Mesh morphing strategies for image-based in-silico musculoskeletal biomechanics

背景和益处预计在未来二十年中，肌肉骨骼干预将有相当大的增长，这将增加对开发、改进和有针对性干预的需求。这样的发展可以通过计算机模型来实现，例如特定患者的有限元模型。众所周知，这种模型采用六面体网格有利于提高计算分析的准确性。然而，六面体网格的自动构建是一个独立的研究领域，因此，特定图像六面体网格的开发仅限于可用的工具，这些工具需要多个手动步骤，使得该过程的准确性高度依赖于用户。目的通过开发适用于多种工具的网格变形策略，促进下一代针对患者的计算机肌肉骨骼模型的再现性和准确性。新的网格变形策略将允许现有的基于图像的网格，以当前耗时的方式开发，自动适应新的主题，大大加快了模型开发过程，而不会影响网格质量。已经战略性地选择了三个解剖组织部位来开发和测试该工具。每个站点都提供了代表技术发展里程碑的独特挑战（详见下文）。此外，该工具还可以应用于从可用的图像数据库（内部和公共）中快速开发模型，用于所选的每个解剖部位。这个博士项目将开发新的变形或翘曲算法。最先进的技术将通过创建包括灰度值等基于图像的信息的多体网格来推进。该项目将包括几个方面：1/使用网格工具（IA-FEMesh，有限元软件中嵌入的工具或博士期间开发的工具）在每个感兴趣的解剖部位（髋关节，脚踝，脊柱）建立标准的六面体组织网格。2/开发和实现相对通用的变形或翘曲工具。3/使用来自iMBE或公共存储库的（预）临床图像数据库对开发的工具进行髋关节、踝关节和脊柱的验证。使用六面体网格的优越性将在不可压缩材料和组织的精确建模中得到证明。新开发的网格将用于评估四面体网格和六面体网格在接触力学方面的差异。这三个应用程序（脊柱、髋关节和踝关节）代表了IMBE现有图像数据库和建模专业知识的领域。六面体网格发展和变形方法将依次应用于每个解剖区域，并在每个阶段引入额外的复杂性：脊柱：椎间盘负责人类脊柱的大部分运动，其退变是使人衰弱的疼痛的来源。相对简单的整体形状使圆盘成为二维和三维初始网格和变形发展的良好对象。一种可能的扩展是捕捉光盘内部的分层结构。髋关节：覆盖髋关节骨盆一侧的软骨是早期可能发生损伤的部位之一，与晚年的骨关节炎有关。这种组织的形状提出了额外的挑战。应用这里的初始方法将演示变形方法可以实现多少形状变化。一个可能的扩展是包括“唇”组织包围软骨在腔的边缘。踝关节：踝关节容易发生创伤后骨关节炎，骨骼之间的接触面积很小，解剖结构的微小变化与高功能变化有关。变形网格以适应踝关节骨骼将提出在接触面上解决多个网格的额外挑战，以及捕获每个骨骼的形状。

Background and benefitsConsiderable growth in musculoskeletal intervention is predicted in the next two decades, increasing the need to develop, improve and target interventions. Such developments can be achieved through in-silico models, such as patient-specific finite element models. It is well known that a hexahedral mesh for such models is beneficial on the accuracy of the computational analysis. However, the automatic construction of hexahedral meshes is a research area on its own and therefore the development of image-specific hexahedral meshes is restricted to available tools which require multiple manual steps, making the accuracy of the process is highly user-dependent.AimsTo facilitate reproducibility and accuracy in the next-generation of in-silico patient-specific musculoskeletal models by developing mesh morphing strategies, applicable to meshes from multiple tools.The new mesh morphing strategies will allow existing image-based meshes, developed in the current time-consuming fashion, to be automatically adapted to a new subject, significantly speeding up the model development process, without compromising on mesh quality.Three anatomic tissue sites have been strategically chosen for the development and testing of the tool. Each site provides unique challenges representing milestones in the technical development (detailed below). In addition, the tool can be applied to rapidly develop models from the available image databases (both internal and public) for each anatomic site chosen.Research projectThis PhD project will develop new morphing or warping algorithms. State-of-the-art will be advanced by creating multi-body meshes including image-based information such as grey-scale values.The project will consist of several aspects:1/ Building standard hexahedral mesh of the tissues on each anatomic site of interest (hip, ankle, spine) using meshing tools (IA-FEMesh, tools embedded in Finite Element Software's, or tools developed during the PhD).2/ Development and implementation of relatively generic morphing or warping tools.3/ Validation of the developed tool using (pre-)clinical image databases from iMBE, or public repositories, on the hip, the ankle and the spine. The superiority of using hexahedral meshes will be demonstrated in accurately modelling incompressible materials and tissues. The new developed meshes will be used to assess the difference in contact mechanics between tetrahedral meshes and hexahedral meshes.The three applications (spine, hip and ankle) represent areas of existing image databases and modelling expertise within the IMBE. The hexahedral mesh development and morphing methods will be applied to each anatomical area in turn, with additional complexity introduced at each stage:Spine: The intervertebral disc is responsible for the majority of the movement possible in the human spine and its degeneration is the source of debilitating pain. The relative simple overall shape makes the disc a good subject for initial mesh and morphing development in two and three dimensions. A possible extension is to capture the layered structure within the disc.Hip: The cartilage covering the pelvic side of the hip joint is one of the sites where damage can occur at an early age, linked with osteoarthritis in later life. The shape of this tissue presents additional challenges. Applying the initial methods here will demonstrate how much shape change the morphing methodology can achieve. A possible extension is to include the 'labral' tissue which surrounds the cartilage at the edge of the cavity.Ankle: The ankle joint is prone to post-traumatic osteoarthritis, with very small contact areas between bones and small anatomical changes linked to high functional change. Morphing a mesh to fit to the ankle bones will present the additional challenge of resolving multiple meshes at a contact surface, as well as capturing the shape of each bone.