness level assessment report. These preliminary results show that the team is in a leading position in AI white-box evaluation and cognitive modeling, which can be directly applied to the development and evaluation system of AI agents in this project.

In terms of medicine, the participating medical university and Chinese medicine university teams have rich experience in geriatrics and Chinese medicine informatization. For example, Professor Z's team has undertaken the project of "Alzheimer's Disease Prevention and Treatment Technology" of the National Key R&D Program, and established a multimodal early assessment database of cognitive impairment, which provides valuable data support for the Alzheimer's disease scenario of this project. Professor W's team has built a TCM knowledge graph system and developed an intelligent TCM syndrome recognition algorithm, which has reached the domestic advanced level in the field of TCM knowledge engineering, which will greatly promote the TCM knowledge graph work of this project. In terms of the knowledge base of western medicine, the team can share the national drug population knowledge base and the electronic medical record database of a tertiary hospital (a data sharing agreement has been signed), covering major chronic disease diagnosis and treatment standards and real-world data, which is conducive to enriching the content of the project knowledge graph.

On the basis of application demonstration, local governments and institutions have also given support: Hainan Province has demonstrated experience in the construction of community Wisdom elderly care and health islands, and a number of Wisdom medical pilot communities can be directly connected to the project for further improvement. A county pension center has participated in the "Internet + pension" pilot, and has basic equipment and a team of informants, which can quickly carry out rural demonstrations. The AI company DeepSeek has built a health management cloud platform with a certain user base, and the project can be quickly loaded into its platform for scale verification and product transformation after the technology is mature.

In terms of intellectual property rights and achievements, the team has more than 10 authorized invention patents (such as "a knowledge reasoning method based on DIKWP model", etc.) and 5 software copyrights. This lays the foundation for the intellectual property layout of the new technology of this project, and the subsequent new patents can also be combined with existing patents to form a complete technical protection network. In addition, the team has close ties with the World Association for Artificial Awareness (WAC), the Chinese Association for Artificial Intelligence, the Chinese Society of Traditional Chinese Medicine and other organizations, and can use its platform to promote the project results to standards and industries (such as publishing standards at the WAC annual meeting, establishing group standards through the society, etc.) to amplify the impact.

9.2 Domestic and foreign cooperation and resource support

This project focuses on openness and cooperation, and forms a synergy with the help of domestic and foreign resources:

Domestic Cooperation Mechanism:

l It has established cooperation intentions with the China Academy of Chinese Medical Sciences, China Geriatric Health Care Association and other institutions, and has provided support in knowledge base sharing, expert consultation, demonstration and promotion. For example, Professor Yu's team at the Academy of Chinese Medical Sciences is willing to provide the data of its latest dictionary of Chinese medicine; The Geriatric Health Association will organize experts to demonstrate the project's intervention plan to ensure that it is in line with the national guidelines for geriatric health.

l Maintain communication with Huawei and other technology companies to explore cooperation in 5G, cloud computing and big data infrastructure. Huawei intends to provide 5G network optimization and cloud storage support for the project demonstration, and the two parties have signed a strategic cooperation memorandum, and may jointly apply for industry application demonstration in the future.

l Industry-University-Research-Application Alliance: The project plans to establish an "Active Medical Industry-University-Research Alliance" in the implementation of the project, inviting relevant universities, hospitals, and enterprises to join and regularly exchange progress, demand and supply. This will provide a long-term mechanism for the continuous iteration and promotion of the project results, and ensure that there are carriers to continue to deepen cooperation after the project ends.

In terms of international cooperation:

l The project will actively participate in international AI standardization activities. At present, Professor Yucong Duan is an expert of the ISO/IEC Artificial Intelligence Standards Working Group, and the evaluation standards and semantic models output of the project are expected to be proposed by him in the international standards organization. In addition, the World Artificial Awareness Association is promoting the establishment of an international artificial awareness standard system, and the results of the project can be incorporated into it as part of the international norm to improve the international influence of Chinese scholars.

l Cooperate with foreign research teams: We have contacted the Digital Health Center of a medical school of a university in the United States and an AI laboratory in Europe to jointly organize an online seminar, and the other party will act as a consultant in the project to share experience in the fields of proactive prevention and medical AI supervision. This international exchange of peers helps the project to absorb global best practices and stay on the cutting edge.

l International Conference and Publication: The team plans to host an international symposium on active medicine and artificial intelligence (to be held in Hainan in 2029), inviting experts from all over the world to discuss the concept of active health. We will also publish papers on international platforms such as IEEE journals to introduce the results of the project, and strive to promote the concept of "Active Medicine" and China's experience.

Policy and government resource support: The project has received strong support from relevant departments of the Hainan Provincial Government. As a global health demonstration province, Hainan has been included in the list of scientific research and application projects supported by the Provincial Health Commission, which can coordinate the participation of medical institutions and communities in the province. The Provincial Department of Industry and Information Technology has also reserved application scenarios for the project in the digital economy project, allowing the deployment of provincial health data resources for verification. These policy supports will effectively reduce the resistance to project advancement. The Ministry of Science and Technology and the National Health Commission also attach great importance to the active health project, and we will report the progress to the competent authorities in a timely manner and seek guidance and help.

To sum up, with the strong foundation in the early stage and the guarantee of extensive cooperation resources, the project has sufficient conditions to proceed smoothly and achieve the expected results. Through the close integration of industry, academia, research and government and the interaction at home and abroad, we will build the project into a platform for collaborative innovation to maximize resource utilization and optimize the output of results.

10. References

Yucong Duan et al. Active Medicine, Active Health and Passive Medicine (Technical Report). ResearchGate, 2025.

"Guidelines for the Application of 2020 Targeted Projects for the Key Special Project of "Active Health and Aging Science and Technology Response".Annex 8 of CAS Development [2020] No. 74

WANG Zhuhua. Large Language Model Awareness Level "Knowledge" White Box DIKWP Assessment 2025 Report Released Science and Technology Daily, 2025

Prof. Yucong Duan: DIKWP artificial consciousness model leads the future of AI, and 114 patents are expected to be implemented in the industry China Convergence Media Industry Network, 2025

Yucong Duan. A Brief Analysis of the Theory, Structure and Application of DIKWP Semantic Mathematics (Research Report). ResearchGate, 2025.

Yucong Duan. Capability mapping meta-analysis of DIKWP white-box assessment and LLM black-box benchmarking (Research Report). ResearchGate, 2025.

Zhong Daidi. The project of "Demonstration of Active Health Technology Innovation and Application in Different Economic Regions" was launched Chongqing University News Network, 2024

Yucong Duan. Active Medicine: The Philosophical Upgrade of Medicine in the AI Era. World Artificial Consciousness Conference Keynote, 2024.

LI Peng. "Active Health" helps you prevent illness before it happens Xinhua Daily Telegraph, 2021

National Health Commission. Healthy China Action (2019-2030). 2019.