Hyperspherical Variational Co-embedding for Attributed Networks

Network-based information has been widely explored and exploited in the information retrieval literature. Attributed networks, consisting of nodes, edges as well as attributes describing properties of nodes, are a basic type of network-based data, and are especially useful for many applications. Examples include user profiling in social networks and item recommendation in user-item purchase networks. Learning useful and expressive representations of entities in attributed networks can provide more effective building blocks to down-stream network-based tasks such as link prediction and attribute inference. Practically, input features of attributed networks are normalized as unit directional vectors. However, most network embedding techniques ignore the spherical nature of inputs and focus on learning representations in a Gaussian or Euclidean space, which, we hypothesize, might lead to less effective representations. To obtain more effective representations of attributed networks, we investigate the problem of mapping an attributed network with unit normalized directional features into a non-Gaussian and non-Euclidean space. Specifically, we propose a hyperspherical variational co-embedding for attributed networks (HCAN), which is based on generalized variational auto-encoders for heterogeneous data with multiple types of entities. HCAN jointly learns latent embeddings for both nodes and attributes in a unified hyperspherical space such that the affinities between nodes and attributes can be captured effectively. We argue that this is a crucial feature in many real-world applications of attributed networks. Previous Gaussian network embedding algorithms break the assumption of uninformative prior, which leads to unstable results and poor performance. In contrast, HCAN embeds nodes and attributes as von Mises-Fisher distributions, and allows one to capture the uncertainty of the inferred representations. Experimental results on eight datasets show that HCAN yields better performance in a number of applications compared with nine state-of-the-art baselines.

Mechanical Strength Study of a Cranial Implant Using Computational Tools

The human head is sometimes subjected to impact loads that lead to skull fracture or other injuries that require the removal of part of the skull, which is called craniectomy. Consequently, the removed portion is replaced using autologous bone or alloplastic material. The aim of this work is to develop a cranial implant to fulfil a defect created on the skull and then study its mechanical performance by integrating it on a human head finite element model. The material chosen for the implant was PEEK, a thermoplastic polymer that has been recently used in cranioplasty. A6 numerical model head coupled with an implant was subjected to analysis to evaluate two parameters: the number of fixation screws that enhance the performance and ensure the structural integrity of the implant, and the implant’s capacity to protect the brain compared to the integral skull. The main findings point to the fact that, among all tested configurations of screws, the model with eight screws presents better performance when considering the von Mises stress field and the displacement field on the interface between the implant and the skull. Additionally, under the specific analyzed conditions, it is observable that the model with the implant offers more efficient brain protection when compared with the model with the integral skull.

Fluid-structure interaction simulation of tissue degradation and its effects on intra-aneurysm hemodynamics

Tissue degradation plays a crucial role in vascular diseases such as atherosclerosis and aneurysms. Computational modeling of vascular hemodynamics incorporating both arterial wall mechanics and tissue degradation has been a challenging task. In this study, we propose a novel finite element method-based approach to model the microscopic degradation of arterial walls and its interaction with blood flow. The model is applied to study the combined effects of pulsatile flow and tissue degradation on the deformation and intra-aneurysm hemodynamics. Our computational analysis reveals that tissue degradation leads to a weakening of the aneurysmal wall, which manifests itself in a larger deformation and a smaller von Mises stress. Moreover, simulation results for different heart rates, blood pressures and aneurysm geometries indicate consistently that, upon tissue degradation, wall shear stress increases near the flow-impingement region and decreases away from it. These findings are discussed in the context of recent reports regarding the role of both high and low wall shear stress for the progression and rupture of aneurysms.

Effects of Revision Rod Position on Spinal Construct Stability in Lumbar Revision Surgery: A Finite Element Study

Revision surgery (RS) is a necessary surgical intervention in clinical practice to treat spinal instrumentation–related symptomatic complications. Three constructs with different configurations have been applied in RS. One distinguishing characteristic of these configurations is that the revision rods connecting previous segments and revision segments are placed alongside, outside, or inside the previous rods at the level of facetectomy. Whether the position of the revision rod could generate mechanical disparities in revision constructs is unknown. The objective of this study was to assess the influence of the revision rod position on the construct after RS. A validated spinal finite element (FE) model was developed to simulate RS after previous instrumented fusion using a modified dual-rod construct (DRCm), satellite-rod construct (SRC), and cortical bone trajectory construct (CBTC). Thereafter, maximum von Mises stress (VMS) on the annulus fibrosus and cages and the ligament force of the interspinous ligament, supraspinous ligament, and ligamentum flavum under a pure moment load and a follower load in six directions were applied to assess the influence of the revision rod position on the revision construct. An approximately identical overall reducing tendency of VMS was observed among the three constructs. The changing tendency of the maximum VMS on the cages placed at L4-L5 was nearly equal among the three constructs. However, the changing tendency of the maximum VMS on the cage placed at L2-L3 was notable, especially in the CBTC under right bending and left axial rotation. The overall changing tendency of the ligament force in the DRCm, SRC, and CBTC was also approximately equal, while the ligament force of the CBTC was found to be significantly greater than that of the DRCm and SRC at L1-L2. The results indicated that the stiffness associated with the CBTC might be lower than that associated with the DRCm and SRC in RS. The results of the present study indicated that the DRCm, SRC, and CBTC could provide sufficient stabilization in RS. The CBTC was a less rigid construct. Rather than the revision rod position, the method of constructing spinal instrumentation played a role in influencing the biomechanics of revision.

