Search

Orthodontic aligners undergo deformation during installation, producing an unexpected component of elastic restoring force that causes unintended changes in the dentition. The aim of this study was to investigate the relationship between strain and elastic recovery of the aligner. We distinguished the contributions to aligner deformation due to molding and installation by measuring the thickness distribution of an aligner after molding using micro-CT and tracking changes in grid patterns drawn on the sheet used to fabricate the aligner. The aligner was installed on a device that simulated canine movement. Although canine strain was quite strong around the cusp, and in premolar, it was observed mainly in their centers. Furthermore, after molding, thickness distribution of the aligner was found. But, it is no clear relationship between the thickness distribution and the strain distribution. Our method of analysis can help improve aligner design and establish molding method to deliver optimal orthodontic treatment.

The present study was aimed to develop a probabilistic finite element method (FEM) that predicts the variability in the fatigue life of additively manufactured clasp so that it can be used as a virtual test in the design phase before manufacturing. Titanium alloy (Ti-6Al-4V) clasp with integrated chucking part, which was designed for experimental fatigue test to validate the computational method, was investigated. To predict the lower bound, an initial spherical defect was assumed in the region where stress concentration was predicted. The Smith-Watson-Topper (SWT) method, BäumelSeeger rule, elasto-plastic FEM, and zooming FEM were used. The influence of assumed initial defect on the fatigue life was significant, and the large variability in the fatigue life was predicted. This study demonstrated that the proposed practical computational method can simulate the large variability in the fatigue life of titanium alloy clasp, which is useful in its design before manufacturing.

Convolution neural networks (CNNs) have proven effectiveness, but they are not applicable to all datasets, such as those with heterogeneous attributes, which are often used in the finance and banking industries. Such datasets are difficult to classify, and to date, existing high-accuracy classifiers and rule-extraction methods have not been able to achieve sufficiently high classification accuracies or concise classification rules. This study aims to provide a new approach for achieving transparency and conciseness in credit scoring datasets with heterogeneous attributes by using a one-dimensional (1D) fully-connected layer first CNN combined with the Recursive-Rule Extraction (Re-RX) algorithm with a J48graft decision tree (hereafter 1D FCLF-CNN). Based on a comparison between the proposed 1D FCLF-CNN and existing rule extraction methods, our architecture enabled the extraction of the most concise rules (6.2) and achieved the best accuracy (73.10%), i.e., the highest interpretability&ndash, priority rule extraction. These results suggest that the 1D FCLF-CNN with Re-RX with J48graft is very effective for extracting highly concise rules for heterogeneous credit scoring datasets. Although it does not completely overcome the accuracy&ndash, interpretability dilemma for deep learning, it does appear to resolve this issue for credit scoring datasets with heterogeneous attributes, and thus, could lead to a new era in the financial industry.

In the numerical prediction of the homogenized macroscopic properties of an arbitrary heterogeneous material with periodic microstructure, a general formulation of the first-order perturbation-based stochastic homogenization method is presented in a discretized form based on the finite element method in order to consider the variability or uncertainty of the mechanical properties of the material models. Many random parameters are defined for each material model and for each component of the stress–strain matrix of the constituent’s material model. The first-order terms of the characteristic displacement are thoroughly studied both theoretically and numerically, and are also used in the verification of the developed computer code. The comparison with the Monte Carlo simulation also supports the proposed formulation.

Aiming at the future applications of a lattice structure manufactured by selective laser sintering for bone scaffold in the biomedical field, by putting highlight on the possible imperfections influenced by different manufacturing strategies, a numerical prediction method was studied and compared with compressive test results. To consider the geometrical imperfections as well as the uncertainty in sintered material property, the first-order perturbation based stochastic homogenization (FPSH) method was employed. To be able to compare products made by three different printing service companies, maraging steel was used in this study, which was commonly available. Several geometrical imperfections were categorized into kink, notch and hole, and representative dimensions were parameterized and measured statistically using micro-CT images. They were also correlated to the building direction and support types. The numerical prediction revealed the reduction of stiffness in a stochastic way qualitatively. The constructed database of the geometrical imperfections and the FPSH method will be useful in the next occasion of manufacturing unexperienced product.

