Wang Wenxia, Zou Wei, Hu Danfeng, Wang Jiajun
1 School of Electronic and Information Engineering, Soochow University, Suzhou, China.
2 College of Information Engineering, Henan University of Science and Technology, Luoyang, China.
Proc Inst Mech Eng H. 2018 Mar;232(3):215-229. doi: 10.1177/0954411917752026. Epub 2018 Jan 9.
Meshes play a crucial role in determining the accuracy of the elastic modulus reconstruction in the elastography when the finite element method is employed. In this article, we propose an adaptive mesh refinement strategy which can ensure the coincidence of the meshes with the shape of the inclusions in the observed tissue. This strategy is based on the intensity distribution of the strain image where the variance of the strain distribution in each element of the meshes is used to measure the homogeneity of the element, that is, the larger the strain variance is the more inhomogeneous the element will be and hence more detailed information will be included in this element. For more accurate reconstruction of such detailed information, mesh refinement procedure is implemented in such elements. Besides, two refinement steps are employed for the reconstruction to improve the fitness of the reconstructed image and the observed tissue. Simulation results show that the two-stage adaptive mesh refinement algorithm performs well without needing any prior information about the internal geometric shape in tissue. Not only Young's moduli of models but also shapes of the inclusions can be reconstructed perfectly and quickly with our proposed method.
在采用有限元方法进行弹性成像时,网格在确定弹性模量重建的准确性方面起着至关重要的作用。在本文中,我们提出了一种自适应网格细化策略,该策略可以确保网格与观察组织中内含物的形状相吻合。此策略基于应变图像的强度分布,其中网格每个单元中的应变分布方差用于衡量该单元的均匀性,即应变方差越大,该单元的不均匀性就越大,因此该单元将包含更详细的信息。为了更准确地重建此类详细信息,在这些单元中实施网格细化过程。此外,重建采用两个细化步骤以提高重建图像与观察组织的拟合度。模拟结果表明,两阶段自适应网格细化算法无需关于组织内部几何形状的任何先验信息即可良好运行。使用我们提出的方法,不仅可以完美且快速地重建模型的杨氏模量,还可以重建内含物的形状。