A multi-resolution physics-informed recurrent neural network: formulation and application to musculoskeletal systems.

This work presents a multi-resolution physics-informed recurrent neural network (MR PI-RNN), for simultaneous prediction of musculoskeletal (MSK) motion and parameter identification of the MSK systems. The MSK application was selected as the model problem due to its challenging nature in mapping the high-frequency surface electromyography (sEMG) signals to the low-frequency body joint motion controlled by the MSK and muscle contraction dynamics. The proposed method utilizes the fast wavelet transform to decompose the mixed frequency input sEMG and output joint motion signals into nested multi-resolution signals. The prediction model is subsequently trained on coarser-scale input-output signals using a gated recurrent unit (GRU), and then the trained parameters are transferred to the next level of training with finer-scale signals. These training processes are repeated recursively under a transfer-learning fashion until the full-scale training (i.e., with unfiltered signals) is achieved, while satisfying the underlying dynamic equilibrium. Numerical examples on recorded subject data demonstrate the effectiveness of the proposed framework in generating a physics-informed forward-dynamics surrogate, which yields higher accuracy in motion predictions of elbow flexion-extension of an MSK system compared to the case with single-scale training. The framework is also capable of identifying muscle parameters that are physiologically consistent with the subject's kinematics data.

Investigating the Correlation between Force Output, Strains, and Pressure for Active Skeletal Muscle Contractions.

Experimental observations suggest that the force output of the skeletal muscle tissue can be correlated to the intra-muscular pressure generated by the muscle belly. However, pressure often proves difficult to measure through in-vivo tests. Simulations on the other hand, offer a tool to model muscle contractions and analyze the relationship between muscle force generation and deformations as well as pressure outputs, enabling us to gain insight into correlations among experimentally measurable quantities such as principal and volumetric strains, and the force output. In this work, a correlation study is performed using Pearson's and Spearman's correlation coefficients on the force output of the skeletal muscle, the principal and volumetric strains experienced by the muscle and the pressure developed within the muscle belly as the muscle tissue undergoes isometric contractions due to varying activation profiles. The study reveals strong correlations between force output and the strains at all locations of the belly, irrespective of the type of activation profile used. This observation enables estimation on the contribution of various muscle groups to the total force by the experimentally measurable principal and volumetric strains in the muscle belly. It is also observed that pressure does not correlate well with force output due to stress relaxation near the boundary of muscle belly.

A Feature-Encoded Physics-Informed Parameter Identification Neural Network for Musculoskeletal Systems.

Identification of muscle-tendon force generation properties and muscle activities from physiological measurements, e.g., motion data and raw surface electromyography (sEMG), offers opportunities to construct a subject-specific musculoskeletal (MSK) digital twin system for health condition assessment and motion prediction. While machine learning approaches with capabilities in extracting complex features and patterns from a large amount of data have been applied to motion prediction given sEMG signals, the learned data-driven mapping is black-box and may not satisfy the underlying physics and has reduced generality. In this work, we propose a feature-encoded physics-informed parameter identification neural network (FEPI-PINN) for simultaneous prediction of motion and parameter identification of human MSK systems. In this approach, features of high-dimensional noisy sEMG signals are projected onto a low-dimensional noise-filtered embedding space for the enhancement of forwarding dynamics prediction. This FEPI-PINN model can be trained to relate sEMG signals to joint motion and simultaneously identify key MSK parameters. The numerical examples demonstrate that the proposed framework can effectively identify subject-specific muscle parameters and the trained physics-informed forward-dynamics surrogate yields accurate motion predictions of elbow flexion-extension motion that are in good agreement with the measured joint motion data.

Multiscale modeling of passive material influences on deformation and force output of skeletal muscles.

Passive materials in human skeletal muscle tissues play an important role in force output of skeletal muscles. This paper introduces a multiscale modeling framework to investigate how age-associated variations on microscale passive muscle components, including microstructural geometry (e.g., connective tissue thickness) and material properties (e.g., anisotropy), influence the force output and deformations of the continuum skeletal muscle. We first define a representative volume element (RVE) for the microstructure of muscle and determine the homogenized macroscale mechanical properties of the RVE from the separate mechanical properties of the individual components of the RVE, including muscle fibers and connective tissue with its associated collagen fibers. The homogenized properties of the RVE are then used to define the elements of the continuum muscle model to evaluate the force output and deformations of the whole muscle. Conversely, the regional deformations of the continuum model are fed back to the RVE model to determine the responses of the individual microscale components. Simulations of muscle isometric contractions at a range of muscle lengths are performed to investigate the effects of muscle architectural changes (e.g., pennation angles) due to aging on force output and muscle deformation. The correlations between the pennation angle, the shear deformation in the microscale connective tissue (an indicator for the lateral force transmission), the angle difference between the fiber direction and principal strain direction and the resulting shear deformation at the continuum scale, as well as the force output of the skeletal muscle are also discussed.