Abstract:The development of biophysical models for clinical applications is rapidly advancing in the research community, thanks to their predictive nature and their ability to assist the interpretation of clinical data. However, high-resolution and accurate multi-physics computational models are computationally expensive and their personalisation involves fine calibration of a large number of parameters, which may be space-dependent, challenging their clinical translation. In this work, we propose a new approach which relies on the combination of physics-informed neural networks (PINNs) with three-dimensional soft tissue nonlinear biomechanical models, capable of reconstructing displacement fields and estimating heterogeneous patient-specific biophysical properties. The proposed learning algorithm encodes information from a limited amount of displacement and, in some cases, strain data, that can be routinely acquired in the clinical setting, and combines it with the physics of the problem, represented by a mathematical model based on partial differential equations, to regularise the problem and improve its convergence properties. Several benchmarks are presented to show the accuracy and robustness of the proposed method and its great potential to enable the robust and effective identification of patient-specific, heterogeneous physical properties, s.a. tissue stiffness properties. In particular, we demonstrate the capability of the PINN to detect the presence, location and severity of scar tissue, which is beneficial to develop personalised simulation models for disease diagnosis, especially for cardiac applications.

What problem does this paper attempt to address?

### What problem does this paper attempt to solve? This paper aims to solve the problem of estimating material properties (such as tissue stiffness) in nonlinear biomechanical models of soft tissues, especially the personalized parameter estimation in clinical applications. Specifically, the paper proposes a new method that combines Physics - Informed Neural Networks (PINNs) with three - dimensional nonlinear biomechanical models of soft tissues to reconstruct displacement fields and estimate patient - specific biophysical properties (such as stress and strain). This method can extract information from limited clinical data and combine physical laws (represented by partial differential equations), thereby improving computational efficiency and accuracy. #### Main challenges 1. **High computational cost of high - resolution and multi - physical - field computational models**: These models require fine - tuning of a large number of parameters, especially spatially - dependent parameters, which makes them difficult to apply in a clinical environment. 2. **Accurately estimating material properties from limited clinical data**: In the clinical environment, only limited displacement and strain data are usually available. How to use these data for accurate parameter estimation is a challenge. #### Solutions The authors propose a method based on PINNs, which solves the above problems in the following ways: - **Combining physical laws and neural networks**: Using PINNs to embed physical laws (such as partial differential equations) into neural networks, so that physical constraints are considered during the training process. - **Handling limited data**: By combining limited displacement and strain data and physical models, the robustness and convergence of the model are improved. - **Estimating heterogeneous material properties**: It can not only estimate homogeneous material properties, but also handle spatially - dependent heterogeneous material properties, such as detecting and quantifying the presence, location, and severity of scar tissue. #### Application scenarios This method is particularly suitable for cardiac applications and can help develop personalized simulation models for disease diagnosis, especially for heart disease diagnosis. By detecting the presence and location of scar tissue, more accurate diagnosis information can be provided to doctors, so as to develop more effective treatment plans. ### Summary The main objective of the paper is to develop an efficient and accurate method to estimate patient - specific biophysical properties by combining PINNs with nonlinear biomechanical models of soft tissues, thereby promoting the development of personalized medicine, especially in the diagnosis and treatment of heart diseases.

Physics-informed Neural Network Estimation of Material Properties in Soft Tissue Nonlinear Biomechanical Models

Identifying heterogeneous micromechanical properties of biological tissues via physics-informed neural networks

Biomechanics-informed Non-rigid Medical Image Registration and its Inverse Material Property Estimation with Linear and Nonlinear Elasticity

Physics-Informed Neural Networks for Material Model Calibration from Full-Field Displacement Data

Non-invasive Inference of Thrombus Material Properties with Physics-Informed Neural Networks

Non-rigid Medical Image Registration using Physics-informed Neural Networks

Learning solution of nonlinear constitutive material models using physics-informed neural networks: COMM-PINN

Identifying Constitutive Parameters for Complex Hyperelastic Materials using Physics-Informed Neural Networks

Modeling of additive manufacturing processes with time‐dependent material properties using physics‐informed neural networks

Elasticity imaging using physics-informed neural networks: Spatial discovery of elastic modulus and Poisson's ratio

Data-driven Modeling of the Mechanical Behavior of Anisotropic Soft Biological Tissue

Modeling finite-strain plasticity using physics informed neural network and assessment of the network performance

Enhanced physics‐informed neural networks for hyperelasticity

Physics-informed Neural Networks for Heterogeneous Poroelastic Media

A Physics-Guided Neural Operator Learning Approach to Model Biological Tissues from Digital Image Correlation Measurements

Learning solutions of thermodynamics-based nonlinear constitutive material models using physics-informed neural networks

Neural Network Approaches for Soft Biological Tissue and Organ Simulations

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

A generic physics-informed neural network-based constitutive model for soft biological tissues

Effective data sampling strategies and boundary condition constraints of physics-informed neural networks for identifying material properties in solid mechanics