Advanced computational workflow for the multi-scale modeling of the bone metabolic processes

Recurrent neural network to predict hyperelastic constitutive behaviors of the skeletal muscle

Hyperelastic constitutive laws have been commonly used to model the passive behavior of the human skeletal muscle. Despite many efforts, the use of accurate finite element formulations of hyperelastic constitutive laws is still time-consuming for a real-time medical simulation system. The objective of the present study was to develop a deep learning model to predict the hyperelastic constitutive behaviors of the skeletal muscle toward a fast estimation of the muscle tissue stress.A finite element (FE) model of the right psoas muscle was developed. Neo-Hookean and Mooney-Rivlin laws were used. A tensile test was performed with an applied body force. A learning database was built from this model using an automatic probabilistic generation process. A long-short term memory (LSTM) neural network was implemented to predict the stress evolution of the skeletal muscle tissue. A hyperparameter tuning process was conducted. Root mean square error (RMSE) and associated relative error was quantified to evaluate the precision of the predictive capacity of the developed deep learning model. Pearson correlation coefficients (R) was also computed.The nodal displacements and the maximal stresses range from 70 to 227 mm and from 2.79 to 5.61 MPa for Neo-Hookean and Monney-Rivlin laws, respectively. Regarding the LSTM predictions, the RMSE ranges from 224.3 ± 3.9 Pa (8%) to 227.5 [Formula: see text] 5.7 Pa (4%) for Neo-Hookean and Monney-Rivlin laws, respectively. Pearson correlation coefficients (R) of 0.78 [Formula: see text] 0.02 and 0.77 [Formula: see text] 0.02 were obtained for Neo-Hookean and Monney-Rivlin laws, respectively.The present study showed that, for the first time, the use of a deep learning model can reproduce the time-series behaviors of the complex FE formulations for skeletal muscle modeling. In particular, the use of a LSTM neural network leads to a fast and accurate surrogate model for the in silico prediction of the hyperelastic constitutive behaviors of the skeletal muscle. As perspectives, the developed deep learning model will be integrated into a real-time medical simulation of the skeletal muscle for prosthetic socket design and childbirth simulator.

A Multidisciplinary Journey towards Bone Tissue Engineering

Millions of patients suffer yearly from bone fractures and disorders such as osteoporosis or cancer, which constitute the most common causes of severe long-term pain and physical disabilities. The intrinsic capacity of bone to repair the damaged bone allows normal healing of most small bone injuries. However, larger bone defects or more complex diseases require additional stimulation to fully heal. In this context, the traditional routes to address bone disorders present several associated drawbacks concerning their efficacy and cost-effectiveness. Thus, alternative therapies become necessary to overcome these limitations. In recent decades, bone tissue engineering has emerged as a promising interdisciplinary strategy to mimic environments specifically designed to facilitate bone tissue regeneration. Approaches developed to date aim at three essential factors: osteoconductive scaffolds, osteoinduction through growth factors, and cells with osteogenic capability. This review addresses the biological basis of bone and its remodeling process, providing an overview of the bone tissue engineering strategies developed to date and describing the mechanisms that underlie cell-biomaterial interactions.

Integration of mechanics and biology in computer simulation of bone remodeling

Bone remodeling is a complex physiological process that spans across multiple spatial and temporal scales and is regulated by both mechanical and hormonal cues. An imbalance between bone resorption and bone formation in the process of bone remodeling may lead to various bone pathologies. One powerful and non-invasive approach to gain new insights into mechano-adaptive bone remodeling is computer modeling and simulation. Recent findings in bone physiology and advances in computer modeling have provided a unique opportunity to study the integration of mechanics and biology in bone remodeling. Our objective in this review is to critically appraise recent advances and developments and discuss future research opportunities in computational bone remodeling approaches that enable integration of mechanics and cellular and molecular pathways. Based on the critical appraisal of the relevant recent published literature, we conclude that multiscale in silico integration of personalized bone mechanics and mechanobiology combined with data science and analytics techniques offer the potential to deepen our knowledge of bone remodeling and provide ample opportunities for future research.