Cancer Cluster Morphology

项目描述（申请人提供）：SBIR一期项目将开发准确检测和描述癌症集群和热点的方法、软件和教学材料。目前最先进的技术通常依赖于十多年前定义的方法，这些方法不能准确描述簇的形态（例如形状、超额风险和边界），并且已知会将低风险区域错误地分类为癌症簇的一部分。虽然现在有经过验证的方法可用于准确检测聚集性和量化其形态，但在地方和州卫生部门进行聚集性调查的一线公共卫生专业人员很少使用这些技术。这有几个原因。首先，许多这样的专业人员不知道新方法，不知道它们相对于现有方法的优势，也不知道如何使用它们。其次，直观易用的商业级软件在很大程度上是不可用的。最后，缺乏针对这些一线应急人员的教育举措，包括易于理解的短期课程、基于网络的教育工具和应用研究。这项研究将提出解决这些需求的方法、软件和教育计划。在第一阶段，我们将开发准确描述癌症聚类形态的方法和原型软件，并将使用已知聚类真实程度的模拟数据评估基于分类准确性，I型和II型错误的技术。在第二阶段，我们将全面开发软件，建立针对公共卫生专业人员的教育课程工具，并举办一系列培训讲习班，以迅速传播知识和工具。相关性：这项研究的科学和技术创新有望彻底改变我们的能力，使我们能够基于对癌症集群形态的准确理解，做出合理的癌症监测和控制决策。

DESCRIPTION (provided by applicant): This SBIR Phase I project will develop methods, software and instructional materials for the accurate detection and description of cancer clusters and hotspots. Present state-of-the-art often relies on methods that were defined over a decade ago, do not accurately describe cluster morphology (e.g. shape, excess risk, and boundaries) and that are known to misclassify areas of low risk as part of a cancer cluster. While proven methods are now available for accurately detecting clusters and for quantifying their morphology, these techniques are rarely used by front-line public health professionals who are undertaking cluster investigations in local and state health departments. This arises for several reasons. First, many such professionals are not aware of the new methods, their advantages over present approaches, nor of how to use them. Second, enabling commercial-grade software that is intuitive and easy to use is largely unavailable. Finally, educational initiatives are lacking that target these front-line responders with easy to understand short courses, web-based educational tools, and applied studies. This research will put in place methods, software, and educational programs that address these needs. In Phase I, we will develop methods and prototype software for accurately describing cancer cluster morphology, and will evaluate the techniques based on classification accuracy, type I and type II error using simulated data for which the true extent of clustering is known. In Phase II, we will fully develop the software, establish educational course tools targeting public health professionals, and conduct a series of training workshops to quickly disseminate knowledge and tools. Relevance: The scientific and technological innovations from this research are expected to revolutionize our ability to make sound cancer surveillance and control decisions based on an accurate understanding of the morphology of cancer clusters.