Computer-aided Numerical Analysis Research Articles

Deep learning for diagnosis of malign pleural effusion on computed tomography images

BackgroundThe pleura is a serous membrane that surrounds the lungs. The visceral surface secretes fluid into the serous cavity and the parietal surface ensures a regular absorption of this fluid. If this balance is disturbed, fluid accumulation occurs in the pleural space called “Pleural Effusion”. Today, accurate diagnosis of pleural diseases is becoming more critical, as advances in treatment protocols have contributed positively to prognosis. Our aim is to perform computer-aided numerical analysis of Computed Tomography (CT) images from patients showing pleural effusion images on CT and to examine the prediction of malignant/benign distinction using deep learning by comparing with the cytology results. MethodsThe authors classified 408 CT images from 64 patients whose etiology of pleural effusion was investigated using the deep learning method. 378 of the images were used for the training of the system; 15 malignant and 15 benign CT images, which were not included in the training group, were used as the test. ResultsAmong the 30 test images evaluated in the system; 14 of 15 malignant patients and 13 of 15 benign patients were estimated with correct diagnosis (PPD: 93.3%, NPD: 86.67%, Sensitivity: 87.5%, Specificity: 92.86%). ConclusionAdvances in computer-aided diagnostic analysis of CT images and obtaining a pre-diagnosis of pleural fluid may reduce the need for interventional procedures by guiding physicians about which patients may have malignancies. Thus, it is cost and time-saving in patient management, allowing earlier diagnosis and treatment.

Biomechanical comparison using finite element analysis of different screw configurations in the fixation of femoral neck fractures

Abstract. In this research, biomechanical behaviors of five different configurations of screws used for stabilization of femoral neck fracture under axial loading have been examined, and which configuration is best has been investigated. A point cloud was obtained after scanning the human femoral model with a three dimensional (3-D) scanner, and this point cloud was converted to a 3-D femoral model by Geomagic Studio software. Femoral neck fracture was modeled by SolidWorks software for five different configurations: dual parallel, triple parallel, triangle, inverted triangle and square, and computer-aided numerical analysis of different configurations were carried out by ANSYS Workbench finite element analysis (FEA) software. For each configuration, mesh process, loading status (axial), boundary conditions and material model were applied in finite element analysis software. Von Mises stress values in the upper and lower proximity of the femur and screws were calculated. According to FEA results, it was particularly advantageous to use the fixation type of triangle configuration. The lowest values are found as 223.32 MPa at the lower, 63.34 MPa at the upper proximity and 493.24 MPa at the screws in triangle configuration. This showed that this configuration creates minimum stress at the upper and lower proximity of the fracture line. Clinically, we believe that the lowest stress values which are created by triangle configuration encompass the most advantageous method. In clinical practices, it is believed that using more than three screws does not provide any benefit. Furthermore, the highest stresses are as follows: at upper proximity 394.79 MPa in triple parallel configuration, for lower proximity 651.2 MPa in square configuration and for screw 2459 MPa in inverted triangle.

Transparent color-coded three-dimensional diffusion-weighted magnetic resonance imaging for the numerical analysis of reversible and irreversible hyperacute ischemic stroke—case report

Although diffusion-weighted (DW) magnetic resonance (MR) imaging can detect hyperacute ischemic parenchyma with high sensitivity, the ability of DW images to reveal subtle change in abnormal diffusion may be limited by the conventional visual evaluation. To overcome the limitation, we have developed a method of transparent color-coded three-dimensional (3D) DW MR imaging for the computer-aided numerical analysis of hyperacute ischemic stroke. The 3D images were reconstructed from volume data of source DW images by using a parallel volume-rendering algorithm with transluminal imaging technique. By selecting a threshold range from a signal intensity opacity chart of volume-rendering data set, several high signal intensity areas were depicted and assigned to different colors, transparently through contours of the brain. This imaging was applied in a case of a recanalized middle cerebral artery (M2) occlusion with partially reversible ischemic parenchyma accompanied by partial recovery from ischemic neurological deficit. Complex and dynamic change in hyperacute ischemic parenchyma, with regression of subtle high signal intensity areas and progression of ischemic parenchyma, was depicted three-dimensionally. Transparent color-coded 3D DW MR imaging may provide computer-aided numerical analysis of hyperacute ischemic stroke appearing as a high signal intensity area on the source DW images.

