5G and AI Pave the Way for Future Smart Hospitals: CAS, Zhejiang Digital Research Institute, Jianpei, and Meinian Collaborate on Technological Innovation

On August 29, the inaugural Yangtze River Delta Health Industry High-Quality Development Conference and the 2019 West Lake Health Forum · Zhejiang International Health Industry Summit Grandly Opened at the Hangzhou International Expo Center. Themed “5G Ushering in a New Era of Medical AI,” the forum was jointly hosted by the Zhejiang Provincial Health Commission, the Zhejiang Provincial Development and Reform Commission, and the Zhejiang Provincial Department of Economy and Information Technology. It was organized by Hangzhou Jianpei Technology Co., Ltd. and the Zhejiang Health Service Promotion Association, with support from the Yangtze River Delta Health Industry Alliance, the Zhejiang Medical Association, the Xiaoshan Economic and Technological Development Zone, Meinian Onehealth (Health 100), Hangzhou Bay Information Port, Intel, and other organizations.

As the dawn of the 5G era ushers in a new chapter, surging technological forces are reshaping the boundaries of imagination. Seizing this opportunity to overcome challenges and propel healthcare to new heights has become the key to unlocking the future of smart healthcare. As an annual international summit bridging healthcare and AI, West Lake Health Forum brings together experts from domestic and international medical, academic, and industrial sectors. The forum delves into cutting-edge topics such as “5G Empowering Medical AI” and “Hospital Intelligence in the 5G Era,” focusing on multiple dimensions including 5G technology, AI enablement, smart healthcare, and industrial openness. From a global perspective, participants jointly explore the innovation and development of “5G + Healthcare” and “Healthcare + AI.”

Following the conference, VCBeat reviewed the presentations delivered by several speakers, making appropriate edits and abridgments. The summarized content is presented below.

In the past, artificial intelligence (AI) remained largely confined to academia. With the advent of the big data era, AI transitioned from being unusable to usable. However, it has become evident that deploying AI in real-world applications is challenging. Commercialization is not something that can be achieved overnight; there is still a long road ahead from making AI merely usable to making it highly effective and user-friendly.

In the healthcare sector, there are numerous ways to integrate artificial intelligence (AI), including accelerating new drug development, assisting physicians with diagnostic and therapeutic decision-making, and leveraging AI and big data for health management, medical rehabilitation, and wearable devices, thereby enabling the delivery of healthcare services in home settings.

In the aforementioned integration, President Xu Bo pointed out that there are tens of thousands of disease types in medicine, each with its own specialized domain and expertise. How, then, can artificial intelligence play a more significant role in this field? The application of medical big data provides the most fundamental and comprehensive solution to facilitating the practical implementation of AI in healthcare.

There is a profound chasm between medical artificial intelligence and the internet. Integrating the two presents a formidable challenge, effectively reshaping healthcare relationships. In this process of transformation, individuals (patients), devices, technologies, and hospital regulatory authorities are all actively involved, forming a new collaborative ecosystem.

Originally, there was an interactive relationship between patients and healthcare professionals. The equipment used by healthcare staff consisted of simple, unidirectional, non-intelligent devices that could be operated according to standardized procedures. However, with the integration of artificial intelligence, we have observed the emergence of a direct relationship between patients and medical devices.

For instance, in intelligent consultation, when we query a robot, we engage in a direct interaction with the machine, necessitating a fundamental reshaping of policies, regulations, ethics, liability, and legal frameworks. Previously, the relationship between healthcare professionals, equipment, and technology was unidirectional: doctors followed fixed procedures to deliver results. However, the integration of artificial intelligence has endowed equipment systems with intelligence, transforming this into a bidirectional interactive relationship, which imposes new requirements on healthcare professionals. With such new devices, new medical scenarios, and new doctor-patient dynamics, it is essential to establish a new system that is trustworthy, reliable, high-quality, and efficient, requiring coordinated efforts among policy-making, application, and industry sectors. Therefore, we argue that the integration of medicine and artificial intelligence is still in its nascent stage.