A probabilistic multiscale framework is proposed to characterize the probabilistic behaviors of macroscopic elastic properties of multiphase composite materials. The first-order perturbation-based stochastic homogenization method is extended to incorporate inherent randomness existed in multiphase constituents materials on basis of our previous work for porous material. Moreover, a generalized stochastic method is introduced to combine above with morphological nonparametric uncertainties. A numerical example for coated particulate composite material demonstrated the feasibility and effectiveness of proposed methods.

Multiscale analysis of a semi-periodically perforated plate based on a homogenization theory is experimentally validated both macroscopically and microscopically, using a full-field strain measurement. To do this, a plate-fin-type perforated plate with a misaligned microstructure is considered as a semi-periodically perforated plate. Then, a homogenization theory that can analyze macroscopic behavior and microscopic stress and strain distributions of the perforated plate is presented. To validate the theory, a tensile test of a plate-fin-type semi-periodically perforated plate made of epoxy resin is conducted. During the test, the microscopic deformation of the specimen is observed with a digital image correlation (DIC) full-field measurement system, from which microscopic strain distribution of the specimen is calculated. It is shown that the obtained strain distribution satisfies the unit-cell periodicity except at edges of the periodic structure, and that the strain distribution is in good agreement with the result of analysis using the homogenization theory. It is also shown that the macroscopic stress–strain relationships obtained by the experiment and analysis agree well, supporting the macroscopic and microscopic validity of the multiscale analysis using the homogenization theory.

Scars developing on body surfaces not only restrict body movement, but are also problematic from a cosmetic standpoint. Hence, revision is conducted by removing the scar and re-suturing the resultant defects. In performing scar revision, care should be taken to prevent the re-sutured wounds from developing hypertrophy again. Scars often present a pattern where hard, red parts are separated by soft parts in between. As the hard and soft parts may be analogized as islands and seas respectively, we call this the “Island-Like” scar. Two strategies can be taken to treat scars of this type. The first is to remove the entire scar—including both hard and soft parts; the second is to remove only the hard parts and leave the soft parts untouched. The authors conducted a biomechanical study using finite element analyses and found that as a body moves, greater stresses occur in the peri-wound regions with the first strategy than with the second strategy. A wound’s likelihood to develop hypertrophy increases as the stresses working on it increase. Hence, it is hypothesized that the second strategy carries less risk of the operated wounds developing re-hypertrophy than the first strategy. Based on this logic, in performing scar revision for scars consisting of hard and soft parts, it is recommended only to remove only hard parts and not to operate on soft parts in between.

Surgical success of drilling in oral implantology depends on the sense of force on the fingers or feeling of a dental clinician, which is related to the quality of trabecular bone in the jawbone expressed by apparent mechanical characteristics. As the mechanical properties of trabecular bone depend on the bone volume fraction, microstructure, and many other factors closely related to individual differences, a probabilistic numerical procedure to assess drilling force is proposed. Using a micro-CT-based jawbone model, a first-order perturbation-based stochastic homogenization method was employed to estimate the possible scattering of apparent mechanical properties of the trabecular bone region. The complicated drilling problem was simplified to sequential linear static FEAs, to which the predicted apparent Young’s modulus and shearing moduli in the drilling direction were applied. The FEAs demonstrated that the homogenized mechanical properties showed anisotropy, which might lead to differences in the drilling forces at different drilling angles. The numerically estimated drilling forces were shown by the expected value, 50 %-probability result, and 90%-probability result and revealed that one patient among two or ten patients would probably have poor bone quality. There was a remarkable difference in the drilling forces between the expected value and the 90%-probability result.