Study of microbial community structures in UASB sludge treating municipal wastewater by denaturing gradient gel electrophoresis of 16S rDNA

The structures of microbial communities in lab-scale upflow anaerobic sludge blanket (UASB) reactors for treating municipal wastewater with different ratios of COD soluble/COD total were studied using denaturing gradient gel electrophoresis (DGGE) of 16S rRNA genes. The microbial structure of the inoculum sludge obtained from a full-scale UASB reactor of treating potato processing wastewater was compared with the structures of sludges collected from three lab-scale UASB reactors after eight months feeding with raw municipal wastewater, with CEPS (chemically enhanced primary sedimentation) pretreated municipal wastewater, and with a synthetic municipal sewage, respectively. Computer-aided numerical analysis of the DGGE fingerprints showed that the bacterial community underwent major changes. The sludges for treating raw and CEPS pretreated wastewater had very similar bacterial and archaeal communities (82% and 96% similarity) but were different from that for treating the synthetic sewage. Hence, despite similar % COD in the particulate form in the synthetic and the real wastewater, the two wastewaters were selected for different microbial communities. Prominent DGGE bands of Bacteria and Archaea were purified and sequenced. The 16S rRNA gene sequences of the dominant archaeal bands found in the inoculum, and UASB sludge fed with raw sewage, CEPS pretreated wastewater, and synthetic sewage were closely associated with Methanosaeta concilii. In the UASB sludge fed with synthetic sewage, another dominant band associated with an uncultured archaeon 39-2 was found together with M. concilii.

Analysis of Flows in a Heater Unit of an Automobile Air Conditioning System.

Facilitating cool-hot air mixing inside the heater unit is one of important factors when considering the design of air conditioning systems. In order to produce efficient mixing in the limited space inside the heater unit, conventional guides are provided to control mixing. Because of the difficulties of detailed analysis of air flow inside the heater unit, the shapes and installation positions of these guides were determined empirically. The development of experimental flow analysis utilizing the latest image processing technology and computer-aided numerical analysis contribute remarkably to the design of air conditioning system based upon qualitative and quantitative analysis of actual flow phenomena. In this report, laser-assisted flow visualization and the velocity distribution measuring technique using the speckle method are applied to an experimental analysis of guide effects to cool-hot air mixing in a heater unit model. Additionally, the results of 3-dimensional numerical analysis of the heater unit air flow using standard software are compared with the experimental ones to verify the validity of the method as an effective tool for the heater unit air flow analysis.

Computer-Aided Numerical Analysis and Design for Cold Extrusion of a Spur Gear

The die profile for extrusion of a spur gear section through a continuously curved die is described by a parametric form in the cylindrical coordinate system using a blending function and Fourier series expansion. A finite element analysis is made for steady state three-dimensional extrusion of a spur gear section. The distribution of stress and strain components is determined. In treating the incompressibility, the penalty constraint method based on the rigid-plastic material model is used. The initial estimate for the initial velocity field is obtained from the assumption of linear viscous material. The work hardening effect is considered by integrating the effective strain rate along each streamline through interpolation by the least-square method. The elastic deformation of the die has been computed by the elastic finite element analysis. It is shown that the computer aided analysis and design of the extrusion process could help to achieve more precise production of gears with improved productivity.

Demonstration of excitation-dependent grain-boundary recombination velocity in polycrystalline silicon

The dependence of grain-boundary recombination velocity on the excitation (excess carrier density) at the grain boundary is experimentally demonstrated using electron-beam-induced-current (EBIC) measurements. To facilitate quantitative interpretation of the EBIC measurements, the underlying carrier transport problem in the adjacent grains, subject to the nonlinear boundary condition at the grain boundary, is solved using computer-aided numerical analysis. By comparing the theoretical and experimental EBIC responses, values for the surface-state density at unpassivated and passivated grain boundaries and for the diffusion length in the grains in Wacker (cast) polycrystalline silicon are derived.

