A one-dimensional flow model enhanced by machine learning for simulation of vocal fold vibration

TL;DR

This paper introduces a machine learning-enhanced 1D flow model for simulating vocal fold vibrations, combining physics-based equations with data-driven parameter estimation for improved accuracy and speed.

Contribution

It develops a novel hybrid 1D flow model trained on 3D FSI simulation data, incorporating analytical expressions for glottal effects, advancing vocal fold vibration simulation.

Findings

The enhanced 1D model accurately replicates 3D FSI results.

The model performs robustly on both idealized and subject-specific geometries.

It offers a faster alternative to traditional 3D simulations.

Abstract

We describe a one-dimensional (1D) unsteady and viscous flow model that is derived from the momentum and mass conservation equations, and to enhance this physics-based model, we use a machine learning approach to determine the unknown modeling parameters. Specifically, we first construct an idealized larynx model and perform ten cases of three-dimensional (3D) fluid--structure interaction (FSI) simulations. The flow data are then extracted to train the 1D flow model using a sparse identification approach for nonlinear dynamical systems. As a result of training, we obtain the analytical expressions for the entrance effect and pressure loss in the glottis, which are then incorporated in the flow model to conveniently handle different glottal shapes due to vocal fold vibration. We apply the enhanced 1D flow model in the FSI simulation of both idealized vocal fold geometries and subject-specific anatomical geometries reconstructed from the MRI images of rabbits' larynges. The 1D flow model is evaluated in both of these setups and is shown to have robust performance. Therefore, it provides a fast simulation tool superior to the previous 1D models.