A conjoint analysis on a Web survey in 2010 with 160 participants and 8 questions shows that a Marginal Willingness To Pay (MWTP) on a Next Generation Network (NGN) of Flet’s Hikari Next is lower than its actual price as provided by NTT East. Conjoint analysis is one application of conditional logit. MWTP is WTP when one unit of property is increased, and MWTP is used to verify how much each person evaluates each property. Higher transmission speed, digital TV availability, and higher Internet security affect the MWTP positively, and the amount of the monthly payment of NGN affects the MWTP negatively. Therefore, it is assumed that the NGN of Flet’s Hikari Next by NTT East will be accepted by potential customers and diffused as a major communications service of NTT East and NTT West. NGN providers may be able to maximize their benefits by adjusting the NGN of Flet’s Hikari Next property to meet customer demands, especially in digital TV availability, higher transmission speed, and communications security.

Keloids gradually change their shapes as they grow. We hypothesize that the change of keloid morphology reflects the incremental change of the stress patterns occurring in peri-keloid regions due to movement of the keloid-carrying body part. To examine the validity of this hypothesis, we used three-dimensional finite element analysis to calculate the stresses occurring in the peri-keloid regions of keloids on the chest in response to respiratory movement. The stresses concentrate at the peri-keloid regions close to the bilateral ends of the keloids. By reviewing this result in reference to our hypothesis, we can explain why keloids on the chest are likely to present crab or butterfly shapes. Although we know that keloids grow in response to mechanical stresses, our hypothesis differs from existing ones in that it focuses on morphological transformation. Our hypothesis is helpful for physicians in performing treatment for keloids, because they can predict what part of a keloid is likely to grow and perform preventive treatment in reference to the hypothesis.

The aim of this study was to elucidate the association between the bone structure at implant insertion sites and stress distribution around the mandibular canal by means of three-dimensional finite element (3D FE) analysis. Four FE models were created with slice data using micro-computed tomography (micro-CT), and 3D FE analysis was performed. Mechanical analysis showed that the load reached the mandibular canal via the trabecular structure in all FE models. High levels of stress were generated around the mandibular canal when the distance between the mandibular canal and the implant decreased. High stress levels were also observed when cortical bone thickness and bone volume/total volume (BV/TV) were low. Our findings suggest that load is transmitted to the mandibular canal regardless of differences in the thickness of cortical bone or cancellous bone structure, but excessive load may be generated in bone with thin cortical and coarse cancellous structures.

The aim of this study was to investigate the biomechanical role of trabecular bone around dental implants in the mandible.The model in this study was made using micro-computed tomography data taken from a cadaver in whom endosseous implants had been in place for 15 years prior to death. Morphologic analysis and three-dimensional (3D) finite element analysis were performed to calculate the peri-implant loading path of the model in which the trabecular structure was accurately simulated.As seen through multiscale analysis using the homogenization method, the trabecular bone architecture around implants was isotropic for the most part. Also, 3D finite element analysis showed that compressive stresses oblique to the implant axis were transmitted to the lower constrained surface; tensile stresses oblique to the implant axis were transmitted to the upper constrained surface, and they intersected each other with vertical loading. The highest stress in cancellous bone was observed on perpendicular loading, and stress produced in trabeculae decreased approaching horizontal loading.Cancellous bone architecture around the implant was generally isotropic. 3D finite element analysis showed that cancellous bone trabeculae around implants dispersed stress by forming load transfer paths. The results suggest that trabecular bone plays a major role in supporting functional pressure exerted via the implant.

For the design of porous titanium component, differently from its use as functional materials, structural strength prediction algorithm is proposed based on the micro-stress concentration analyzed by the multiscale simulation. The modeling of real microstructure is carried out by the image-based technique with X-ray micro-CT. The multiscale computational method consists of asymptotic homogenization and finite element mesh superposition (FEMS) techniques. A criterion is proposed to predict the nonlinearity initiation point in the load and displacement curve by means of micro-stress distribution expressed by histogram. Only the constituent's yield strength is referred regardless of the microscopic morphology. The originality of this paper lies in the validation of the homogenization process, the modeling guideline of micro-mesh superposition onto macro-mesh in FEMS, and the strength prediction algorithm. L-shaped components with different pore diameter were discussed in both experiment and simulation.

