Delivering large scale digital clinical trials for artificial intelligence

为人工智能提供大规模数字化临床试验

• 批准号: 10065901
• 金额: $ 6.26万
• 依托单位: NEWTON'S TREE LTD
• 依托单位国家: 英国
• 项目类别: Collaborative R&D
• 财政年份: 2023
• 资助国家: 英国
• 起止时间: 2023 至 无数据
• 项目状态: 已结题
• 来源: https://gtr.ukri.org/projects?ref=10065901
• 关键词: Delivering; large; scale; digital; clinical

The project aims to speed up the deployment, evaluation, and commissioning of AI in healthcare by creating a near real-time evaluation toolkit for AI applications. The cost of clinical trials for AI applications can be prohibitively expensive for small and medium-sized businesses, making it difficult for them to prove the safety and efficacy of their products.To address this issue, the project has three main goals: to determine the appropriate performance metrics for identifying safe and trustworthy AI applications in healthcare, to gather feedback from clinical users in near real-time without interrupting their workflows, and to generate evidence of efficacy at a large scale while maintaining data privacy.The project will use mixed methods research and a privacy-preserving federated AI software evaluation toolkit to achieve these goals, and once completed, will be used to evaluate AI at a large scale with the help of SMEs, healthcare providers, and commissioners.

该项目旨在通过为AI应用程序创建一个近实时的评估工具包，加快AI在医疗保健领域的部署、评估和调试。人工智能应用的临床试验成本对于中小型企业来说可能非常昂贵，这使得他们很难证明其产品的安全性和有效性。为了解决这个问题，该项目有三个主要目标：为了确定适当的性能指标，以确定医疗保健中安全和值得信赖的AI应用程序，为了在不中断临床用户的工作流程的情况下近乎实时地收集他们的反馈，并在保持数据隐私的同时大规模地产生功效证据。该项目将使用混合方法研究和隐私-保留联合人工智能软件评估工具包以实现这些目标，一旦完成，将在中小企业、医疗保健提供商和专员的帮助下用于大规模评估人工智能。