So, why is it said that medicine and artificial intelligence are just getting started? Three aspects can explain this:

First, most medical AI products are still in the experimental stage and remain some distance away from meeting clinical workflow requirements and achieving practical implementation. Currently, the majority of diagnostic solutions rely either on text or on imaging data, lacking the comprehensive approach of real-world physicians, who integrate patient consultations, pathological findings, and various other diagnostic data through multimodal analysis to reach a conclusion. Consequently, most products remain experimental, lacking standardized validation protocols as well as third-party benchmarking data and evaluations. Furthermore, constructing open databases suitable for training and evaluating medical AI systems is a critical step in the advancement of artificial intelligence.

Second, medical AI has a broad range of application scenarios. Currently, most products focus on improving local efficiency, lacking significant global achievements and failing to generate a driving effect.

Third, currently, few genuine medical artificial intelligence products in China have obtained certification from the CFDA.

So, at this stage, where do the challenges lie for medical artificial intelligence technology?

Looking back on the 60-year development of artificial intelligence, the entire process has been inseparable from three major technical schools: symbolism, connectionism, and behaviorism. All three schools draw inspiration from human thinking, behavior, or the structure of the brain. While each research approach has its own advantages, none can solve all problems.

For example, early symbolic artificial intelligence attempted to encode physicians’ knowledge into symbols within computers. This school of thought treated symbolic computation as a cognitive process, transforming experts into knowledge bases for reasoning and judgment to provide information to users. The core challenge lies in the difficulty of representing and acquiring differential knowledge in computers.

The most typical example is a traditional Chinese medicine (TCM) expert system from the late 1980s. Researchers encoded accumulated TCM knowledge into a computer, creating a highly complex system that covered etiology explanations, association queries, knowledge acquisition, and the construction of a TCM information knowledge base. However, this expert system failed. The main reasons include: first, the abundance of empirical common sense made it difficult to formalize; second, the rule coverage was insufficient; and third, the complexity of the rules led to numerous contradictions.

The core of what we now refer to as artificial intelligence is deep learning. The most significant difference between this approach and expert systems lies in its being data-driven rather than knowledge-driven. Its output logic does not rely on simple linear relationships; instead, it processes massive volumes of information through complex deep network architectures, employing high-dimensional nonlinear transformations to ultimately provide clinical decision support.

With the aid of deep learning technologies, medical artificial intelligence has advanced rapidly. Taking speech recognition as an example, its accuracy remained virtually unchanged between 2000 and 2010, prior to the emergence of deep learning; however, after 2010, the error rate of this technology declined sharply. Such advancements have opened up broad application scenarios for AI in healthcare. Departments such as ultrasound, radiology, and operating rooms have gradually begun to adopt AI technologies, which help free physicians’ hands to improve diagnostic efficiency while enhancing the quality and standardization of electronic medical records.

Perception-based technological products, such as those used in medical imaging, have also seen significant improvements. Previously, it took radiologists 3–6 hours to review the images for a single case, a task that relied heavily on the physician’s experience. However, with the adoption of artificial intelligence, the system can now identify an image in just 16 milliseconds, achieving a lesion detection rate of 99.5% and delivering results in real time. This has substantially enhanced the patient experience. Additionally, AI-assisted segmentation and measurement of certain images in medical imaging play a crucial role in clinical practice.

However, mere perception and recognition do not address the ultimate challenge of judgment and decision-making. In terms of cognition, artificial intelligence primarily leverages large-scale medical knowledge graphs to enhance the accuracy and automation of assisted diagnosis and treatment. It can be stated that medical knowledge graphs are the cornerstone of smart healthcare and the foundation supporting more precise medical services.

The construction of medical knowledge graphs involves data quality assessment, medical reasoning, medical knowledge fusion, knowledge extraction, and knowledge representation. Taking the pancreatitis knowledge graph as an example, after obtaining deep learning results, researchers employ automatic feature extraction methods to build the knowledge graph. However, the accuracy of current automated extraction is not yet sufficiently high, necessitating a certain degree of manual intervention for verification and correction.

However, deep learning has its limitations and still lags behind human understanding and cognition. Its drawbacks include high computational resource consumption, poor handling of uncertainty, reliance on large volumes of high-quality annotated data, susceptibility to adversarial examples, and lack of interpretability in results.

For instance, under normal circumstances, artificial intelligence can easily recognize images of animals such as cats and dogs; however, if white noise is artificially inserted into the image, human vision cannot discern the difference, whereas machines will classify it as a novel object.

So, where will the focus of future development in medical AI lie?

