Ruijin Hospital Launches Medical Health Digital Twin Alliance, Initiates 50 Million Human Health Digital Twin Plan

“We need to build a ‘global digital navigation’ for human health: one that mirrors the present, tracks the future, guides interventions, and keeps learning.” On October 30, the Academic Symposium of the Healthcare Digital Twin Alliance was held at Ruijin Hospital Affiliated to Shanghai Jiao Tong University School of Medicine. Ning Guang, academician of the Chinese Academy of Engineering and president of Ruijin Hospital Affiliated to Shanghai Jiao Tong University School of Medicine, said this in his address. The symposium focused on the research progress and clinical application of healthcare digital twins and invited experts in related fields and editors of top international journals to discuss, around the integration of population health and artificial intelligence, the development pathways of healthcare digital twins in disease mechanism research, dynamic monitoring, risk prediction, scientific evaluation, and clinical translation. At the meeting, the list of alliance committee members, the core work plan, and research progress were released, and the unveiling of the alliance headquarters and a member signing ceremony were held. How can health information accumulated from repeated examinations, long-term follow-up, and daily monitoring be connected to detect disease risks earlier? This is the question that healthcare digital twins hope to answer. By integrating longitudinal health data with dynamic models, researchers attempt to depict the trajectories of individual health changes and provide a basis for disease prevention and personalized diagnosis and treatment. The Healthcare Digital Twin Alliance (ICDT-HM), initiated by Ruijin Hospital, brings together research forces from universities, medical institutions, and related fields at home and abroad. At this symposium, Chinese and foreign academicians, experts and scholars from universities and medical institutions such as Stanford University, the University of Oxford, Macau University of Science and Technology, Wenzhou Medical University, and Duke-NUS Medical School, as well as editors from the Nature series journals, participated in exchanges on site and online. The alliance will promote the deep integration of large-scale longitudinal population cohorts and artificial intelligence, build international consensus on data standards, cross-population validation, and data governance, and gradually construct predictable and simulatable digital twins of human health. Academician Ning Guang said that perhaps the most profound value of digital twins lies precisely here: first walking through the future in the virtual world so that diseases that could have been avoided need not occur in reality. Ning Guang introduced that Ruijin Hospital has proposed the concept of “AI-native hospitals,” making AI the “coding gene” of new hospitals and health service systems. By gathering multidimensional data such as biological, physiological, psychological, behavioral, social, and environmental data, it explores the construction of an individual’s “digital life” coordinates, enabling health to shift from the binary judgment of “sick” and “not sick” to a measurable, predictable, and navigable continuous trajectory. Speaking of open sharing, Ning Guang said: “Our principle is: comply with the law and regulations, keep data securely local, and let models and knowledge go global.” He also emphasized: “No matter how precise a twin is, it is only a mirror image; no matter how massive the data, it is only a projection. The original intention of medicine is always the warm human being in front of the mirror.” Luo Dajin, director of the Shanghai Municipal Science and Technology Commission, said in his address that healthcare digital twins stand at the intersection of artificial intelligence, life sciences, and data elements. He expects the alliance to bring together first-class universities, medical institutions, and builders of national population cohorts at home and abroad, focus on major diseases such as metabolic diseases, cardiovascular and cerebrovascular diseases, and cancer, integrate multidimensional data such as clinical diagnosis and treatment, long-term follow-up, multi-omics, and wearable devices, and strengthen basic research and clinical validation; promote the construction of large population cohorts, unify data standards, and explore effective pathways for privacy protection, compliant authorization, trusted circulation, and security governance; and deepen international cooperation to promote the translation and application of healthcare digital twin achievements. Francesca Cesari, senior editor for biology, clinical, and social sciences at Nature, emphasized in a video address that study design, data quality, external validation, and reproducibility are as important as data scale, and that predictive performance cannot replace clinical evidence. Which populations and scenarios a model applies to, what limitations exist, and whether it can improve patient outcomes all need to be carefully evaluated. The value of digital twins should be reflected in better-grounded clinical decisions, more equitable and effective medical services, and a reduction in the burden of disease. Zhang Kang, vice dean of the Faculty of Medicine at Macau University of Science and Technology, dean of the Institute