Abstract
The constitutive modelling of soft biological tissues has rapidly gained attention over the last 20 years. Current constitutive models can describe the mechanical properties of arterial tissue. Predicting these properties from microstructural information, however, remains an elusive goal. To address this challenge, we are introducing a novel hybrid modelling framework that combines advanced theoretical concepts with deep learning. It uses data from mechanical tests, histological analysis and images from second-harmonic generation. In this first proof of concept study, our hybrid modelling framework is trained with data from 27 tissue samples only. Even such a small amount of data is sufficient to be able to predict the stress–stretch curves of tissue samples with a median coefficient of determination of R2 = 0.97 from microstructural information, as long as one limits the scope to tissue samples whose mechanical properties remain in the range commonly encountered. This finding suggests that deep learning could have a transformative impact on the way we model the constitutive properties of soft biological tissues.
Highlights
How can we predict the mechanical response of a new, unknown tissue sample based solely on its histological and microstructural imaging data? To answer this question, we introduce a novel hybrid modelling framework that combines advanced theory-based constitutive modelling [12] with Deep learning (DL)
Our hybrid model achieved a predictive accuracy for the validation samples in the region of interest (ROI) with a median of R2 = 0.966, s.d. ± 0.01
Significant advances have been made in constitutive modelling of soft biological tissues in the last decades
Summary
Soft tissue biomechanics has substantially contributed to our understanding of various cardiovascular diseases such as aneurysms and dissections. The waviness and cross-linking of collagen fibres affect the stiffening of arterial tissue under tensile loading While some phenomena such as fibre dispersion can well be captured with advanced microstructure-based constitutive models [1], the incorporation of more subtle microstructural and histological information for the purpose of predictive constitutive modelling of arterial tissue remains elusive to date. In soft tissue biomechanics, such huge amounts of data are often not available due to the significant costs typically associated with experiments To overcome such problems, physics-informed machine learning methods have recently been proposed for constitutive modelling [11]. Physics-informed machine learning methods have recently been proposed for constitutive modelling [11] These can achieve a high level of predictive accuracy from a surprisingly small set of training data, since their architecture has physical knowledge that enables them to use the available training data with the utmost efficiency.
Sign-in/Register to view highlights & summary 
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
