A three-dimensional vocal fold posturing model based on muscle mechanics and magnetic resonance imaging of a canine larynx
A three-dimensional vocal fold posturing model based on muscle mechanics and magnetic resonance imaging of a canine larynx
复制标题
DOI：
10.1121/10.0001093
复制
发表时间：
2020-04-01
影响因子：
2.4
通讯作者：
Zheng Xudong
中科院分区：
文献类型：
--
作者：
Geng Biao;Pham Ngoc;Zheng Xudong
在这项工作中，一个高保真的三维连续体模型的犬喉部框架被开发用于模拟喉部姿势。通过磁共振成像（MRI）构建每块肌肉和软骨，该模型在解剖学上是高度真实的。采用有限元法对肌肉力学进行建模。通过模拟单个和组喉部肌肉在系统激活下的声带姿态对该模型进行了测试，该模型准确预测了来自犬活体喉部的声带姿态参数。为了证明其应用，该模型随后被用于研究类杓肌运动、内侧表面形态和声带外展的肌肉控制。结果表明，传统分类的内收肌和外展肌可以对声带姿势产生相反的影响，使得说话和唱歌时高度复杂的喉部调节成为可能。这些结果表明，一个现实的综合喉部模型是可行的，这是迈向基于因果物理的声音产生模型的关键一步。
In this work, a high-fidelity three-dimensional continuum model of the canine laryngeal framework was developed for simulating laryngeal posturing. By building each muscle and cartilage from magnetic resonance imaging (MRI), the model is highly realistic in anatomy. The muscle mechanics is modeled using the finite-element method. The model was tested by simulating vocal fold postures under systematic activations of individual as well as groups of laryngeal muscles, and it accurately predicted vocal fold posturing parameters reported from in vivo canine larynges. As a demonstration of its application, the model was then used to investigate muscle controls of arytenoid movements, medial surface morphology, and vocal fold abduction. The results show that the traditionally categorized adductor and abductor muscles can have opposite effects on vocal fold posturing, making highly complex laryngeal adjustments in speech and singing possible. These results demonstrate that a realistic comprehensive larynx model is feasible, which is a critical step toward a causal physics-based model of voice production.