Copper impurity levels in silicon

This paper presents a new model of copper impurity levels in silicon, and suggests a new method to analyze the parameters of multilevel impurities using Hall data. On the assumption that the energy levels of copper do not depend on the conduction type of silicon and the concentration of copper, an impurity level model which can explain all the Hall data of many different copper-doped silicon samples is proposed using a computer-aided numerical analysis. Experimental data and analysis indicate that copper exists in silicon in the form of four independent species, viz. substitutional species (Cu I), interstitial species (Cu IV) and two copper-associated species Cu II and Cu III. The usual species Cu I has three acceptor levels and a donor level below the intrinsic level. Cu II has two levels, an acceptor and a donor level respectively, below the donor level of Cu I. Cu III produces an acceptor level at a slightly distant position from the conduction band edge. That the intersitial species Cu IV behaves as a shallow donor is well-known, but its behavior is predominant only in samples diffused at temperatures below 900°C because of the precipitation of its species. Not only the energy levels of copper, but also the degeneracy ratios associated with these levels have been simultaneously determined, and it has been indicated that the degeneracy ratios have an important effect on the temperature dependence of the carrier concentration.

Analysis of minority-carrier transport in polysilicon devices

Key assumptions are made, with justification, to simplify the three-dimensional, nonlinear boundary-value problem that defines minority-carrier transport, including recombination, in polysilicon devices. These assumptions enable the separation of the grain-boundary recombination analysis, which is based on quasi-equilibrium in the space-charge region, from the intragrain transport analysis, which is done by partitioning the grain into subregions in which the minority-carrier flow is predominantly one-dimensional. The analyses are coupled through the effective minority-carrier recombination velocity at the grain boundary, which generally is dependent on the minority-carrier density in the quasi-neutral grain. Limitations of the model implied by the quasi-equilibrium assumption are effectively removed by recognizing that when conditions obtain that negate quasi-equilibrium, the effective recombination velocity is fixed at the minority-carrier kinetic-limit velocity. The model development is facilitated by computer-aided numerical analysis of the grain-boundary recombination and is supported by qualitative discussion of the underlying physics.

The effects of caudocephalad (+ Gz) acceleration on the initially curved human spine

A beam-column model of the human spine subjected to a caudocephalad (+ G z ) acceleration is analyzed by the finite-difference numerical technique. It was shown that the previous analytical treatment of this problem by the assumed-mode method is valid only for either very low levels of the acceleration pulse or for very early times in the response, i.e. the loading is such that the initial configuration of the spine is little changed by the dynamics. Numerical results generated, using a 20 g step acceleration input, show that the initial configuration is so appreciably changed as to invalidate the results of the assumed-mode analysis. Any future extensions of similar continuum models of the human spine, to include other additional effects, should treat the use of classical analysis with caution and consider the efficacy of computer-aided numerical analysis by either the present finite-difference solution or the currently popular finite-element method.

A Dominant Mode Analysis of Microstrip (Short Papers)

A computer-aided numerical analysis of the dominant mode propagating in microstrip transmission line is reported, and the theoretical results are corroborated by experimental measurements. A region of almost complete circular polarization is found to exist at the surface of the dielectric, although most of the energy is concentrated directly beneath the strip conductor where the fields possess almost complete linear polarization.

An approach to the classification of lactobacilli using computer-aided numerical analysis.

An electronic computer was used to analyze the data obtained for 82 strains of streptococci, lactobacilli, and propionibacteria. The lactobacilli were divided into three major taxonomic groups which correspond to the subgenera Thermobacterium, Streptobacterium, and Betabacterium described by Orla-Jensen (14). Three clusters observed among the streptococcal strains also correlate with Sherman's (19) classification of pyogenic, fecal, and lactic groups of streptococci. The results are in agreement with present classifications. No justification for including the genus Lactobacillus in the same family as genus Propionibacterium was found.