打通韩国医疗数据跨境研究壁垒，支持安全高效协作。
NSTRI Global Collaborative Research Data Platform
- 用医学嵌入和AI搜索实现跨数据集快速定位
- 整合10个韩医数据集，支持多模态分析
- 适合跨国医疗AI团队，兼顾隐私与效率
首尔国立大学医院（SNUH）运营的国家战略性技术研究院（NSTRI）数据平台，旨在解决国际研究者获取韩国医疗数据的难题。该平台提供经过去标识化的韩国医疗数据安全访问，并融合国际数据集，助力开发更具公平性与泛化能力的机器学习模型。系统包含四大核心AI组件：基于领域医学嵌入的智能数据搜索引擎、韩英医学术语翻译系统、全面的药物检索引擎以及大语言模型驱动的医学研究助手。平台采用容器化安全研究舱架构，保障数据隐私的同时提升研究效率。目前平台接入来自SNUH的10个不同医学数据集，按权限分级管理并统一标准，支持跨数据集分析。该基础设施为全球医疗研究协作提供安全合规的支撑。
原文摘要 · Abstract (English)
The National Strategic Technology Research Institute (NSTRI) Data Platform operated by Seoul National University Hospital (SNUH) addresses the challenge of accessing Korean healthcare data for international research. This platform provides secure access to pseudonymized Korean healthcare data while integrating international datasets, enabling the development of more equitable and generalizable machine learning models. The system features four key AI-powered components: an intelligent data search engine utilizing domain-specific medical embeddings, a Korean-English medical translation system, a comprehensive drug search engine, and an LLM-powered medical research assistant. The platform implements containerized environments within a secure research pod architecture, ensuring data protection while maintaining research efficiency. The platform currently provides access to 10 distinct medical datasets from SNUH, categorized by access permissions and standardized for cross-dataset analysis. This infrastructure enables global collaborative healthcare research while maintaining strict data protection standards.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。