Before discussing development, Professor Xu Bo believes it is essential to affirm the following two fundamental principles:

First, regardless of how medical artificial intelligence evolves, physicians remain the central agents, while AI serves as a tool. Although artificial intelligence will ultimately play a significant role in the future development of healthcare, it will remain a tool for a considerable period. No matter how AI advances, physicians are the primary subjects in medical practice.

Secondly, modern medicine encompasses at least tens of thousands of disease entities. It is impossible for artificial intelligence (AI) to address every single domain individually. Therefore, it is essential to empower physicians through platforms that aggregate the collective expertise of diverse clinicians, thereby driving the development of medical AI. Thus, only by leveraging such collective intelligence via AI platforms can we resolve the core challenges in medical AI, using intelligentization to lead informatization.

Professor Xu Bo’s team is currently focusing on the following key initiatives:

First, how can human-computer interaction improve the completeness of medical data? Professor Xu Bo believes that there is significant potential for human-computer interaction to enhance the integrity of medical data, making it a top priority.

The second focus is on hybrid cognitive intelligence technology that integrates physicians’ expertise with clinical data. Although abundant clinical data are currently available, deep learning models derived from these data often function as “black boxes” lacking interpretability. Therefore, Professor Xu Bo is prioritizing the integration of physicians’ knowledge with clinical data to address the challenges of mutual conversion among data, knowledge, and models.

Third, the open platform for medical artificial intelligence. As AI serves as an enabling technology, Professor Xu Bo aims to build a medical AI platform for physicians, organically integrating clinical practice, scientific research, and intelligent technologies.

Nowadays, AI platforms in healthcare are placing higher demands on data sharing, data linkage, data security, data annotation, and benchmarking. Current AI platforms may fail to meet the needs of physicians and clinical applications. Moreover, there is currently no dedicated benchmark platform for medical data annotation and training, either domestically or internationally. However, such infrastructure would significantly propel the future development of medical AI. Therefore, advancing this area is an urgent priority.

Researcher Jiang Wei’s presentation was primarily divided into three parts: first, the current state of 5G smart healthcare development; second, how 5G technology empowers smart healthcare; and third, strategic approaches for the development of 5G healthcare.

The "Made in China 2025" planning outline proposes comprehensive breakthroughs in 5G mobile technology. Beyond 5G, cloud computing, big data, artificial intelligence, and virtual/augmented reality technologies will all become critical infrastructure for future digital transformation. Therefore, under the three-year action plan for building a cyber power, substantial national infrastructure investment will be directed toward 5G infrastructure development. Guided by national strategy, various provinces have initiated preliminary layouts for 5G-enabled healthcare. For instance, Zhejiang Province, in its implementation opinions issued on April 28 to accelerate the development of the 5G industry, explicitly stated that it would launch demonstration projects for "5G + Smart Healthcare."

Researcher Jiang Wei believes that the application of 5G technology will be more concentrated in the construction of smart hospitals and operating rooms. As of August 30, 20 provinces and cities across China, along with 46 hospitals, had launched 5G-related projects. In June 2019, following the Changning earthquake, Sichuan Provincial People's Hospital, China Mobile Sichuan Branch, and China Mobile (Chengdu) Industrial Research Institute formed a joint rescue team. They deployed 5G-equipped ambulances to the disaster area and raced against time to save the injured through the nation's first 5G emergency rescue system. In July 2019, Shanghai Ruijin Hospital also conducted a series of live-streamed surgeries leveraging the 5G network.

What can 5G technology actually contribute to the smart healthcare solutions currently being explored? We must first clarify the key features of 5G, namely high bandwidth, massive connectivity, and low-latency data transmission.

The 5G era emerges from the maturity of 4G. Everything currently achievable with 4G will undoubtedly be enhanced by 5G in the future; furthermore, areas beyond the reach of 4G will gain significant expansion potential by leveraging the unique technical features of 5G networks. Advanced applications such as next-generation ambulances, high-level remote consultations, and telesurgery can only be realized under the conditions provided by 5G technology.

Specifically, 5G has the potential to optimize medical workflows, expand the scope of services, and enhance operational efficiency. For instance, in scenarios involving diagnostic and treatment guidance, leveraging 5G’s enhanced mobile broadband capabilities enables physicians to conduct remote consultations, provide instruction, and deliver emergency care. This helps break through physical spatial constraints and facilitates the decentralization of medical resources. Regarding remote expert guidance and training, while 4G previously offered only two-dimensional video, future integration of 5G with augmented reality (AR) and virtual reality (VR) technologies may provide primary-care physicians with more immersive educational experiences.

