Towards rapid prediction of personalised muscle mechanics: integration with diffusion tensor imaging
Towards rapid prediction of personalised muscle mechanics: integration with diffusion tensor imaging
复制标题
DOI：
10.1080/21681163.2018.1519850
复制
发表时间：
2020-09-02
影响因子：
1.6
通讯作者：
Zhang, Ju
中科院分区：
文献类型：
--
作者：
Fernandez, Justin;Mithraratne, Kumar;Zhang, Ju
扩散张量成像(DTI)已被广泛用于描述肌肉力学中的三维肌束结构。然而，与连续介质模型相关的计算开销使得它们在图形和医学可视化中的使用变得棘手。本研究将连续体肌肉力学与偏最小二乘回归相结合，建立一个快速的力学统计模型。我们以人的腓肠肌(内侧头和外侧头)为例，通过DTI获得信息。与有限元模型模拟相比，我们的统计模型预测了肌肉形状(均方根误差在0.063毫米以内)、肌腱力量(误差在1%以内)和组织应变(在肌肉收缩过程中最大误差在8%以内)。本文介绍的技术是将昂贵的连续介质力学与生物力学中的快速刚体解相结合的一步。虽然肌肉力是刚体解算器的主要目标，但现在可以集成包括3D肌肉形状和应力/应变场在内的附加信息。其主要优点之一是考虑了肌肉与其他软组织的相互作用，不需要估计肌力矩臂，并且不再简化详细丰富的3D连续统束结构，而是在生物力学模拟中充分发挥作用。这对肌肉骨骼生物力学、整形外科和医学可视化都有影响。
Diffusion tensor imaging (DTI) has been widely used to characterise the 3D fascicle architecture in muscle mechanics. However, the computational expense associated with continuum models make their use in graphics and medical visualisation intractable. This study presents an integration of continuum muscle mechanics with partial least-squares regression to create a fast mechanostatistical model. We use the human gastrocnemius muscle (medial and lateral heads) as an example informed though DTI. Our statistical models predicted muscle shape (within 0.063 mm root-mean-square (RMS) error), musculotendon force (within 1% error), and tissue strain (within 8% max error during muscle contraction), compared to a finite-element model simulation. The technique presented here is a step towards integrating expensive continuum mechanics with fast rigid body solutions in biomechanics. While muscle force is a primary objective in rigid body solvers the additional information including 3D muscle shape and stress/strain fields may now be integrated. One of the key benefits is that muscle interaction with other soft tissues is accounted for, muscle moment arms are not estimated, and the detailed rich 3D continuum fascicle architecture is no longer simplified but plays a full role in biomechanics simulation. This has implications for musculoskeletal biomechanics, orthopaedics and medical visualisation.