Real-world-data Enabled Assessment for heaLth regulatory decision-Making

REALM项目的总体目标是为监管机构、应用程序开发人员、医疗保健专业人员和政策官员创建一个协作框架，以共同创建和评估医疗保健软件。我们建议创建一个包容性平台，为医疗保健软件的评估和认证提供一个透明的生态系统，开发人员和监管（和健康技术评估）机构都可以访问标准化的技术堆栈和数据。这将通过首先绘制和分析欧盟、国家机构和世界各地关于医疗保健实践中软件的法规、立法工作和指导方针来实现。这将指导路线图建立一个包容、公平和多利益攸关方的生态系统。集成架构的支架将与达尔文合作开发，基于标准化数据模型和优化的数据驱动方法，以有效利用医疗监管实践中的真实世界数据（RWD）。该架构将由四个部分组成：两个技术基础设施，一个生活实验室和一个上市后监测模块。i）将建立基于云的联合数据资源目录，将当前可用的RWD数据和合成数据汇集在一起，以满足平台的数据需求。ii）将建立监管机构，汇集标准化工具，以培训、测试、评估和监测医疗/保健软件。iii）用于试验医疗/保健软件技术评估的生活实验室环境，将人机交互作为系统的一部分。iv）与RWD进行上市后监督，以确保经认证软件的质量标准。最后，基于所提出的架构，将在3个国家（荷兰、比利时和希腊）实施5个真实的医疗器械软件（MDSW）项目。

The overall aim of the REALM project is to create a collaborative framework for regulatory authorities, application developers, healthcare professionals and policy officers to co-create and evaluate software for medical and healthcare use. We propose to create an inclusive platform leading to a transparent ecosystem for evaluation and certification of software in healthcare where both the developers and the regulatory (and Health Technology Assessment) bodies have access to a standardized set of technology stack and data. This will be achieved by first mapping and analyzing regulations, legislative efforts and guidelines from EU, national bodies and around the world on software in healthcare practice. These will guide the roadmap towards building an inclusive, fair and multi-stakeholder ecosystem. The scaffold for an integrated architecture will be developed in collaboration with DARWIN, based on standardized data models and optimized data driven methodologies for the effective use of real-world data (RWD) in healthcare regulatory practice. The architecture will consist of four components: two technological infrastructures, a living lab and a post-marketing surveillance module. i) a federated cloud-based data resources catalog will be established, to bring together currently available RWD data and synthetic data to facilitate the data needs of the platform. ii) Regulatory Toolbox will be established to bring together standardized tools to train, test, evaluate and monitor medical/healthcare software. iii) Living lab environment for piloting medical/healthcare software technology assessment taking into account human-software interactions as part of the system. iv) Post-marketing surveillance with RWD for ensuring quality standards of the certified software in practice. Finally, building on the proposed architecture, five real Medical Device Software (MDSW) projects are going to be implemented across 3 countries (Netherlands, Belgium and Greece).