Furthermore, 5G will drive changes in the operation of future hospitals and their overall evolution model. For instance, 5G may break down geographical barriers to diagnosis and treatment, gradually transforming the current healthcare structure into a regional medical center model centered on health management.

Current ambulances have largely achieved the core capability of transmitting emergency medical data, enabling interactive exchange of medical information between hospitals and ambulances. However, at future emergency scenes, 5G technology may intelligently optimize ambulance routing, connecting patients, onboard equipment, and remote experts to establish a coordinated, real-time emergency response system.

Ward-round robots also represent a key direction for 5G applications in healthcare. For instance, in wards housing patients with highly radioactive or contagious diseases—areas previously inaccessible to physicians—5G-enabled ward-round robots may prove invaluable.

In the past, the da Vinci Surgical System—characterized by its flexibility, intelligence, and high sensitivity—was confined to limited operative fields. However, 5G technology may enable da Vinci robots in medical institutions to serve a broader population of patients outside the hospital setting.

There is a significant shortage of ultrasound physicians in primary healthcare settings, a gap that cannot be bridged in the short term. By leveraging 5G technology to connect tertiary hospitals with primary care institutions and transmit ultrasound images over the network, patients in grassroots communities can access expert-level ultrasound diagnostic services. This scenario can help alleviate, to some extent, the shortage of ultrasound physicians.

Hospital management is another application scenario. Leveraging 5G coverage, hospitals can utilize the enhanced mobile broadband capabilities to deploy material-handling robots and guidance robots, thereby reducing human resource consumption.

5G is a foundational technology. Although 5G can deliver signals into hospitals, there remains a "last meter" gap between the network and end-user devices, as connections between devices still need to be converted into basic Wi-Fi signals. Therefore, to truly achieve a 5G network, all relevant equipment within this last-meter range must be upgraded to comply with 5G standards.

The advent of 5G will inevitably trigger another reshuffling of standards. Hospitals must not only resolve lingering ambiguities from previous standardization efforts but also address the new disruptions brought by 5G, carefully considering how to integrate 5G with existing healthcare informatics standards to ensure interoperability between systems. Therefore, we cannot afford to wait until old issues are fully resolved before tackling new ones; instead, we should adopt a second- or third-curve development model, progressing through a spiral of continuous improvement.

Since 5G networks require substantial investments in manpower and capital from telecom operators, it is unrealistic to expect all applications to be empowered by 5G simultaneously. Instead, we should take into account actual conditions, focusing on current pain points and the maturity of existing technologies, to selectively explore innovative 5G-enabled medical applications.

The number of devices in the Internet of Everything (IoE) will far exceed that of currently connected devices; therefore, we must prioritize the security of device connectivity. Consequently, it is imperative to establish robust security mechanisms and advance 5G standardization. In the past, our focus has been on exploring digital transformation, including the digitization of healthcare and medical devices. However, as we integrate an increasing number of devices, new challenges will inevitably emerge.

Finally, Researcher Jiang Wei summarized the future of 5G:

First, infrastructure determines the actual implementation of upper-layer applications. As an infrastructure, 5G will inevitably have a profound impact on upper-layer applications.

Second, leverage technological synergy to achieve value stacking. With the advent of 5G, it is essential to integrate it with existing technologies to drive new innovations.

Third, continuously enhance interoperability between systems through the application of information standards.

Fourth, the selection of key applications requires a combination of technology maturity and medical pain points.

5. Prioritize security development alongside application deployment, and promote adoption through demonstration projects.

After gaining an understanding of the inherent characteristics and development of 5G, Cheng Guohua, Chairman of JIANPEI, provided a more in-depth analysis of application scenarios for the convergence of 5G and AI within the broader context of multi-technology integration.

JIANPEI has long been actively exploring big data in medical imaging, witnessing firsthand the evolution of AI over time. Cheng Guohua categorizes these changes into three aspects.