This study is focused on the multi-scale modelling strategy for complex and heterogeneous microstructures of real materials by automatic image-based modelling and finite element mesh superposition method. The synergetic application of the conventional asymptotic homogenization method and the authors' mesh superposition method has been proposed to obtain the microscopic responses under high gradient of macroscopic fields at the macroscopic crack tip and/or interface, for instance. For complex and random microstructures, automatic image-based voxel meshing by means of x-ray CT is commonly required; however, it cannot always adapt to the mathematical theory of microscopic modelling in the mesh superposition method. Therefore, a modelling technique for mesh refinement is proposed in this paper using additional elements for insulation in consideration of the theoretical background of the mesh superposition method.In this paper, we provide the modelling procedure and its theoretical consideration of mesh refinement for flexible modelling of real materials. To demonstrate the technique, a numerical example of a porous ceramic component with random microstructure and macroscopic crack is illustrated.

Two-dimensional finite element models were created for a three-unit posterior fixed partial denture. An experimental resin-impregnated glass fiber was used as the fiber-reinforced composite (FRC) for the framework. The FRC was evaluated using varying combinations of position and thickness, alongside with two types of veneering composite. A load of 50 N simulating bite force was applied at the pontic in a vertical direction. Tensile stress was examined using a finite element analysis program. Model without FRC showed tensile stress concentrations within the veneering composite on the cervical side of the pontic--from the connector area to the bottom of the pontic. Model with FRC at the top of the pontic had almost the same stress distribution as the model without FRC. Models with 0.4-0.8 mm thick FRC positioned at the bottom of the pontic showed maximum tensile stresses reduced by 4-19% within the veneering composite.

The continuous fiber reinforced thermoplastic composites, such as textile composites, have many attractive properties and functions and they are used widely nowadays in industrial applications. The important point in the design and manufacturing is that the composites as well as the reinforcing textiles undergo large deformation during thermoforming, which is much influential on the overall properties. Hence, the large deformation characteristics must be investigated at thermoforming temperature. For this sake, the digital image-based measurement technique of large strain distribution is developed. It is applicable to arbitrary complex shaped components with the help of 3D-CAD. As the large strain definition, Biot's strain is adopted that is calculated by the polar decomposition of the deformation gradient tensor based on continuum mechanics. The developed strain measurement technique is applied to the deep-drawing of aramid rib-knitted fabric reinforced polypropylene composites. The comparison between rib and plain-knitted fabrics reinforced composites is studied and many differences are found. The large deformation characteristics are also investigated from microscopic point of view under uniaxial tension at thermoforming temperature.

This paper studies a three-scale computational method that simultaneously considers the microstructure of heterogeneous materials, the macroscopic component, and the fracture origin such as interface or crack. The synergetic application of the asymptotic homogenization and mesh superposition methods to problems with strong scale mixing is emphasized. The scale gap between the microstructure and the component is very large, but the fracture origin is at the middle scale between them. The overall behavior is analyzed by means of the homogenization of the heterogeneity expressed by the unit cell model, while the fracture origin is modeled directly with the microscopic heterogeneity by another microscopic mesh. The microscopic mesh is superposed onto the macroscopic mesh. This mesh superposition method can analyze the non-periodic microscopic stress at the crack tip under a non-uniform macroscopic strain field with high gradient. Hence, the present three-scale method can accurately focus on the behaviors at arbitrary scale differently from the conventional hierarchical model. A demonstrative example of porous thin film on a substrate with an interface crack was solved and the microscopic stress was analyzed at the crack tip considering the random dispersion of pores and the high gradient of macroscopic strain field. To solve the large-scale 3D problem with approximately 80,000 solid elements, a renumbering technique and the out-of-core skyline solver was employed.

