Department of Electrical Engineering, San Jose State University, San Jose, CA, USA.
Med Phys. 2012 Apr;39(4):1748-68. doi: 10.1118/1.3690467.
In developing breast imaging technologies, testing is done with phantoms. Physical phantoms are normally used but their size, shape, composition, and detail cannot be modified readily. These difficulties can be avoided by creating a software breast phantom. Researchers have created software breast phantoms using geometric and/or mathematical methods for applications like image fusion. The authors report a 3D software breast phantom that was built using a mechanical design tool, to investigate the biomechanics of elastography using finite element modeling (FEM). The authors propose this phantom as an intermediate assessment tool for elastography simulation; for use after testing with commonly used phantoms and before clinical testing. The authors design the phantom to be flexible in both, the breast geometry and biomechanical parameters, to make it a useful tool for elastography simulation.
The authors develop the 3D software phantom using a mechanical design tool based on illustrations of normal breast anatomy. The software phantom does not use geometric primitives or imaging data. The authors discuss how to create this phantom and how to modify it. The authors demonstrate a typical elastography experiment of applying a static stress to the top surface of the breast just above a simulated tumor and calculate normal strains in 3D and in 2D with plane strain approximations with linear solvers. In particular, they investigate contrast transfer efficiency (CTE) by designing a parametric study based on location, shape, and stiffness of simulated tumors. The authors also compare their findings to a commonly used elastography phantom.
The 3D breast software phantom is flexible in shape, size, and location of tumors, glandular to fatty content, and the ductal structure. Residual modulus, maps, and profiles, served as a guide to optimize meshing of this geometrically nonlinear phantom for biomechanical modeling of elastography. At best, low residues (around 1-5 KPa) were found within the phantom while errors were elevated (around 10-30 KPa) at tumor and lobule boundaries. From our FEM analysis, the breast phantom generated a superior CTE in both 2D and in 3D over the block phantom. It also showed differences in CTE values and strain contrast for deep and shallow tumors and showed significant change in CTE when 3D modeling was used. These changes were not significant in the block phantom. Both phantoms, however, showed worsened CTE values for increased input tumor-background modulus contrast.
Block phantoms serve as a starting tool but a next level phantom, like the proposed breast phantom, will serve as a valuable intermediate for elastography simulation before clinical testing. Further, given the CTE metrics for the breast phantom are superior to the block phantom, and vary for tumor shape, location, and stiffness, these phantoms would enhance the study of elastography contrast. Further, the use of 2D phantoms with plane strain approximations overestimates the CTE value when compared to the true CTE achieved with 3D models. Thus, the use of 3D phantoms, like the breast phantom, with no approximations, will assist in more accurate estimation of modulus, especially valuable for 3D elastography systems.
在开发乳腺成像技术时，使用体模进行测试。通常使用物理体模，但它们的大小、形状、组成和细节不易修改。通过创建软件乳腺体模可以避免这些困难。研究人员已经使用几何和/或数学方法创建了软件乳腺体模，用于图像融合等应用。作者报告了一种使用机械设计工具构建的 3D 软件乳腺体模，用于使用有限元建模 (FEM) 研究弹性成像的生物力学。作者提出该体模作为弹性成像模拟的中间评估工具；在使用常用体模进行测试后和临床测试之前使用。作者设计该体模在乳腺几何形状和生物力学参数方面具有灵活性，使其成为弹性成像模拟的有用工具。
作者使用基于正常乳腺解剖插图的机械设计工具开发 3D 软件体模。软件体模不使用几何基元或成像数据。作者讨论了如何创建此体模以及如何修改它。作者演示了一个典型的弹性成像实验，即在模拟肿瘤上方的乳房顶部表面施加静态应力，并使用线性求解器的平面应变近似值在 3D 和 2D 中计算正常应变。特别是，他们通过基于模拟肿瘤的位置、形状和刚度进行参数研究来研究对比传递效率 (CTE)。作者还将他们的发现与常用的弹性成像体模进行了比较。
3D 乳腺软件体模在形状、大小、肿瘤位置、腺体到脂肪含量以及导管结构方面具有灵活性。残余模量、图谱和剖面图作为指导，用于优化该几何非线性体模的网格划分，以进行弹性成像的生物力学建模。在最好的情况下，在体模内发现低残余物（约 1-5 KPa），而在肿瘤和小叶边界处发现高残余物（约 10-30 KPa）。从我们的有限元分析中，乳腺体模在 2D 和 3D 中产生的 CTE 优于块状体模。它还显示了深部和浅部肿瘤的 CTE 值和应变对比度的差异，并显示了使用 3D 建模时 CTE 的显著变化。在块状体模中，这些变化并不显著。两种体模在增加输入肿瘤背景模量对比度时，CTE 值均变差。
块状体模是一个起点工具，但像所提出的乳腺体模这样的下一级体模将在临床测试之前作为弹性成像模拟的有价值的中间工具。此外，鉴于乳腺体模的 CTE 指标优于块状体模，并且因肿瘤形状、位置和刚度而异，这些体模将增强对弹性成像对比度的研究。此外，与使用 3D 模型获得的真实 CTE 相比，使用具有平面应变近似值的 2D 体模高估了 CTE 值。因此，使用无近似值的乳腺体模等 3D 体模将有助于更准确地估计模量，这对于 3D 弹性成像系统尤其有价值。