The first change stems from the exponential growth in computing power. With the advent of AI, GPUs have evolved into dedicated AI chips, while customized chips such as FPGAs have also carved out their own niche. Google’s Tensor Processing Unit (TPU) accelerators have become one of the mainstream solutions, and Intel has begun developing CPUs specifically designed for AI. Across various fields, data is experiencing exponential growth. The emergence of AI has driven comprehensive industrial upgrading; without the increase in computing power, many advancements would be unimaginable, and numerous computational tasks would be prohibitively expensive.

The second change stems from the fact that medical AI “brains” at various levels have begun to take shape. With the advent of AI, healthcare informatics has undergone profound transformations; all data generated through informatization can now be analyzed using artificial intelligence, effectively constructing a centralized “brain.”

From the hospital perspective, researchers are undertaking a similar endeavor: evolving the “medical brain” into a thinking hospital. When multiple hospitals are interconnected, they form an intelligent hospital consortium and community, constituting a higher-level brain.

JIANPEI is currently developing a "Medical Brain" to build a remote assisted diagnosis platform for Zhejiang Province. JIANPEI not only connects prefecture-level cities in Zhejiang to medical cloud computing, providing them with algorithm libraries and computational accelerators, but also facilitates infrastructure integration for grassroots medical institutions, such as township health centers in Zhejiang. The foundation of this entire Medical Brain consists of JIANPEI’s neural networks and its imaging cloud infrastructure. In summary, JIANPEI Technology’s Imaging Cloud and Diagnostic Cloud together form the brain of medical imaging big data, which constitutes the core architecture of "Zhuo Yisheng" (the AI film-reading robot developed by JIANPEI Technology).

The capabilities of the "Medical Brain" have evolved rapidly in recent years, yet they still require concerted collaboration among enterprises, hospitals, and government bodies. In this regard, JIANPEI has opened access to Zhuo Yisheng’s database and algorithm library, seeking collaborative partnerships with hospitals, universities, and industry peers.

The third change: AI edge applications are blossoming everywhere.

The evolution brought by AI is not limited to the "brain"; it also drives the evolution of edge applications. In this analogy, the "brain" resides in the cloud, while edge applications operate at the outermost periphery.

Take smartwatches as an example. Traditional watches lacked medical functionalities, whereas internet-connected smartwatches can leverage sensors to help us measure blood pressure and blood oxygen levels. Smartphones are also gradually becoming an integral part of health management; for instance, patients can use the camera function to capture images for the diagnosis of skin diseases and cataracts.

There are many similar smart devices, and 5G is the core that connects them. Through this connectivity, AI can play a broader role, extending its benefits to every member of the public.

To address the issue of healthcare inequality, the state has introduced a tiered diagnosis and treatment system, promoted the development of internet-based healthcare, and advanced the construction of medical consortia. Tiered diagnosis and treatment is a landmark policy; however, if triage is conducted manually, it will push the healthcare system to another extreme—where the supply of physicians fails to meet demand. Therefore, AI is the solution to this problem.

Through data analysis, AI can automatically triage patients to hospitals at various levels via intelligent triage systems. With the support of 5G, true telemedicine has emerged. Taking remote ultrasound as an example, image acquisition heavily relies on the operator’s technique. Without the low-latency and high-reliability characteristics of 5G, remote ultrasound would be difficult to control; 5G effectively addresses this challenge.

Furthermore, leveraging 5G network slicing and massive connectivity technologies, AI enables the dynamic creation of network slices at any given moment to rapidly connect patients with the systems, devices, and healthcare personnel required for their treatment. Consequently, we can establish a virtual hospital for each patient, thereby delivering patient-centered medical services.

For the general public, how can health management be shifted toward a prevention-first approach? This is where edge AI comes into play. For instance, smartphones can help users monitor their heart rate while running, and smart bands can report epileptic seizures. However, to unlock the deeper value of these data, they must be connected to a central medical intelligence system; otherwise, the data will remain fragmented. Today, 5G and AI effectively build an invisible bridge between devices, bringing us closer to the era of the Internet of Everything.

Overall, 5G+AI has addressed many pain points and represents a new transformation. Cheng Guohua believes that the new era of 5G+AI-driven transformation began to unfold gradually in 2019.

The first transformation is the revolution in health management. The integration of 5G and AI will elevate healthcare to the status of essential infrastructure, akin to water, electricity, and gas. Users can access these services through subscription, connectivity, and payment models. For instance, bedrooms in the new era could monitor whether a user has spinal disc herniation or sufficient sleep duration. Smart pillows could help users fall asleep faster and improve sleep quality. Healthcare services can intervene in these areas, fostering new business models. This represents a disruptive shift in health management.

