Elastic parameter identification of three-dimensional soft tissue based on deep neural network

In the field of virtual surgery and deformation simulation, the identification of elastic parameters of human soft tissues is a critical technology that directly affects the accuracy of deformation simulation. Current research on soft tissue deformation simulation predominantly assumes that the elasticity of tissues is fixed and already known, leading to the difficulty in populating with the elasticity measured or identified from specific tissues of real patients. Existing elasticity modeling efforts struggle to be implemented on irregularly structured soft tissues, failing to adapt to clinical surgical practices. Therefore, this paper proposes a new method for identifying human soft tissue elastic parameters based on the finite element method and the deep neural network, UNet. This method requires only the full-field displacement data of soft tissues under external loads to predict their elastic distribution. The performance and validity of the algorithm are assessed using test data and clinical data from rhinoplasty surgeries. Experiments demonstrate that the method proposed in this paper can achieve an accuracy of over 99% in predicting elastic parameters. Clinical data validation shows that the predicted elastic distribution can reduce the error in finite element deformation simulations by more than 80% at the maximum compared to the error with traditional uniform elastic parameters, effectively enhancing the computational accuracy in virtual surgery simulations and soft tissue deformation modeling.

Physics-informed UNets for discovering hidden elasticity in heterogeneous materials

Soft biological tissues often have complex mechanical properties due to variation in structural components. In this paper, we develop a novel UNet-based neural network model for inversion in elasticity (El-UNet) to infer the spatial distributions of mechanical parameters from strain maps as input images, normal stress boundary conditions, and domain physics information. We show superior performance - both in terms of accuracy and computational cost - by El-UNet compared to fully-connected physics-informed neural networks in estimating unknown parameters and stress distributions for isotropic linear elasticity. We characterize different variations of El-UNet and propose a self-adaptive spatial loss weighting approach. To validate our inversion models, we performed various finite-element simulations of isotropic domains with heterogenous distributions of material parameters to generate synthetic data. El-UNet is faster and more accurate than the fully-connected physics-informed implementation in resolving the distribution of unknown fields. Among the tested models, the self-adaptive spatially weighted models had the most accurate reconstructions in equal computation times. The learned spatial weighting distribution visibly corresponded to regions that the unweighted models were resolving inaccurately. Our work demonstrates a computationally efficient inversion algorithm for elasticity imaging using convolutional neural networks and presents a potential fast framework for three-dimensional inverse elasticity problems that have proven unachievable through previously proposed methods.

The role of shear viscosity as a biomarker for improving chronic kidney disease detection using shear wave elastography: A computational study using a validated finite element model

The application of ultrasound shear wave elastography for detecting chronic kidney disease, namely renal fibrosis, has been widely studied. A good correlation between tissue Young's modulus and the degree of renal impairment has been established. However, the current limitation of this imaging modality pertains to the linear elastic assumption used in quantifying the stiffness of renal tissue in commercial shear wave elastography systems. As such, when underlying medical conditions such as acquired cystic kidney disease, which may potentially influence the viscous component of renal tissue, is present concurrently with renal fibrosis, the accuracy of the imaging modality in detecting chronic kidney disease may be affected. The findings in this study demonstrate that quantifying the stiffness of linear viscoelastic tissue using an approach similar to those implemented in commercial shear wave elastography systems led to percentage errors as high as 87%. The findings presented indicate that use of shear viscosity to detect changes in renal impairment led to a reduction in percentage error to values as low as 0.3%. For cases in which renal tissue was affected by multiple medical conditions, shear viscosity was found to be a good indicator in gauging the reliability of the Young's modulus (quantified through a shear wave dispersion analysis) in detecting chronic kidney disease. The findings show that percentage error in stiffness quantification can be reduced to as low as 0.6%. The present study demonstrates the potential use of renal shear viscosity as a biomarker to improve the detection of chronic kidney disease.

