Developing Contractual Service Framework for Bone Histomorphometry

描述（由申请人提供）：本项目旨在开发能够自动化骨组织形态测量的软件技术，目前骨组织形态测量主要是通过繁琐和费力的交互过程手动完成的。我们的计划是使用最先进的图像处理技术来自动检测、测量和定量生物效应，直接对从显微设备获得的染色图像进行检测。我们计划采用三步方法进行自动定量：(i)改进我们的样品处理和成像步骤，并用传统的手动方法验证计算机推导的分析结果；（ii）使用一组更大的具有不同遗传背景的室内小鼠，开发自动组织形态学分析管道和数据库，以对研究人员有吸引力的格式保存原始数据和分析结果；（iii）由外部机构的四名研究人员对该策略进行beta测试，他们将样品提交到分析管道进行处理和分析。第三步将利用公司现有的可视化数据流框架并建立一个商业模型，该模型不仅提供最终图像分析结果的交付，而且还提供整个分析过程本身，以确保科学家的分析来源。我们提议的研发工作的潜在影响是巨大的。目前，内部组织形态测量的手工处理需要很长时间，容易出错，而且运行成本太高。此外，由于以内部为导向的个性化操作，将分析结果与社区范围内的金标准进行比较非常困难。我们的开发可以创建一个集中的、成本有效的服务企业，科学家可以通过它以成本快速获取测量数据，其格式可以很容易地重新检查和验证。此外，集中式服务可以存储大量已发表的“参考”图像和测量值，以便科学家可以更有效地将音型变化与参考文献进行比较。当这种集中的基于合同的图像处理被建立和广泛使用时，组织形态测量的成本将大大降低，而科学的质量将大大提高。公共卫生相关性：感知到的社会、教育和科学效益：我们提出的软件可能会彻底改变骨组织形态学。科学家将能够在更短的时间内进行图像分析，并提高量化质量。此外，使用感知服务模型，由于成本和速度的显著降低，科学家们将能够在比目前可用的更大的范围内利用组织形态计量学。至于对科学的具体影响，我们提出的自动化方法一旦建立，就可以作为一种高通量机制来评估所有转基因小鼠骨结构改变的证据，并观察骨结构的变化。

DESCRIPTION (provided by applicant): This project aims at developing software technology capable of automating bone histomorphometry, which at the moment is mostly done manually through cumbersome and laborious interactive processes. Our plan is to use state-of-the-art image processing techniques to automate detection, measurement and quantification of biological effects directly over the stained images acquired from microscopic devices. We plan to employ a three-step approach for the automated quantification: (i) refine our sample processing and imaging steps and validate the computer derived analytical results with the traditional manual approach; (ii) using a larger set of in house mice with different genetic backgrounds, develop automated histomorphometric analytical pipelines and a database to house the raw data and the analysis outcomes in a format that would be attractive to investigators; and (iii) beta test this strategy with four investigators at outside institutions who submit samples to the analytical pipeline for processing and analysis. The third step will utilize the company's existing visual dataflow framework and build a business model which offers not only the delivery of the final image analysis outcome, but also the entire analysis process itself to ensure analysis provenance to the scientists. The potential impact of our proposed R&D work is enormous. Currently the manual processing of in-house histomorphometry requires lengthy man hours, is prone to error and is becoming too costly to run. In addition, due to in-house oriented individualized operations, comparing the analysis outcome against community-wide gold standards has been very difficult. Our development can create a centralized, cost effective service enterprise with which scientists can rapidly acquire measurement data at cost in a format that can easily be re-examined and validated. Furthermore, the centralized service can store a large collection of published "reference" images and measurement values so that scientists can more efficiently compare phonotypical changes against the references. When this centralized contractual based image processing is established and widely-used, the cost of histomorphometry could become significantly lower, while the quality of science dramatically increases. PUBLIC HEALTH RELEVANCE: Perceived societal, educational, and scientific benefits: Our proposed software can potentially revolutionize bone histomorphometry. Scientists will be able to do image analysis in a much shorter time with an improvement in quantification quality. In addition, using the perceived service model scientists will be able to utilize histomorphometry at far greater scale than what is current available due to the significant reduction in cost and speed. As for the specific impact on the science, once established our proposed automated method could serve as a high throughput mechanism for evaluating all the genetically modified mice for evidence of altered bone structure and observing the change in bone architecture.