Multi-scale analysis using the asymptotic homogenization method is becoming a matter of concern for microstructural design and analysis of advanced heterogeneous materials. One of the problems is the lack of the experimental verification of the multi-scale analysis. Hence, it is applied to the porous alumina with needle-like pores to compare the predicted homogenized properties with the experimental result. The complex and random microstructure was modeled three-dimensionally with the help of the digital image-based modeling technique. An appropriate size of the unit microstructure model was investigated. The predicted elastic properties agreed quite well with the measured values. Next, a four-point bending test was simulated and finally the microscopic stress distribution was predicted. However, it was very hard to evaluate the calculated microscopic stress quantitatively. Therefore, a numerical algorithm to help understanding the three-dimensional and complex stress distribution in the random porous microstructure is proposed. An original histogram display of the stress distribution is shown to be effective to evaluate the stress concentration in the porous materials.

To analyse the macroscopic and microscopic behaviours of heterogeneous materials and components, a multi-scale computational method is studied. Although asymptotic homogenization theory has been the main tool during the last decade to solve various multi-scale problems, the assumption of the periodicity of the microscopic unit cell and the incapability of considering the scale effect have resulted in the limitations to this theory's applications. These problems should be overcome because advanced materials are often used as joint or laminated components and the interface crack problem must be analysed. For this sake, a novel multi-scale finite element method is proposed that uses the enhanced mesh superposition method together with the asymptotic homogenization theory. The finite element mesh superposition method uses the global mesh and the local mesh that is superimposed arbitrarily onto the global mesh. The enhanced method allows the adoption of different constitutive laws for the two meshes. The advantage of the homogenization theory to predict the homogenized material model accurately based on the complex microstructure is still utilized. The homogenized material model is used for the global mesh, whilst the microscopic heterogeneity and the crack are considered in the local mesh with the material properties of the constituents. The formulation, modelling strategy, implementation and numerical accuracy of the proposed method is described. A porous ceramic is studied in the numerical example.

This paper proposes a novel, integrated, computational design methodology of graded microstructures of heterogeneous materials for the emergence of macroscopic function. In the first step of the design procedure, some discrete microstructures among a large number of graded microstructures are determined by the genetic algorithm as the optimization method. In this determination procedure, the homogenized modeling is adopted considering the micro-macro coupling by the homogenization method. This method enables us to study complex microstructures. Homogenized properties such as elastic properties, coefficient of thermal expansion and thermal conductivity can be calculated rigorously based on continuum mechanics. Calculated homogenized properties are stored in the micro-macro correlative database. The genetic algorithm can select the best geometrical arrangement of multiple microstructures from the pre-calculated database to construct the graded microstructure architecture. By using the database, the microscopic and macroscopic analyses are separated from each other, which reduces considerably the computational cost for the micro–macro coupled design. In the second step, continuously graded microstructures are designed using a feature-based 3D-CAD system by interpolating the discretized graded microstructures. In addition, a solid model is produced by the stereolithography technique to help in understanding the computationally designed complex microstructures. Brief descriptions of the formulation of the homogenization method for heat conduction and thermal stress problems are shown. A design problem for a plate with graded microstructures in its thickness direction is shown. The objective function for this example is the control of the wrap of the plate under the condition of temperature distribution.

The objective of this study is to propose an intelligent material system that can perform a self-repairing operation against initial damage that might occur in GFRP laminates. Specifically, we highlight the development of a novel actuator to be used in such self-repairing operations. There has been much research on sensor systems to detect damage and some actuators to control the shape of the structure or to control the vibration. However, no literature can be found on the actuator aiming at the self-repair of damage. For this purpose, a small grain particle-type adhesive is embedded in a glass/epoxy composite (GFRP) laminate. The diameter of the particle is approximately 50 gim. Hence, in the developed intelligent material system, coldsetting epoxy resin is used as the matrix, uni-directionally arranged glass fiber is used as the reinforcement, and a thermosetting epoxy particle is used as a repairing actuator. The volume fraction of the particles in the matrix was approximately 40%. The embedded particles can repair the damage, when melted by heat. Basic characteristics of the particles were investigated first, and we confirmed that the embedded particles in the matrix can melt by heat and flow to repair the crack. We also confirmed that the embedded particles do not deteriorate the stiffness of the GFRP laminate. Then we investigated the efficiency of the repairing operation against initial damage such as microscopic matrix cracks and delamination, by conducting two typical tests, i.e., static three-point bending of [0/90] laminate and tensile fatigue of [O/90], laminate. Damage was observed by CCD camera. As a result, the decrease in the stiffness due to initial damage has been recovered and consequently increased the residual fatigue life.