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

Elasticity imaging is a technique that discovers the spatial distribution of mechanical properties of tissue using deformation and force measurements under various loading conditions. Given the complexity of this discovery, most existing methods approximate only one material parameter while assuming homogeneous distributions for the others. We employ physics-informed neural networks (PINN) in linear elasticity problems to discover the space-dependent distribution of both elastic modulus (E) and Poisson's ratio (ν) simultaneously, using strain data, normal stress boundary conditions, and the governing physics. We validated our model on three examples. First, we experimentally loaded hydrogel samples with embedded stiff inclusions, representing tumorous tissue, and compared the approximations against ground truth determined through tensile tests. Next, using data from finite element simulation of a rectangular domain containing a stiff circular inclusion, the PINN model accurately localized the inclusion and estimated both E and ν. We observed that in a heterogeneous domain, assuming a homogeneous ν distribution increases estimation error for stiffness as well as the area of the stiff inclusion, which could have clinical importance when determining size and stiffness of tumorous tissue. Finally, our model accurately captured spatial distribution of mechanical properties and the tissue interfaces on data from another computational model, simulating uniaxial loading of a rectangular hydrogel sample containing a human brain slice with distinct gray matter and white matter regions and complex geometrical features. This elasticity imaging implementation has the potential to be used in clinical imaging scenarios to reliably discover the spatial distribution of mechanical parameters and identify material interfaces such as tumors. STATEMENT OF SIGNIFICANCE: Our work is the first implementation of physics-informed neural networks to reconstruct both material parameters - Young's modulus and Poisson's ratio - and stress distributions for isotropic linear elastic materials by having deformation and force measurements. We comprehensively validate our model using experimental measurements and synthetic data generated using finite element modeling. Our method can be implemented in clinical elasticity imaging scenarios to improve diagnosis of tumors and for mechanical characterization of biomaterials and biological tissues in a minimally invasive manner.

Hybrid Discrete-Continuum Multiscale Model of Tissue Growth and Remodeling

Tissue growth and remodeling (G&R) is often central to disease etiology and progression, so understanding G&R is essential for understanding disease and developing effective therapies. While the state-of-the-art in this regard is animal and cellular models, recent advances in computational tools offer another avenue to investigate G&R. A major challenge for computational models is bridging from the cellular scale (at which changes are actually occurring) to the macroscopic, geometric-scale (at which physiological consequences arise). Thus, many computational models simplify one scale or another in the name of computational tractability. In this work, we develop a discrete-continuum modeling scheme for analyzing G&R, in which we apply changes directly to the discrete cell and extracellular matrix (ECM) architecture and pass those changes up to a finite-element macroscale geometry. We demonstrate the use of the model in three case-study scenarios: the media of a thick-walled artery, and the media and adventitia of a thick-walled artery, and chronic dissection of an arterial wall. We analyze each case in terms of the new and insightful data that can be gathered from this technique, and we compare our results from this model to several others. STATEMENT OF SIGNIFICANCE: This work is significant in that it provides a framework for combining discrete, microstructural- and cellular-scale models to the growth and remodeling of large tissue structures (such as the aorta). It is a significant advance in that it couples the microscopic remodeling with an existing macroscopic finite element model, making it relatively easy to use for a wide range of conceptual models. It has the potential to improve understanding of many growth and remodeling processes, such as organ formation during development and aneurysm formation, growth, and rupture.

Three-Dimensional Anisotropic Hyperelastic Constitutive Model Describing the Mechanical Response of Human and Mouse Cervix