of Artificial Intelligence Interdisciplinary Applications, and chief scientist of the National Clinical Medical Research Center for Ocular Diseases, introduced the alliance’s development blueprint and core work, with a focus on sharing the “50M Human Health Digital Twin Initiative.” The “50 Million People Digital Twin Initiative” is a long-term scientific research plan proposed by the alliance for precision health, aiming to gradually build a research system integrating longitudinal health data and dynamic models through multi-party collaboration, providing support for disease prediction, precision prevention, and personalized medicine. Fifty million people is the planning target, and the related construction will be explored and advanced in stages. The alliance will promote the deep integration of large-scale longitudinal population cohorts and artificial intelligence, build international consensus on data standards, cross-population validation, and data governance, and gradually construct predictable and simulatable digital twins of human health. Professor Wang Weiqing, director of the Department of Endocrine and Metabolic Diseases at Ruijin Hospital Affiliated to Shanghai Jiao Tong University School of Medicine, introduced the research foundation and latest progress of the China Metabolic Analytics Project (ChinaMAP). The ChinaMAP official website and digital hub deeply integrate deep whole-genome sequencing data from more than 10,000 cases, large prospective cohort studies, landmark national multicenter RCT studies, and nationally representative metabolic disease surveillance data. Through this digital shared portal, ChinaMAP has achieved centralized display of data results across regions, studies, and omics, creating an open ecosystem for the global scientific research community. The open sharing of this major metabolic disease research resource will provide core strategic support for revealing the metabolic disease pathogenesis mechanisms unique to the Chinese population, formulating precision intervention strategies, and promoting the leapfrog development of China's precision medicine endeavor. In the morning's keynote and thematic report sessions, Michelle A. Williams, member of the National Academy of Medicine and professor at Stanford University, shared population health research from pregnancy to midlife, introducing five studies conducted based on maternal-infant intelligent agents. Joseph C. Wu, member of the National Academy of Medicine, member of the American Academy of Arts and Sciences, and director of the Stanford Cardiovascular Institute, shared research methods from genomic medicine to "clinical trials in a petri dish," introduced related research approaches, and explored new paths for medical research. Ziad Obermeyer, director of the UC Berkeley Medical Artificial Intelligence Laboratory, shared how, from a data science perspective, to move basic research toward the clinic and conduct research exploration starting from real clinical problems. Eric K. Oermann, director of the Health AI Research Laboratory at New York University, delivered a keynote report on the scientific evaluation of medical artificial intelligence, extending the focus from model capabilities to evaluation methods and sharing his thoughts on research and validation of medical artificial intelligence. In the afternoon of the same day, the keynote and thematic report sessions were held under the chairmanship of Bi Yufang, vice president of Shanghai Jiao Tong University School of Medicine and director of the Shanghai Institute of Endocrine and Metabolic Diseases. Multiple experts combined their respective research directions to share perspectives from medical engineering, patient prediction, population health, and proteomics research. Philip Awadalla, professor of genetics at the University of Oxford, took population multi-omics and cellular and network modeling as his theme, shared related research progress, and provided the attendees with an academic perspective for understanding health data and modeling research at different levels. Gao Wei, professor in the Department of Medical Engineering at Caltech, delivered a thematic report on flexible materials and biosensing technology, introducing achievements and research on skin-mounted wearable biosensors and non-invasive continuous detection of metabolites and biomarkers through body fluids such as sweat. Alexandre Loupy, director of the Paris Institute for Transplantation and Organ Regeneration at Université Paris Cité, shared related research on multimodal patient prediction systems and introduced the process by which such systems move from research to validation and evaluation. In the thematic discussion session, experts and editors from the Korea Institute of Science and Technology, Duke-NUS Medical School, Sichuan Academy of Medical Sciences and Sichuan Provincial People's Hospital, Nature Medicine, Nature Cancer, NEJM AI, and Cell engaged in face-to-face exchange and discussion, discussing the clinical trustworthiness of cutting-edge medical artificial intelligence technologies and their evidence base, with a focus on mechanistic explanation in data-driven research, the reliability of models across different populations, and the validation and evaluation required for clinical application, exchanging views from the perspectives of research methods, evidence quality, and academic dissemination.