Intelligent Applications Based on Medical Large Models to Enhance People's Healthcare Experience and Health Status
China
Healthcare
Implementing Organisation
Alipay (Hangzhou) Information Technology Co., Ltd. / Ant Group
China, Zhejiang/Shanghai, Hangzhou
Implementing Point of Contact
Nannan DU
Contact Person
Contributor of the Impact Story
International Telecommunication Union (ITU)
Year of implementation
2024
Problem statement
Patient experience is compromised due to the gap between supply and demand for medical resources. Complex administrative workflows limit doctors' clinical focus, with doctors spending approximately 70% of their time on administrative work and only 30% on clinical care. Uneven geographic distribution of medical resources affects accessibility and convenience of high-quality healthcare for residents in different regions.
Submission Overview
Alipay/Ant Group is a leading Chinese technology company specializing in digital payment and financial services. The company has developed a self-trained medical large language model with multimodal capabilities (text, image, audio, video), built on a data system containing hundreds of billions of tokens of professional medical texts and hundreds of millions of multimodal clinical documents.
AI Technology Used
RAG (Retrieval-Augmented Generation), Medical Knowledge Graphs, AI Agents, Multimodal Large Language Models, KG-CoT (Knowledge Graph Chain of Thought), PRM-MCTS, Reinforcement Learning, Diffusion Models, Neural Radiance Fields, and Voice Cloning
Key Outcomes
Access & Reach
Ant Group's medical large model addresses healthcare accessibility gaps by deploying an AI health assistant covering the full medical journey. The solution provides smart family doctor capabilities for residents, intelligent clinical/administrative assistants for doctors, and online specialist agents for underserved areas. Since launch in June 2024, it has reached over 13 million users, delivered 78+ million service instances, and achieved 98% user satisfaction. The system achieves 93% diagnostic accuracy and over 95% personalized service satisfaction, while integrating with 1,000+ healthcare facilities.
Impact Metrics
Number of Users Reached via Ant Group's AI solutions
Baseline Value
NA
Post-Implementation
Over 13 million users since launch in June 2024
Number of Service Instances Delivered via Ant Group's AI solutions
Baseline Value
NA
Post-Implementation
More than 78 million service instances were delivered
Diagnostic Accuracy of Ant Group's AI solutions
Baseline Value
NA
Post-Implementation
93 % diagnostic accuracy was recorded using KG-CoT and PRM-MCTS technologies.
Number of healthcare facilities integrated via Ant Group's AI solutions
Baseline Value
NA
Post-Implementation
Intelligent agents integrated across over 1,000 healthcare facilities
Implementation Context
Deployed across 1,000+ healthcare facilities in China via Zhejiang Provincial Health Commission, Hangzhou Municipal Medical Insurance Bureau, and Shanghai Jiao Tong University School of Medicine Renji Hospital
Residents with healthcare needs across all demographics, doctors in clinical practice, and patients in underserved and remote areas.
Key Partnerships
Zhejiang Provincial Health Commission, Hangzhou Municipal Medical Insurance Bureau, and Shanghai Jiao Tong University School of Medicine Renji Hospital
Replicability & Adaptation
1. Adapt medical knowledge base to local clinical guidelines and regulations 2. Localize language support beyond Chinese and English 3. Partner with local medical institutions for domain-specific fine-tuning 4. Ensure compliance with local data privacy and digital health regulations
Supporting Materials
* The data presented is self-reported by the respective organisations. Readers should consult the original sources for further details.