Development and Dissemination of Robust Brain MRI Measurement Tools

摘要：本申请响应RFA: PAR-07-249，“与国家生物医学计算中心合作”。该项目的目标是开发并广泛分发一个软件包，用于使用计算神经解剖学方法在MR图像中对大脑结构进行鲁棒测量。该项目将与国家医学图像计算联盟（NA-MIC）合作，使用NA-MIC软件工程流程开发软件，利用NA-MIC工程基础设施，并将该软件集成到3D切片器中，这是一个在NA-MIC中开发的架构良好的应用环境。这个特殊的软件包将包括一个名为HAMMER的脑图像配准和扭曲算法，以及一个用于分割白质病变（WMLs）的算法，白质病变可能由多种病理引起，包括血管病理和多发性硬化症。HAMMER获得了IEEE信号处理学会2006年最佳论文奖。HAMMER已成功应用于多个大型临床研究和临床试验，涉及5000多张脑磁共振图像，被来自20多个国家102个机构的318名用户下载。WML分割算法已成功应用于“Action to Control Cardiovascular Risk in Diabetes- memory in Diabetes”（ACCORD-MIND）子研究，数据来自4个中心，共650例患者，历时8年。为这两种算法设计一个易于使用、功能强大的软件包，并将其整合到3D切片机中，将使大量需要在各种神经成像研究中使用先进图像分析方法的最终用户受益。为了提高算法对临床图像数据高度可变的质量和特征的鲁棒性，需要进一步的算法开发。为了增加非计算机分析方法专家的易用性，并将该软件集成到Slicer平台中，计划进行重大的软件工程工作。将调查三个目标。第一个目标是进一步发展和扩展新的图像分析方法，旨在提高HAMMER配准和WML分割算法的鲁棒性和性能，使其易于应用于各种临床研究。第二个和第三个目标是为这两个算法设计单独的软件模块，并将它们合并到3D切片器中。这两个模块将被设计为(1)具有一致的跨平台交互和脚本化界面，(2)允许最终用户以交互方式探索适合其数据的参数，(3)使开发人员能够添加新功能。这两个模块的稳健性将通过软件工程工具和各种临床研究数据（从不同中心获得）进行广泛的测试和改进。最终的软件将免费提供源代码和预编译程序。公共卫生相关性：该项目的目标是开发并广泛分发一个软件包，用于使用计算神经解剖学方法在MR图像中健壮地测量大脑结构。

DESCRIPTION (provided by applicant): Development and Dissemination of Robust Brain MRI Measurement Tools Abstract: This application responds to RFA: PAR-07-249, "Collaborations with National Centers for Biomedical Computing". The goal of this project is to develop and widely distribute a software package for robust measurement of brain structure in MR images by using computational neuroanatomy methods. The project will collaborate with the National Alliance for Medical Image Computing (NA-MIC) to develop the software using the NA-MIC Software Engineering Process, leverage the NA-MIC engineering infrastructure, and integrate this software into the 3D Slicer, a well-architected application environment being developed in NA-MIC. The particular software package will include both a brain image registration and warping algorithm, called HAMMER, and an algorithm for the segmentation of white matter lesions (WMLs), which can arise from a variety of pathologies including vascular pathology and multiple sclerosis. HAMMER received 2006 Best Paper Award from IEEE Signal Processing Society. HAMMER has been successfully applied to many large clinical research studies and clinical trials involving over 5,000 MR brain images and has been downloaded by 318 users from 102 institutions in over 20 countries. The WML segmentation algorithm has been successfully applied to "Action to Control Cardiovascular Risk in Diabetes-Memory in Diabetes" (ACCORD-MIND) sub- study, with data acquired from 4 centers on 650 patients over a period of 8 years. Designing an easy-to-use, robust software package for these two algorithms and incorporating it into the 3D Slicer will benefit a large community of end-users that need access to advanced image analysis methods in various neuroimaging studies. To increase the robustness of the algorithms to the highly variable quality and characteristics of clinical image data, further algorithm development is necessary. To increase ease of use by non-experts in computer analysis methods and integrate this software into the Slicer platform, significant software engineering efforts are planned. Three aims will be investigated. The first aim is to further develop and extend novel image analysis methods aiming at improving the robustness and performance of HAMMER registration and WML segmentation algorithms, so that they can be easily applied to various clinical research studies. The second and third aims are to design separate software modules for these two algorithms, and to incorporate them into the 3D Slicer. These two modules will be designed (1) with consistent cross-platform interactive and scripted interfaces, (2) allowing end-users to interactively explore the suitable parameters for their data, (3) enabling developers to add new functions. The robustness of these two modules will be extensively tested and improved by both software engineering tools and various clinical research data (acquired from different centers). The final software will be freely available in both source code and pre-compiled programs. PUBLIC HEALTH REVELANCE: The goal of this project is to develop and widely distribute a software package for robust measurement of brain structure in MR images by using computational neuroanatomy methods.