For the simulation of mechanical behaviours of textile composite materials and structures, a novel hierarchical modelling technique is proposed. Not only the global deformation but also the stresses at multiscales are analysed precisely. Four-level hierarchy is defined for the textile composites such as woven and knitted fabric composites. The stresses at the mesostructure, which is a periodic unit cell of textile composite materials consisting of fibre bundles and matrix, can be evaluated accurately by the homogenization method and finite-element mesh superposition technique. The latter technique makes it possible to overlay arbitrary local fine mesh on the global rough mesh. Anisotropic damage mechanics is also utilized for strength evaluation at the mesoscale. Three-dimensional modelling of the mesostructure of woven and knitted fabric composite materials is shown. Localization analysis has been carried out within practical computational time and cost by the proposed hierarchical modelling.

The aim of this study was to clarify the influence of shear modulus on the analytical accuracy in peri-implant jaw bone simulation. A 3D finite element (FE) model was prepared based on micro-CT data obtained from images of a jawbone containing implants. A precise model that closely reproduced the trabecular architecture, and equivalent models that gave shear modulus values taking the trabecular architecture into account, were prepared. Displacement norms during loading were calculated, and the displacement error was evaluated. The model that gave shear modulus values taking the trabecular architecture into account showed an analytical error of around 10-20% in the cancellous bone region, while in the model that used incorrect shear modulus, the analytical error exceeded 40% in certain regions. The shear modulus should be evaluated precisely in addition to the Young modulus when considering the mechanics of peri-implant trabecular bone structure.

We demonstrate the fabrication of polymer microneedle arrays using soft lithography. A photomask was designed to use Fresnel diffraction of UV light to create sharp, tapered hollows in SU-8, a negative photoresist, after development. Polymer microneedles were formed using these SU-8 structures as a mold. These polymer needles may be applicable as flexible electrodes in brain-machine interfaces because they are more likely to survive movement of the skin than conventional brittle silicon needles. Similar needles, made from medicinal substances, could be used for transdermal drug administration. For these applications, the needles must be long, sharp, and stiff enough to penetrate the stratum corneum (∼20 μm in thickness) and reach the viable epidermis (200-300 μm in thickness), but must not reach the dermis, which contains sensitive nerve endings. We successfully manufactured 20×20 microneedle arrays of polydimethylsiloxane with a needle length of 200 μm. We experimentally verified that these manufactured electrodes successfully penetrated the stratum corneum of a cultured skin.

Although the biomechanical behavior of the acetabular cup (AC) implant is determinant for the surgical success, it remains difficult to be assessed due to the multiscale and anisotropic nature of bone tissue. The aim of the present study was to investigate the influence of the anisotropic properties of peri-implant trabecular bone tissue on the biomechanical behavior of the AC implant at the macroscopic scale. Thirteen bovine trabecular bone samples were imaged using micro-computed tomography (μCT) with a resolution of 18 μm. The anisotropic biomechanical properties of each sample were determined at the scale of the centimeter based on a dedicated method using asymptotic homogenization. The material properties obtained with this multiscale approach were used as input data in a 3D finite element model to simulate the macroscopic mechanical behavior of the AC implant under different loading conditions. The largest stress and strain magnitudes were found around the equatorial rim and in the polar area of the AC implant. All macroscopic stiffness quantities were significantly correlated (R