Bone mechanical strength estimation using complexity analysis of bones

本研究项目的目的是开发一种基于骨骼复杂性分析的骨骼机械强度评估方法。骨骼的机械强度包括骨骼的质量、数量和结构。因此，在项目开始时，我们选择了micro - X-ray CT作为数量和结构的评估，选择了声学仪器作为质量的评估。在项目的中期，我们发现在微x射线CT图像的伪影减少和声学仪器的基本发展方面很重要，因此我们建立了项目的两个子主题。项目成果如下：1)伪影约简：伪影约简算法基本上是两个步骤的迭代，即在正弦图空间进行自适应滤波和在图像空间进行阈值分割。该算法有效地减少了伪影。经过伪影还原后，对微x射线CT图像进行非线性矢量量化和线性识别。结果表明，除5个参数中的1个参数外，机械强度的相关系数均在0.8左右。声学仪器我们评估非线性波形畸变的思路是发射和接收不同初始相位的超声加伯波。然而，我们最近发现了仪器精度的困难，因此我们通过计算机模拟验证了这一想法的有效性

The aim of this research project is to develop an evaluation method of mechanical strength of bones using complexity analysis of bones. The mechanical strength of bones consists of quality, quantity and structure of bones. Therefore, in the beginning of the project, we chose micro X-ray CT for evaluation of quantity and structure and acoustical instrumentation for quality. In the middle of the project, we found important in artifact reduction of micro X-ray CT images and basic development of the acoustical instrumentation, and we thereby build the two subthemes of the project. The achievements of the project are as follows.1) Artifact reduction :The artifact reduction algorithm was basically iteration of two steps, namely an adaptive filtering m sinogram space and a thresholding in image space. The artrfact was efficiently reduced with the algorithm.2) micro X-ray CTAfter the artifact reduction, micro X-ray CT images were evaluated by a nonlinear vector quantization and a linear recognition. As a result, the mechanical strength was successiully predicted with correlation coefficient around 0.8 except one out of five parameters.3) Acoustical instrumentationOur idea for evaluation of nonlinear waveform distortion was transmitting and receiving ultrasonic Garbor waves with different initial phases. However, we newly found difficulty in instrumentation accuracy, so that we validated the effectiveness of this idea by computer simulation