Federated digital pathology platform for AD/ADRD research and diagnostics

用于 AD/ADRD 研究和诊断的联合数字病理学平台

• 批准号: 10734939
• 负责人: Cody Bumgardner
• 金额: $ 177.9万
• 依托单位: UNIVERSITY OF KENTUCKY
• 依托单位国家: 美国
• 财政年份: 2023
• 资助国家: 美国
• 起止时间: 2023-09-20 至 2025-08-31
• 项目状态: 未结题
• 来源: https://reporter.nih.gov/project-details/10734939
• 关键词: African; African ancestry; Age; Alzheimer' s disease related dementia; Area; Artificial Intelligence; Autopsy; Benchmarking; Brain; Classification; Clinical; Collection; Communities; Computer software; Correlation Studies; Data; Data Aggregation; Data Analyses; Data Element; Data Management Resources; Data Set; Diagnosis; Diagnostic; Disease; Environmental Exposure; Evaluation; FAIR principles; Generations; Hand; Human; Image; Image Analysis; Individual; Institution; Kentucky; Machine Learning; Metadata; Methods; Modeling; Pathologic; Pathology; Performance; Population; Population Heterogeneity; Privatization; Procedures; Process; Recording of previous events; Registries; Reproducibility; Research; Research Design; Research Personnel; Resources; Sampling; Scanning; Secure; Services; Site; Slide; Source; Specific qualifier value; Standardization; System; Tauopathies; Training; Traumatic Brain Injury; Universities; Washington; Work; Workload; advanced analytics; annotation system; cohort; data curation; data exchange; data harmonization; data modeling; data sharing; demographics; digital; digital imaging; digital pathology; distributed data; experimental study; federated data; imaging Segmentation; insight; machine learning pipeline; multimodal data; multimodality; neuropathology; novel; open source; protein TDP-43; racial diversity; repository; tau Proteins; tau mutation; tool; white matter; whole slide imaging

Project Summary/Abstract We will connect multiple Alzheimer’s disease and related dementias (AD/ADRD) research centers for optimized and standardized whole slide image (WSI) advanced analytics. We propose these Specific Aims: Specific Aim 1: Generate a federated platform for data sharing and analysis of human digital neuropathological (DNP) slides. Specific Aim 1a: Develop an open-source platform to aggregate data distributed across multiple repositories into a central registry portal. This subaim will follow federated data curation, harmonization, annotation, and standardization employing FAIR (findable, accessible, interoperable, and reusable) principles. Specific Aim 1b: Develop a federated data curation and management system. This subaim provides methods to curate physical data, such as digital images, across federated sites. WSI data and metadata will be shared with private repositories and high-performance clusters. Specific Aim 2: Develop and demonstrate a platform for federated machine learning/artificial intelligence (ML/AI), result evaluation, and central project information, constituting a management hub for AD/ADRD studies. Specific Aim 2a: Develop a federated data annotation and automated AI/ML processing system. This subaim provides methods for users to prepare AI-ready multimodal (WSIs, metadata, demographics, etc.) datasets from distributed sources and conduct multi-site data transfer and federated training. Services and tools developed in Aim 1 will be used to programmatically generate and populate user-defined AI/ML pipelines. Specific Aim 2b: Develop an open-source platform for the generation and review of datasets and associated models. This subaim will provide a project evaluation portal integrating cohort, dataset, and model training results in a unified interface. A public-facing model hub will provide project, data, specifications, and model data to support data sharing and analysis requirements. To optimize and demonstrate the strengths of the novel federated network, five integrated projects are proposed. These will span a diverse sampling of human populations and diseases to leverage the unique strengths of DNP. The following sites will contribute resources, expertise, experimental study design, data, and analyses: University of Kentucky (PIs Nelson and Bumgardner, Co-Is Cheung and Fardo). Generate standardized SOPs useful for DNP diagnoses and research along with ML/AI expertise for a federated network. Northwestern/Nun Study and UTSA Universities (PI Flanagan). A focus on TDP-43 proteinopathy in community-based cohorts and ethno-racially diverse populations. Universityof São Paulo and UCSF (Co-I Suemoto and OSC Grinberg). Evaluate ADNC from the brains of individuals with African and non-African ancestries. University of Toronto (Co-I Kovacs): Use ML to study white matter tau pathologic changes for novel insights into multiple tauopathies. University of Washington (Co-I Keene): Study tau pathologies across ages and environmental exposures with a focus on traumatic brain injuries.

项目概要/摘要 我们将连接多个阿尔茨海默病和相关痴呆症 (AD/ADRD) 研究中心，以实现优化和标准化的全幻灯片图像 (WSI) 高级分析。 我们提出以下具体目标： 具体目标 1：生成用于人类数据共享和分析的联合平台。具体目标 1a：开发一个开源平台，将分布在多个存储库中的数据聚合到中央注册门户中。采用 FAIR（可查找、可访问、可互操作和可重用）原则的协调、注释和标准化 具体目标 1b：开发联合数据管理和管理系统 该子目标提供跨联合站点管理物理数据（例如数字图像）的方法。 WSI 数据和元数据将与私有存储库和高性能集群共享。具体目标 2：开发和演示联合机器平台。学习/人工智能 (ML/AI)、结果评估和中央项目信息，构成 AD/ADRD 研究的管理中心。 具体目标 2a：开发联合数据注释和自动化 AI/ML 处理系统。用户从分布式来源准备人工智能就绪的多模式（WSI、元数据、人口统计等）数据集，并进行多站点数据传输和目标 1 中开发的联合培训。用于以编程方式生成和填充用户定义的 AI/ML 管道。 具体目标 2b：开发一个用于生成和审查数据集及相关模型的开源平台。该子目标将提供一个集成队列、数据集和模型的项目评估门户。面向公众的模型中心将提供项目、数据、规范和模型数据，以支持数据共享和分析需求。为了优化和展示新型联邦网络的优势，提出了五个集成项目。 .这些将跨越对人群和疾病进行多样化采样，以利用 DNP 的独特优势 以下站点将提供资源、专业知识、实验研究设计、数据和分析：肯塔基大学（PI Nelson 和 Bumgardner、Co-Is Cheung 和 Fardo）。为西北/尼恩研究和 UTSA 大学 (PI Flanagan) 联合网络生成适用于 DNP 诊断和研究的标准化 SOP，以及 ML/AI 专业知识。圣保罗大学和加州大学旧金山分校（Co-I Suemoto 和 OSC Grinberg）对具有非洲和非非洲血统的个体的大脑进行 ADNC 评估。 )：使用机器学习研究白质 tau 病理变化，以获得对多种 tau 病的新见解华盛顿大学 (Co-I Keene)：研究跨 tau 病理学。年龄和环境暴露，重点关注创伤性脑损伤。