The mechanical function of the uterine cervix is critical for a healthy pregnancy. During pregnancy, the cervix undergoes significant softening to allow for a successful delivery. Abnormal cervical remodeling is suspected to contribute to preterm birth. Material constitutive models describing known biological shifts in pregnancy are essential to predict the mechanical integrity of the cervix. In this work, the material response of human cervical tissue under spherical indentation and uniaxial tensile tests loaded along different anatomical directions is experimentally measured. A deep-learning segmentation tool is applied to capture the tissue deformation during the uniaxial tensile tests. A 3-dimensional, equilibrium anisotropic continuous fiber constitutive model is formulated, considering collagen fiber directionality, fiber bundle dispersion, and the entropic nature of wavy cross-linked collagen molecules. Additionally, the universality of the material model is demonstrated by characterizing previously published mouse cervix mechanical data. Overall, the proposed material model captures the tension-compression asymmetric material responses and the remodeling characteristics of both human and mouse cervical tissue. The pregnant (PG) human cervix (mean locking stretch ζ=2.4, mean initial stiffness ξ=12 kPa, mean bulk modulus κ=0.26 kPa, mean dispersion b=1.0) is more compliant compared with the nonpregnant (NP) cervix (mean ζ=1.3, mean ξ=32 kPa, mean κ=1.4 kPa, mean b=1.4). Creating a validated material model, which describes the role of collagen fiber directionality, dispersion, and crosslinking, enables tissue-level biomechanical simulations to determine which material and anatomical factors drive the cervix to open prematurely. STATEMENT OF SIGNIFICANCE: In this study, we report a 3D anisotropic hyperelastic constitutive model based on Langevin statistic mechanics and successfully describe the material behavior of both human and mouse cervical tissue using this model. This model bridges the connection between the extracellular matrix (ECM) microstructure remodeling and the macro mechanical properties change of the cervix during pregnancy via microstructure-associated material parameters. This is the first model, to our knowledge, to connect the the entropic nature of wavy cross-linked collagen molecules with the mechanical behavior of the cervix. Inspired by microstructure, this model provides a foundation to understand further the relationship between abnormal cervical ECM remodeling and preterm birth. Furthermore, with a relatively simple form, the proposed model can be applied to other fibrous tissues in the future.

Structure-function relationships of TMJ lateral capsule-ligament complex

The human temporomandibular joint (TMJ) lateral capsule ligament (LCL) complex is debated as a fibrous capsule with distinct ligaments or ligamentous thickening, necessitating further evaluation of the complex and its role in TMJ anatomy and mechanics. This study explores the ultrastructural arrangement, biomechanical tensile properties, and biochemical composition of the human LCL complex including region-specific differences to explore the presence of a distinct temporomandibular ligament and sex-specific differences to inform evaluations of potential etiological mechanisms. LCL complex ultrastructural arrangement, biomechanical properties, and biochemical composition were determined using cadaveric samples. Statistical modeling assessed sex- and region-specific effects on LCL complex tissue properties. Collagen fiber coherency, collagen fiber bundle size, and elastin fiber count did not differ between sexes, but females trended higher in elastin fiber count. LCL complex water and sGAG content did not differ between sexes or regions, but collagen content was higher in the anterior region (311.0 ± 185.6 μg/mg) compared to the posterior region (221.0 ± 124.9 μg/mg) (p = 0.045) across sexes and in males (339.6 ± 170.6 μg/mg) compared to females (204.5 ± 130.7 μg/mg) (p = 0.006) across regions. Anterior failure stress (1.1 ± 0.7 MPa) was larger than posterior failure stress (0.6 ± 0.4 MPa) (p = 0.024). Regional differences confirm the presence of a mechanically and compositionally distinct temporomandibular ligament. Baseline sex-specific differences are critical for etiological investigations of sex disparities in TMJ disorders. These results have important biomechanical and clinical ramifications, providing critical baseline tissue material properties, informing the development of TMJ musculoskeletal models, and identifying new areas for etiologic investigations for temporomandibular disorders.

Integration and functional performance of a decellularised porcine superflexor tendon graft in an ovine model of anterior cruciate ligament reconstruction

The objective was to evaluate the performance of decellularised porcine superflexor tendon (pSFT) as an anterior cruciate ligament (ACL) reconstruction device. The ACL of adult sheep was reconstructed with decellularised pSFT or ovine allograft SFT and animals sacrificed at 4, 12 and 26 weeks (n = 4 per group) for biological evaluation and 26 weeks (n = 6) for biomechanical evaluation of the grafts. Both grafts showed good in vivo performance with no major differences at macroscopic evaluation post euthanasia. Histopathology revealed an inflammatory reaction to both grafts at 4 weeks, which reduced by 26 weeks. There was advanced cellular ingrowth from 12 weeks, ligamentisation of intra-articular grafts, ossification and formation of Sharpey's fibers at the graft/bone junctions. Immunohistochemistry showed that at 4 and 12 weeks, the host response was dominated by CD163+ M2 macrophages and a cell infiltrate comprising α-SMA + myofibroblasts, CD34⁺ and CD271+ progenitor cells. At 26 weeks the biomechanical properties of decellularised pSFT and oSFT grafts were comparable, with all grafts failing in the intra-articular region. This study provides new insight into constructive remodelling of tendons used for ACL replacement and evidence of integration and functional performance of a decellularised xenogeneic tendon with potential as an alternative for ACL reconstruction.

