Resources for development and validation of Radiomic analyses & Adaptive Therapy

放射组学分析的开发和验证资源

• 批准号: 8751514
• 负责人: Fred William Prior
• 金额: $ 63.15万
• 依托单位: WASHINGTON UNIVERSITY
• 依托单位国家: 美国
• 财政年份: 2014
• 资助国家: 美国
• 起止时间: 2014-09-01 至 2019-08-31
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/8751514
• 关键词: Accounting; Address; Algorithms; Big Data; Biological Markers; Clinical; Collaborations; Collection; Communities; Community Developments; Community Networks; Computer software; Data; Data Analyses; Data Collection; Data Element; Data Set; Development; Disease; Dose; Educational workshop; Environment; Gene Proteins; Generations; Goals; Grant; Graph; Health; High Performance Computing; Human; Image; Image Analysis; Imagery; Information Resources; Link; Lung; Malignant Neoplasms; Measures; Methodology; Metric; Mining; Modality; Outcome; Phase; Phenotype; Population Analysis; Process; Radiation Oncology; Radiation therapy; Radiotherapy Research; Recording of previous events; Records; Research; Research Infrastructure; Research Personnel; Resource Development; Resources; Retrieval; Sample Size; Secure; Services; Site; Statistical Data Interpretation; System; Techniques; Testing; Therapeutic Studies; Time; Training; Training Support; Translational Research; Universities; Validation; Washington; Work; anticancer research; base; cancer imaging; cancer radiation therapy; cancer therapy; computing resources; data hosting; data modeling; data structure; experience; image archival system; image processing; imaging informatics; improved; member; novel; open source; outcome forecast; public health relevance; repository; response; screening; software development; statistics; success; tool; treatment response; tumor

DESCRIPTION (provided by applicant): Imaging based cancer research is in the beginning phases of a transition from analyses based on human observers to the use of advanced computing platforms and software to automatically extract large sets of quantitative image features relevant to prognosis or treatment response. These feature sets can be used to infer phenotypes or correlate with gene-protein signatures. In parallel, radiation oncology is responding to the quantitative requirements of adaptive therapy, where therapy decisions take into account the early response of the tumor to the treatment. Both of these advanced techniques share similar resource requirement, in particular high capacity information repositories co-located with high performance computing capabilities and tools to perform advanced analytics. Drawing on broad experience in imaging informatics we will expand upon the existing Cancer Imaging Archive (TCIA) platform and computational resources of the Washington University Center for High Performance Computing to provide leading edge research services to Quantitative Imaging Network (QIN) researchers. To facilitate this translational research and share our processes and experiences with the academic cancer research community, we propose to: 1. Provide data hosting, management and access capabilities to QIN researchers including longitudinal data collections and advanced Radiation Therapy (RT) data structure support; 2. Develop a QIN portal to support advanced quantitative image processing and biomarker validation through co-located scalable computing and big data infrastructure; 3. Support biomarker development and validation with advanced analytics that employ a new generation of statistical tools and data modeling techniques; 4. Provide training and support to permit QIN researchers to effectively utilize this suite of advanced capabilities combined with open source community enablement tools. We have assembled a strong team with a long history of collaboration and extensive experience with open source software development methodologies and advanced imaging informatics. The long-term goal of our team is to develop and deploy software and services to drive advanced quantitative image analysis and biomarker development and provide a gateway for researchers to interrogate multi-modality datasets and mine them to create and validate novel hypotheses in cancer research.

描述（由申请人提供）：基于成像的癌症研究正处于从基于人类观察者的分析向使用先进的计算平台和软件自动提取与预后或治疗反应相关的大量定量图像特征的过渡的开始阶段。这些特征集可用于推断表型或与基因-蛋白质签名相关。与此同时，放射肿瘤学正在响应适应性治疗的定量要求，其中治疗决策考虑到肿瘤对治疗的早期反应。这两种先进技术共享类似的资源需求，特别是高容量信息存储库与高性能计算能力和工具共存以执行高级分析。 凭借在成像信息学方面的丰富经验，我们将扩展现有的癌症成像档案（TCIA）平台和华盛顿大学高性能计算中心的计算资源，为定量成像网络（QIN）研究人员提供领先的研究服务。为了促进这种转化研究，并与学术癌症研究界分享我们的过程和经验，我们建议：1。为QIN研究人员提供数据托管、管理和访问功能，包括纵向数据收集和高级放射治疗（RT）数据结构支持; 2.开发一个QIN门户网站，通过协同定位的可扩展计算和大数据基础设施，支持高级定量图像处理和生物标志物验证; 3.通过采用新一代统计工具和数据建模技术的高级分析，支持生物标志物的开发和验证; 4.提供培训和支持，使QIN研究人员能够有效地利用这套先进的功能与开源社区支持工具相结合。 我们组建了一支强大的团队，拥有悠久的合作历史，在开源软件开发方法和先进的成像信息学方面拥有丰富的经验。我们团队的长期目标是开发和部署软件和服务，以推动先进的定量图像分析和生物标志物开发，并为研究人员提供一个网关，以查询多模态数据集并挖掘它们，以创建和验证癌症研究中的新假设。