The second transformation is that of family doctors. The character of “Baymax” from *Big Hero 6* has become deeply ingrained in public consciousness. In the scenarios depicted in this film, the robot Baymax provides the protagonist with a variety of examination services, including NLP-based consultations and real-time heart rate monitoring, much like a family doctor who is always by one’s side.

This offers us a glimpse of a very promising future, which has been gradually unfolding since 2019. Our 5G technology has endowed the real-life “Baymax” with always-on analytical capabilities. Together with “Baymax,” the underlying network of physicians and medical AI systems will jointly serve every resident.

Overall, AI’s analysis of comprehensive bodily big data will serve as an excellent tool for historical tracking and future prediction. Thus, in the era of 5G+AI, scenarios once confined to science fiction films are becoming increasingly close to our daily lives. Therefore, 5G has truly ushered in a new era for medical AI.

Globally, population aging is a contradiction and challenge that the entire world will face in the future. Therefore, from the perspective of digital health, it has become particularly important to consider what we can do in the coming years to address global population aging.

In response to this contradiction and challenge, Yu Rong believes: aging is a trend, chronic diseases are the focus, and prevention is the key.

Future trends in the healthcare industry are characterized by three key features: first, an upgrade in cost containment; second, an upgrade in consumption; and third, an upgrade in innovation. While there is ongoing debate about whether overall consumption is upgrading or downgrading—with varying patterns across different sectors—consumption upgrading in the health sector is undeniable. Therefore, in the realm of backend treatment, enhanced cost containment will inevitably become a dominant trend in the medium to long term.

All countries have shifted their core strategies and tactics toward “early prevention and early intervention,” making disease prevention, preventive treatment of potential diseases, and health promotion central pillars of national health policies. These efforts are reflected in areas such as the prevention and control of major diseases and cancer screening.

In the broader context of emphasizing prevention, health check-ups serve as a critical entry point and an indispensable frontline gateway. However, the current health check-up coverage rate in China stands at only 30%–40%, lagging significantly behind that of developed countries. This indicates substantial market potential, with future demand poised for rapid growth.

The data indicate that the health checkup market is maintaining robust and rapid growth, with its penetration rate gradually increasing. Meanwhile, China’s highly fragmented market structure offers significant room for expansion for professional chain operators.

Meanwhile, health checkups will become the entry point to the future medical ecosystem. By integrating pharmaceuticals, internet healthcare, specialized services, health management, and insurance, a highly efficient, data-driven closed loop will ultimately be formed. This serves as the fundamental cornerstone for building a future ecosystem anchored in prevention.

Therefore, enhancing the preventive health awareness of Chinese residents is a key focus for Meinian’s future. Specifically, Meinian will transition from a quality-driven service provider to a digital-driven technology platform, continuously enriching and improving the substance and quality of health examinations through technological innovation, thereby infusing preventive medicine with new connotations and content.

In 2018, the number of health check-ups conducted by Meinian Onehealth exceeded 3 million. The high-quality data generated from these 3 million visits will create greater value.

These values are rooted in Meinian’s transformation in digital healthcare and artificial intelligence. In the realm of AI, Meinian offers AI-based imaging diagnostics for conditions such as pulmonary nodules and diabetic retinopathy. Through its cloud-based imaging platform and electrocardiogram (ECG) information platform, Meinian can remotely monitor the quality control and operational status of examinations at nearly 700 institutions across China. In traditional Chinese medicine (TCM), Meinian has already implemented intelligent TCM diagnostic devices. The application of AI and big data is just the beginning; Meinian will continue to pursue further intelligence and automation.

Furthermore, these applications are underpinned by a critical infrastructure: the establishment of the Meinian Health Biobank. In the future, Meinian will collaborate with other national-level entities to jointly build a nationwide biobank.

Therefore, Meinian aims to leverage professional and precise health examination traffic and data to build a future-oriented health screening scenario driven by technological innovation and powered by artificial intelligence, thereby providing users with higher-quality and more professional health screening and health management services. Furthermore, based on its own preventive care platform, Meinian seeks to create a closed-loop, full-lifecycle health management service platform, contributing to the Healthy China initiative.