Open-source Software Development Supplement for 3D quantitative analysisof mouse models of structural birth defects through computational anatomy

通过计算解剖学对结构性出生缺陷小鼠模型进行 3D 定量分析的开源软件开发补充

• 批准号: 10839199
• 负责人: DAVID R. BEIER
• 金额: $ 38.7万
• 依托单位: SEATTLE CHILDREN'S HOSPITAL
• 依托单位国家: 美国
• 财政年份: 2023
• 资助国家: 美国
• 起止时间: 2023-09-08 至 2025-12-31
• 项目状态: 未结题
• 来源: https://reporter.nih.gov/project-details/10839199
• 关键词: 3-Dimensional; Address; Adoption; Anatomy; Atlases; Automation; Biomedical Research; Charge; Cloud Computing; Communities; Complex; Computer Hardware; Computer Vision Systems; Computer software; Congenital Abnormality; Contracts; Data Science; Data Set; Detection; Development; Documentation; Engineering; Environment; Fetal Development; Fetus; Foundations; Frustration; Funding; Generations; Genetic; Goals; High Performance Computing; Human; Image Analysis; International; Knockout Mice; Knowledge; Libraries; Linux; Manuals; Measurement; Measures; Methodology; Microscope; Morphology; Mus; Operating System; Organ; Paper; Pharmaceutical Preparations; Phenotype; Population; Preparation; Pythons; Research; Research Personnel; Resources; Sampling; Science; Shapes; Software Engineering; Software Tools; Standardization; Structural Congenital Anomalies; Structural defect; Structure; Techniques; Three-Dimensional Image; Three-Dimensional Imaging; Toxin; United States National Institutes of Health; Variant; Visualization; Work; biomedical imaging; cluster computing; computational anatomy; computer science; data tools; deep learning; experience; fetal; genetic manipulation; graphical user interface; image registration; insight; interest; learning strategy; machine learning model; microCT; model organism; mouse model; novel; open source; parent grant; programs; response; software development; tool; user-friendly

The primary goal of this supplement is to develop user-friendly, UI driven modules as part of the open-source 3D biomedical imaging suite 3D Slicer to facilitate easier discovery of novel phenotypes in quantitative screens of model organisms that have been experimentally manipulated. We will convert our analytical pipelines based on deformable image registration and deep-learning methods, which are developed for our parent grant, into re- usable, general-purpose modules. The end-users of these tools need minimal experience with specifics of biomedical image analysis and/or computer vision concepts, thereby will be able to apply the tools directly to their specific question.

本补充的主要目标是开发用户友好的、UI 驱动的模块，作为开源项目的一部分 3D 生物医学成像套件 3D Slicer 有助于在定量筛选中更轻松地发现新表型 经过实验操作的模型生物。我们将基于以下方式转换我们的分析管道 关于可变形图像配准和深度学习方法，这些方法是为我们的家长资助而开发的，以重新 可用的通用模块。这些工具的最终用户需要最少的具体经验 生物医学图像分析和/或计算机视觉概念，从而能够将这些工具直接应用于 他们的具体问题。