3.1 Strengthening infrastructure and multi-modal data integration

The development of digital health critically hinges on bolstering information infrastructure, particularly technology-based digital systems. This involves fostering the expansion, collaborative sharing, and intelligent upgrading of information infrastructure. While China has made notable strides in digital construction, there remain considerable unmet needs of over one-third of the population residing in rural areas, who may face limitations due to educational levels, economic conditions, or lack of internet access [4]. Therefore, prioritizing infrastructure enhancement is imperative for the upcoming decade. This should encompass hardware such as network facilities, medical equipment (e.g., IoT), information security hardware, and software such as medical information systems, big data, AI applications, and mobile apps. This development will create significant demand for suppliers. By 2024, the market size of China's digital healthcare is expected to expand to 413 billion CNY, more than doubling from 2022. With this huge market, government and healthcare institutions need to conduct long-term planning and establish supervision and management systems to avoid resource waste.

Advancements in AI technology have significantly improved the processing and analysis of multi-modal data. Research has shown that integrating radiologic and histologic features with clinical and genomic data enhances cancer patient risk stratification [5]. The integration and governance of multi-modal data hold substantial value in digital health, as they can enhance diagnostic accuracy, optimize resource allocation, and foster advancements in medical research. With the continuous progression of technology and expanding application scenarios, the potential applications of multi-modal data in digital healthcare are set to increase.

3.2 Integrating digital solutions into the healthcare delivery system

The GIDH underscores the importance of leveraging digital innovations to enhance healthcare accessibility, efficiency, and quality across the continuum of care. Integrating digital solutions into the entire healthcare delivery system is an essential strategy for transforming health services in China. This involves extending beyond traditional hospital care to encompass out-of-hospital services facilitated by digital technologies [6]. Electronic health records, telemedicine, AI-driven diagnostics, and mobile health platforms, when incorporated across all aspects of healthcare delivery, have the potential to enable seamless patient data flow, support real-time decision-making, and facilitate personalized care plans. Thus, in turn, can improve health outcomes and enhance patient experiences.

3.3 Promoting research in consumer health informatics and behavioral science

The utilization of digital health for delivering behavioral interventions related to physical activity, diet, drug use, and mental health is increasing [7]. However, China's diverse regional cultures and customs necessitate research on the application and effectiveness of these digital health interventions in various cultural contexts. Developing health intervention methods that are tailored to the unique needs of Chinese consumers in different regions is essential. Additionally, fostering collaboration between disciplines such as medicine, informatics, and psychology will be crucial for advancing research in health informatics and behavioral science. Interdisciplinary cooperation will help to better understand and address the cultural nuances that influence the effectiveness of digital health interventions in China.

3.4 Developing a comprehensive Chinese terminology ontology system

The development of a widely adopted Chinese terminology ontology system is crucial for ensuring interoperability and supporting data mining efforts. Although digital health technology adoption increased during the COVID-19 pandemic, it fell short of expectations partly due to the lack of standardized data [8]. Most existing medical terminology systems are designed for English speakers, and the current Chinese clinical medical terminologies – such as the Chinese version of ICD-10, Chinese Medical Subject Headings (CMeSH), and the Traditional Chinese Medicine Language System (TCMLS) – primarily consist of medical vocabularies. However, these terminologies exhibit weak semantic and associative relationships and lack a dynamic updating mechanism.

To address these issues, the design of a Chinese medical terminology/ontology system should learn experience from the mature systems such as SNOMED CT and LOINC, along with a continuously updated database and corpus for standardizing clinical data. Furthermore, traditional Chinese medicine terminology must also be standardized. With the support of this terminology ontology system, natural language processing and deep learning technologies can be applied to extract valuable information from the vast amount of clinical data, advancing the capabilities of digital health technologies in China.

3.5 Embracing advanced technologies in healthcare innovation

As we enter the era of AI, the widespread use of Large Language Models (LLM) is becoming increasingly inevitable. The success of ChatGPT in demonstrating the capabilities of LLM-empowered chatbots highlights their potential to revolutionize the healthcare industry [9]. Therefore, it is crucial to proactively explore the applications and viability of integrating chatbots across various healthcare sectors. Although access to ChatGPT is restricted in China, many companies and universities are researching and developing their own medical LLM. Additionally, emerging technologies such as smartphone apps, virtual reality, and social media are reshaping healthcare practices in unforeseen and exciting ways. Expanding access to digital medicine requires integrating technology, operational models, policy, and application scenarios. For example, medical institutions can establish intelligent medical service platforms powered by AI, providing services such as consultation, education, and decision support to meet patients’ diverse needs. Encouraging innovative research to explore these technologies will be vital in leveraging their potential to enhance healthcare delivery and outcomes.

3.6 Quantitative privacy risk assessment in Chinese health and medical data

Developing privacy risk assessment tools customized for Chinese health and medical data is crucial for quantifying privacy risks associated with data use. Assessing both data set quality and sensitive information security is essential for minimizing risk and information loss [10]. While studies have shown that the Health Insurance Portability and Accountability Act (HIPAA) policy has a similar protective outcomes in patient data in China [11], further evidence is required to validate its effectiveness in the Chinese context.

Medical multi-model data storage should be restricted to medical or government institutions. Establishing balanced policies for data use and privacy protection is essential, covering the entire data lifecycle – data generation, transmission, storage, use, sharing, destruction, and management. Technologies such as federated learning can enable data sharing while ensuring data security. These policies and technologies will help create a robust data application ecosystem, ensuring the responsible and ethical use of health and medical data in China.