K. Linka

Automated model discovery for human cardiac tissue: Discovering the best model and parameters

This paper automates cardiac tissue model discovery by training a neural network to identify the best model and parameters for human cardiac tissue.

Democratizing biomedical simulation through automated model discovery and a universal material subroutine

This paper automates material model selection in finite element simulations using constitutive neural networks, improving robustness and accessibility of biomedical simulations.

On automated model discovery and a universal material subroutine

This paper automates model selection for constitutive modeling in finite element analysis, demonstrating a universal material subroutine that generalizes to multiple models and accurately predicts stress in human brain tissue.

Discovering the mechanics of artificial and real meat

This paper compares the mechanical properties of artificial and real meat using tension, compression, and shear tests, finding that artificial chicken matches real chicken in stress values but tofurky is stiffer and less shear-resistant.

Automated model discovery for muscle using constitutive recurrent neural networks

This paper introduces a constitutive recurrent neural network to model viscoelastic behavior of skeletal muscle, outperforming traditional models and vanilla neural networks in prediction accuracy.

Automated model discovery for human brain using Constitutive Artificial Neural Networks

This paper introduces Constitutive Artificial Neural Networks to automatically discover both constitutive models and their parameters for human brain tissue from experimental data.

Principal-stretch-based constitutive neural networks autonomously discover a subclass of Ogden models for human brain tissue

This paper introduces a neural network that automatically discovers constitutive models for human brain tissue, outperforming traditional models in capturing tension, compression, and shear behavior.

Automated model discovery for skin:Discovering the best model, data, and experiment

This paper introduces Constitutive Artificial Neural Networks (CANN) to automatically discover the best constitutive model, parameters, and experiments for skin mechanics.