Modal Analysis of Conceptual Microsatellite Design Employing Perforated Structural Components for Mass Reduction

Mass reduction is a primary design goal pursued in satellite structural design, since the launch cost is proportional to their total mass. The most common mass reduction method currently employed is to introduce honeycomb structures, with space qualified composite materials as facing materials, into the structural design, especially for satellites with larger masses. However, efficient implementation of these materials requires significant expertise in their design, analysis, and fabrication processes; moreover, the material procurement costs are high, therefore increasing the overall program costs. Thus, the current work proposes a low-cost alternative approach through the design and implementation of geometrically-shaped, parametrically-defined metal perforation patterns, fabricated by standard processes. These patterns included four geometric shapes (diamonds, hexagons, squares, and triangles) implemented onto several components of a structural design for a conceptual satellite, with a parametric design space defined by two scale factors and also two aspect ratio variations. The change in the structure’s fundamental natural frequency, as a result of implementing each pattern shape and parameter variation, was the selection criterion, due to its importance during the launcher selection process. The best pattern from among the four alternatives was then selected, after having validated the computational methodology through implementing experimental modal analysis on a scaled down physical model of a primary load-bearing component of the structural design. From the findings, a significant mass reduction percentage of 23.15%, utilizing the proposed perforation concept, was achieved in the final parametric design iteration relative to the baseline unperforated case while maintaining the same fundamental frequency. Dynamic loading analysis was also conducted, utilizing both the baseline unperforated and the finalized perforated designs, to check its capability to withstand realistic launch loads through applying quasi-static loads. The findings show that the final perforated design outperformed the baseline unperforated design with respect to the maximum displacements, maximum Von Mises stresses, and also the computed margin of safety. With these encouraging outcomes, the perforated design concept proved that it could provide an opportunity to develop low-cost satellite structural designs with reduced mass.

Biomechanical feasibility of semi-rigid stabilization and semi-rigid lumbar interbody fusion: a finite element study

Background Semi-rigid lumbar fusion offers a compromise between pedicle screw-based rigid fixation and non-instrumented lumbar fusion. However, the use of semi-rigid interspinous stabilization (SIS) with interspinous spacer and ligamentoplasty and semi-rigid posterior instrumentation (SPI) to assist interbody cage as fusion constructs remained controversial. The purpose of this study is to investigate the biomechanical properties of semi-rigidly stabilized lumbar fusion using SIS or SPI and their effect on adjacent levels using finite element (FE) method. Method Eight FE models were constructed to simulate the lumbosacral spine. In the non-fusion constructs, semi-rigid stabilization with (i) semi-rigid interspinous spacer and artificial ligaments (PD-SIS), and (ii) PI with semi-rigid rods were simulated (PD + SPI). For fusion constructs, the spinal models were implanted with (iii) PEEK cage only (Cage), (iv) PEEK cage and SIS (Cage+SIS), (v) PEEK cage and SPI (Cage+SPI), (vi) PEEK cage and rigid PI (Cage+PI). Result The comparison of flexion-extension range of motion (ROM) in the operated level showed the difference between Cage+SIS, Cage+SPI, and Cage+PI was less than 0.05 degree. In axial rotation, ROM of Cage+SIS were greater than Cage+PI by 0.81 degree. In the infrajacent level, while Cage+PI increased the ROM by 24.1, 27,7, 25.9, and 10.3% and Cage+SPI increased the ROM by 26.1, 30.0, 27.1, and 10.8% in flexion, extension, lateral bending and axial rotation respectively, Cage+SIS only increased the ROM by 3.6, 2.8, and 11.2% in flexion, extension, and lateral bending and reduced the ROM by 1.5% in axial rotation. The comparison of the von Mises stress showed that SIS reduced the adjacent IVD stress by 9.0%. The simulation of the strain energy showed a difference between constructs less than 7.9%, but all constructs increased the strain energy in the infradjacent level. Conclusion FE simulation showed semi-rigid fusion constructs including Cage+SIS and Cage+SPI can provide sufficient stabilization and flexion-extension ROM reduction at the fusion level. In addition, SIS-assisted fusion resulted in less hypermobility and less von Mises stress in the adjacent levels. However, SIS-assisted fusion had a disadvantage of less ROM reduction in lateral bending and axial rotation. Further clinical studies are warranted to investigate the clinical efficacy and safety of semi-rigid fusions.

A discrete Ramos-Louzada distribution for asymmetric and over-dispersed data with leptokurtic-shaped: Properties and various estimation techniques with inference

<abstract> <p>In this paper, a flexible probability mass function is proposed for modeling count data, especially, asymmetric, and over-dispersed observations. Some of its distributional properties are investigated. It is found that all its statistical and reliability properties can be expressed in explicit forms which makes the proposed model useful in time series and regression analysis. Different estimation approaches including maximum likelihood, moments, least squares, Andersonӳ-Darling, Cramer von-Mises, and maximum product of spacing estimator, are derived to get the best estimator for the real data. The estimation performance of these estimation techniques is assessed via a comprehensive simulation study. The flexibility of the new discrete distribution is assessed using four distinctive real data sets ԣoronavirus-flood peaks-forest fire-Leukemia? Finally, the new probabilistic model can serve as an alternative distribution to other competitive distributions available in the literature for modeling count data.</p> </abstract>