The swelling behavior of the mouse cervix: Changing kinetics with osmolarity and the role of hyaluronan in pregnancy

The cervical remodeling process during pregnancy is characterized by progressive compositional and structural changes in the tissues extra-cellular matrix (ECM). Appropriately timed remodeling is critical for healthy gestation and prevention of premature cervical softening leading to preterm birth (PTB). Modification of the ECM glycosaminoglycans (GAGs) content with advancing pregnancy, especially the non-sulfated GAG hyaluronan (HA), is a fundamental change associated with cervical remodeling. While GAGs have numerous physiological roles, the mechanical consequence of evolving GAG content on cervical structure-function behavior remains an open question. Additionally, an understanding of cervical swelling properties, postulated to be regulated in part by GAGs, is required for the appropriate definition of a reference configuration for mechanical tests and to enhance biological understanding. To investigate cervical swelling, osmotic loading tests are conducted on isolated wild type mouse cervices throughout pregnancy. These tests are performed in various osmolarity solutions to assess the influence of the media on swelling kinetics. A genetically altered strain of mice with depletion of cervical HA is also tested to elucidate the contribution of HA to tissue swelling. Results show ex vivo cervical swelling is significant with volume changes ranging from 20 to 100% after 3h of free swelling. The swelling kinetics depend highly on osmolarity of the media and is altered with advancing pregnancy. The contribution of HA to swelling is only significant in hypo-osmotic solution when HA cervical content is high at the end of pregnancy. In summary, it is critical to account for swelling deformation mechanisms after excision in mechanical experiments. STATEMENT OF SIGNIFICANCE: The cervical extracellular matrix (ECM) undergoes drastic changes to fulfill the functional change of the cervix during pregnancy. Inappropriate timing for this transformation can result in preterm birth, a severe clinical challenge. One of the fundamental changes of the cervical ECM is the significant modification of the glycosaminoglycan content, especially hyaluronan (HA), which is thought to contribute significantly to the swelling and mechanical properties of the cervix. This study aims to measure the swelling kinetics of cervical tissue during pregnancy and to investigate the role of HA in these swelling tendencies. Results show the significant swelling of cervical tissue, which evolves as pregnancy progresses, highlighting a key material property feature of the remodeled cervix. Using a mouse strain with a cervical HA depletion, this work shows HA contributes to the swelling trends of late-term cervical tissue, in a hypo-osmotic solution.

Associations between changes in loading pattern, deformity, and internal stresses at the foot with hammer toe during walking; a finite element approach

Over the past decade, Finite Element (FE) modelling has been used as a method to understand the internal stresses within the diabetic foot. Foot deformities such as hammer toe have been associated with increased risk of foot ulcers in diabetic patients. Hence the aim of this study is to investigate the influence of hammer toe deformity on internal stresses during walking. A 3D finite element model of the human foot was constructed based on capturing Magnetic Resonance Imaging (MRI) of a diabetic neuropathic volunteer exhibiting hammer toe. 3D gait measurements and a multi-body musculoskeletal model for the same participant were used to define muscle forces. FE simulations were run at five different instances during the stance phase of gait. Peak plantar pressure and pressure distribution results calculated from the model showed a good agreement with the experimental measurement having less than 11% errors. Maximum von Mises internal stresses in the forefoot hard tissue were observed at the 3^rd and 5^th metatarsals and 4^th proximal phalanx. Moreover, presence of hammer toe deformity was found to shift the location of maximum internal stresses on the soft tissue to the forefoot by changing the location of centre of pressure with internal stress 1.64 times greater than plantar pressure. Hammer toe deformity also showed to reduce the involvement of the first phalanx in internal/external load-bearing during walking. The findings of this study support the association between changes in loading pattern, deformity, and internal stresses in the soft tissue that lead to foot ulceration.

