Data-driven modeling of the mechanical behavior of anisotropic soft biological tissue
نویسندگان
چکیده
Closed-form constitutive models are currently the standard approach for describing soft tissues’ mechanical behavior. However, there inherent pitfalls to this approach. For example, explicit functional forms can lead poor fits, non-uniqueness of those and exaggerated sensitivity parameters. Here we overcome some these problems by designing deep neural networks (DNN) replace such expert models. One challenge using DNNs in context is enforcement stress-objectivity. We meet training our DNN predict strain energy its derivatives from (pseudo)-invariants. Thereby, also enforce polyconvexity through physics-informed constraints on strain-energy loss function. Direct prediction both derivative functions enables computation elasticity tensor needed a finite element implementation. Then, showcase DNN’s ability learning anisotropic behavior porcine murine skin biaxial test data. Through find that multi-fidelity scheme combines high fidelity experimental data with low analytical approximation yields best performance. Finally, conduct simulations tissue expansion model illustrate potential data-driven approaches as ours medical device design. Also, expect open software stemming work will broaden use mechanics.
منابع مشابه
modification of nanoclay for improving the physico-mechanical properties of dental adhesives
هدف اصلی این مطالعه تهیه یک سامانه نوین چسب عاجی دندانی بر پایه نانورس پیوند شده با پلی متاکریلیک اسید، نانورس پیوند شده با پلی اکریلیک اسید، مخلوط نانوسیلیکا و نانورس پیوند شده با پلی متاکریلیک اسید، مخلوط نانوسیلیکا و نانورس پیوند شده با پلی اکریلیک اسید و نانورس پیوند شده با کیتوسان اصلاح شده با گلایسیدیل متاکریلات است. پیوند پلی متاکریلیک اسید و پلی اکریلیک اسید بر ری سطح نانورس در حضور و ...
A Constitutive Model for Anisotropic Soft Biological Tissue Subject to Mechanical Damage
Many soft biological tissues have the important task of transmitting necessary forces to control movements, hold organs in place or ensure a relatively continuous flow of fluids. Examples of these tissues are ligaments, tendons, skin and blood vessels. Biological tissues can be seen as composite materials whose constituents are assembled into a hierarchical structure. At a macroscopic structura...
متن کاملCohesive behavior of soft biological adhesives: experiments and modeling.
Extracellular proteins play a key role in generating and maintaining cohesion and adhesion in biological tissues. These "natural glues" are involved in vital biological processes such as blood clotting, wound healing and maintaining the structural integrity of tissues. Macromolecular assemblies of proteins can be functionally stabilized in a variety of ways in situ that include ionic interactio...
متن کاملHybrid Mechanical and Data-driven Modeling Improves Inverse Kinematic Control of a Soft Robot
Feed-forward control relies on accurate knowledge about the controlled plant, e.g. models of manipulator kinematics or dynamics. However, for many plants, mechanical models do not capture all aspects of a plant or the plant’s intrinsic properties, e.g. soft materials, do hardly allow for exact and efficient mechanical modeling. In this context, machine learning is a suitable technique to extrac...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: Engineering With Computers
سال: 2022
ISSN: ['0177-0667', '1435-5663']
DOI: https://doi.org/10.1007/s00366-022-01733-3