Predicting 3D soft tissue dynamics from 2D imaging using physics informed neural networks

Tissue dynamics play critical roles in many physiological functions and provide important metrics for clinical diagnosis. Capturing real-time high-resolution 3D images of tissue dynamics remains a challenge. This study presents a hybrid physics-informed neural network (PINN) algorithm that infers 3D flow-induced tissue dynamics and other physical quantities from sparse 2D images. The algorithm combines a recurrent neural network model of soft tissue with a differentiable fluid solver, leveraging prior knowledge in solid mechanics by projecting the governing equations onto a discrete eigenmode space. A long short-term memory (LSTM)-based recurrent encoder-decoder connected to a fully connected neural network captures temporal dependence in flow–structure interaction. The approach is demonstrated on synthetic data from a canine vocal fold model and experimental data from excised pigeon syringes, accurately reconstructing 3D vocal dynamics, aerodynamics, and acoustics from sparse 2D vibration profiles.