Maximum tensile stress and strain of skin of the domestic pig-differences concerning pigs from organic and non-organic farming

The purpose of this work has been to determine differences in biomechanical properties of porcine skin from organic and non-organic farming as porcine skin is widely used as a model for human skin. A test apparatus was used, using gravity to stretch and finally tear a dumbbell-shaped specimen of prepared abdominal skin with a testing surface area of 25 × 4 mm. A total of 32 specimens were taken from seven individual pigs, three from organic and four from non-organic farming, in different orientations with respect to the Langer's lines. The tests were performed at a dynamic speed of around 1.66 m/s (corresponding to a nominal strain rate of 67 s^-1). Engineering strain at rupture was higher in pig skin from non-organic farming with values up to 321% as opposed to 90% in organic pig skin. The maximum tensile stress found in non-organic pig skin was lower than in pig skin from organic farming with maximum values of 34 MPa as opposed to 58 MPa. The reason for the difference in biomechanical properties is unclear; the effect of sunlight is discussed as well as other factors like age and exercise. It seems that the biomechanical properties of porcine skin from organic farming are more similar to those of human skin.

Onset of neonatal locomotor behavior and the mechanical development of Achilles and tail tendons

Tendon tissue engineering approaches are challenged by a limited understanding of the role mechanical loading plays in normal tendon development. We propose that the increased loading that developing postnatal tendons experience with the onset of locomotor behavior impacts tendon formation. The objective of this study was to assess the onset of spontaneous weight-bearing locomotion in postnatal day (P) 1, 5, and 10 rats, and characterize the relationship between locomotion and the mechanical development of weight-bearing and non-weight-bearing tendons. Movement was video recorded and scored to determine non-weight-bearing, partial weight-bearing, and full weight-bearing locomotor behavior at P1, P5, and P10. Achilles tendons, as weight-bearing tendons, and tail tendons, as non-weight-bearing tendons, were mechanically evaluated. We observed a significant increase in locomotor behavior in P10 rats, compared to P1 and P5. We also found corresponding significant differences in the maximum force, stiffness, displacement at maximum force, and cross-sectional area in Achilles tendons, as a function of postnatal age. However, the maximum stress, strain at maximum stress, and elastic modulus remained constant. Tail tendons of P10 rats had significantly higher maximum force, maximum stress, elastic modulus, and stiffness compared to P5. Our results suggest that the onset of locomotor behavior may be providing the mechanical cues regulating postnatal tendon growth, and their mechanical development may proceed differently in weight-bearing and non-weight-bearing tendons. Further analysis of how this loading affects developing tendons in vivo may inform future engineering approaches aiming to apply such mechanical cues to regulate engineered tendon formation in vitro.

Seeds-An evolutionary innovation underlying reproductive success in flowering plants

"Seeds nourish, seeds unite, seeds endure, seeds defend, seeds travel," explains the science writer Thor Hanson in his book The Triumph of Seeds (2015). The seed is an ultimate product of land plant evolution. The nursing and protective organization of the seed enable a unique parental care of the progeny that has fueled seed plant radiation. Seeds promote dispersal and optimize offspring production and thus reproductive fitness through biological adaptations that integrate environmental and developmental cues. The composite structure of seeds, uniting tissues that originate from three distinct organisms, enables the partitioning of tasks during development, maturation, and storage, while a sophisticated interplay between the compartments allows the fine-tuning of embryonic growth, as well as seed maturation, dormancy, and germination. In this review, we will highlight peculiarities in the development and evolution of the different seed compartments and focus on the molecular mechanisms underlying the interactions between them.

Tissue and cellular biomechanics during corneal wound injury and repair

Corneal wound healing is an enormously complex process that requires the simultaneous cellular integration of multiple soluble biochemical cues, as well as cellular responses to the intrinsic chemistry and biophysical attributes associated with the matrix of the wound space. Here, we document how the biomechanics of the corneal stroma are altered through the course of wound repair following keratoablative procedures in rabbits. Further we documented the influence that substrate stiffness has on stromal cell mechanics. Following corneal epithelial debridement, New Zealand white rabbits underwent phototherapeutic keratectomy (PTK) on the right eye (OD). Wound healing was monitored using advanced imaging modalities. Rabbits were euthanized and corneas were harvested at various time points following PTK. Tissues were characterized for biomechanics with atomic force microscopy and with histology to assess inflammation and fibrosis. Factor analysis was performed to determine any discernable patterns in wound healing parameters. The matrix associated with the wound space was stiffest at 7days post PTK. The greatest number of inflammatory cells were observed 3days after wounding. The highest number of myofibroblasts and the greatest degree of fibrosis occurred 21days after wounding. While all clinical parameters returned to normal values 400days after wounding, the elastic modulus remained greater than pre-surgical values. Factor analysis demonstrated dynamic remodeling of stroma occurs between days 10 and 42 during corneal stromal wound repair. Elastic modulus of the anterior corneal stroma is dramatically altered following PTK and its changes coincide initially with the development of edema and inflammation, and later with formation of stromal haze and population of the wound space with myofibroblasts. Factor analysis demonstrates strongest correlation between elastic modulus, myofibroblasts, fibrosis and stromal haze thickness, and between edema and central corneal thickness.

