Enhanced fracture risk assessment of spine using stochastically treated DXA image

描述（由申请人提供）：脊柱骨折是最常见的脊椎骨折类型，影响50岁以上三分之一的女性和六分之一的男性。众所周知，通过骨矿物质密度量化的骨量损失与骨折风险增加相关。然而，骨密度本身并不能完全解释骨折风险的变化。除了骨量，骨结构已被确定为骨折风险的另一个关键因素。尽管近年来已经取得了相当大的进展，显示3D成像，如micro-CT，pQCT和micro-MRI，可以提供与骨脆性相关的结构信息，但这些技术在常规临床应用中仍然是不切实际的。因此，如果我们能找到有用的参数，这是与建筑信息，从骨密度的空间分布在2D图像的双能X射线吸收（DXA）扫描，这将是有希望的，利用一个简单的DXA扫描来评估骨脆性的基础上测量的骨密度和分布。我们的长期目标是开发技术，从临床可行的措施高度准确地预测脊柱骨折。本申请的目的是确定如果使用新型随机图像处理方法进行增强，来自人体脊柱DXA扫描的2D图像的骨矿物质密度分布是否可用于提供骨脆性的额外测量。本申请的中心假设是，从DXA扫描的2D图像量化的骨矿物质密度的空间分布与脊柱的结构特性相关，通过将这种密度分布测量与DXA骨矿物质密度数据相结合，从而显著改善了对骨脆性的预测。我们的假设已经制定了强有力的初步数据的基础上，这表明，随机场理论可以用来量化骨密度的空间分布和随机模型中定义的参数显着相关的微结构和强度的骨小梁。将追求两个具体目标来检验中心假设并实现本申请的目标。在具体目标1中，我们将确定DXA脊柱图像的骨矿物质密度空间分布的随机参数与脊柱微结构的相关性。特定目标1的工作假设是，人体脊柱2D DXA图像的基值方差（骨矿物质密度空间分布的一种度量）与骨小梁3D微CT图像的骨微结构相关。在具体目标2中，我们将确定增强DXA方法在预测骨脆性方面的有效性。据推测，从DXA扫描的2D脊柱图像导出的骨矿物质密度的空间分布的量化，结合骨矿物质密度，将比单独使用骨矿物质密度更好地预测骨强度。在这些研究完成后，我们预计，一个经济和有效的方法来评估脊柱骨折的风险将建立从二维图像的DXA扫描。我们预计，这种方法可以改善骨折风险的预测和对治疗反应的监测。此外，该项目将通过提供调查人员进行独立研究的机会，并为学生提供生物医学研究的经验和参与，加强受赠机构的研究环境。 公共卫生相关性：在所有50岁以上的妇女中，35%至50%至少有一处脊柱骨折。因此，识别人群中的高危人群并减少脊柱骨折的数量至关重要。本项目的重点是通过结合骨密度及其分布的测量，提高使用DXA密度计预测脊柱骨折风险的准确性。

DESCRIPTION (provided by applicant): Spine fractures are the most common type of osteoporotic fractures, affecting one in three women and one in six men over the age of 50. It is well known that loss of bone mass, quantified by bone mineral density, is associated with the increasing risk of bone fractures. However, bone mineral density alone cannot fully explain changes in fracture risks. In addition to bone mass, bone architecture has been identified as another critical factor to fracture risk. Although considerable progress has been made in recent years, showing that 3D imaging, such as micro-CT, pQCT and micro-MRI, can provide the architectural information related to bone fragility, these techniques are still impractical in routne clinical applications. Thus, if we can find useful parameters, which are associated with architectural information, from the spatial distribution of bone mineral density in 2D images of Dual-energy X-ray absorptiometry (DXA) scans, it would be promising to utilize a simple DXA scan to assess bone fragility based on the measurements of both bone mineral density and distribution. Our long-term goal is to develop techniques for highly accurate prediction of spine fractures from clinically feasible measures. The objective of this application is to determine whether the distribution of bone mineral density from 2D images of DXA scans of human spines can be used to provide additional measures of bone fragility if enhanced using a novel stochastic image processing approach. The central hypothesis of this application is that the spatial distribution of bone mineral density quantified from 2D images of DXA scans is associated with the architectural properties of the spine, leading to significantly improved prediction of bone fragility by combining this measure of density distribution with DXA bone mineral density data. Our hypothesis has been formulated on the basis of strong preliminary data, which have shown that random field theory can be used to quantify the spatial distribution of bone mineral density and that the parameters defined in the stochastic model are significantly correlated with both microarchitecture and strength of trabecular bone. Two specific aims will be pursued to test the central hypothesis and accomplish the objective of this application. In specific aim 1, we will determine the correlation of the stochastic parameters of spatial distribution of bone mineral density of the DXA spine images with the microarchitecture of the spine. The working hypothesis for specific aim 1 is that the sill variance, a measure of spatial distribution of bone mineral density, from 2D DXA images of human spine is associated with bone micro-architecture from 3D micro-CT images of trabecular bone. In specific aim 2, we will determine the efficacy of the enhanced DXA approach in predicting bone fragility. It is postulated that quantification of spatial distribution of bone mineral density derived from 2D spine images of DXA scans, combined with bone mineral density, will predict bone strength better than using bone mineral density alone. At the completion of these studies, we anticipate that an economical and effective method for assessing the risk of spine fractures will be established from 2D images of DXA scans. We anticipate that this method could lead to improved prediction of fracture risk and monitoring of response to treatment. Additionally, this project will strengthen the research environment at the grantee institution by providing investigators opportunities to carry out independent research, and offering students experience and involvement in biomedical research. PUBLIC HEALTH RELEVANCE: Between 35% and 50% of all women over age 50 had at least one spine fracture. Therefore, it is critical to identify those at highest risk in the populaion and reduce the number of spine fractures. This project focuses on improving the accuracy of predicting fracture risk of spine using DXA densitometers by combining measures of bone mineral density and its distribution.