STATEMENT OF SIGNIFICANCE

Tissue biomechanics during the course of corneal wound healing is documented for the first time through atomic force microscopy, and is correlated with advanced clinical imaging and immunohistochemistry. Parameters obtained from the study are applied in a multivariate statistical model to cluster the data for better classification and monitor the wound repair process. Elastic modulus of the anterior corneal stroma is dramatically altered following wounding and correlates initially with the development of edema and inflammation, and later with formation of stromal haze and population of the wound space with myofibroblasts. Importantly, the occurrence of myofibroblasts is preceded by changes in tissue mechanics, which is important to consider in light of crosslinking procedures applied to treat corneal diseases.

Structural and mechanical characterisation of the peri-prosthetic tissue surrounding loosened hip prostheses. An explorative study

Very little is known about the structure and properties of peri-prosthetic fibrous tissue that is found around loose orthopaedic implants. We describe a method for characterizing the structural organisation (histology, confocal microscopy) as well as the nano- and micro-scale mechanical behaviour (atomic force microscopy, nanoindentation) of peri-prosthetic fibrous tissue. The tissue was collected from 11 patients undergoing revision surgery due to aseptic loosening. Sirius Red and Movat histological staining procedures indicated that the tissue mainly consists of collagen fibres and ground substance. However, large inter- and intra-patient variations in the relative proportions of these tissue components were found, as well as in collagen fibre orientation and possibly also maturation. The nano-scale Young׳s moduli ranged from 0-950kPa, but showed large inter-patient variability. When the results per sample were presented in a probability density function, we could roughly discriminate one peak in the 0-100kPa range and/or one peak in the 100-500Pa range. These nano-scale moduli seem to respectively present the mechanical properties of glycosaminoglycan (GAG) and collagen molecules. The majority of the micro-scale Young׳s moduli ranged between 0.5 and 2.0kPa for all samples. This explorative study provides new insights in (the variations of) structural organisation and mechanical properties of peri-prosthetic tissue.

Fine Needle Elastography (FNE) device for biomechanically determining local variations of tissue mechanical properties

Diseased tissues exhibit changes in mechanical properties and thus possess clinical diagnostic significance. We report the design and development of a Fine Needle Elastography (FNE) prototype device integrated with Fine Needle Aspiration Cytology (FNAC) needle that allows for quantitative and sensitive assessment of tissues and materials based on local variations in elastic, friction, and cutting forces on needle insertion. A piezoelectric force-sensor at the base of FNA needle measures the forces opposing needle penetration with micrometer scale resolution. Measurement precision (±5 μm) and axial resolution (~20 μm) of FNE device was tested using control mm size gelatin matrices and unripe pear in assessing needle penetration resistance, force heterogeneity and optimization of needle penetration velocity. Further, we demonstrated the usefulness of FNE in quantitative, biomechanical differentiation of simulated thyroid tumor nodules in an ultrasound neck phantom. Fluid or solid nodules were probed in the phantom study coupled with ultrasound guidance. Our data shows significantly higher force variations (1-D force heterogeneity; HF,a=6.5 mN, HF,q=8.25 mN and stiffness heterogeneity; HS,a=0.0274 kN/m, HS,q=0.0395 kN/m) in solid nodules compared either to fluid nodules or to regions corresponding to healthy thyroid tissue within the ultrasound phantom. The results suggest future applications of in vivo FNE biopsies based on force heterogeneity to diagnose thyroid tumors in areas where ultrasound instrumentation or access to a qualified pathologist for FNAC are unavailable, as well as an ancillary diagnostic tool in thyroid cancer management.