Interpretation of MR images using self-organizing maps and knowledge-based expert systems

A new image segmentation system is presented to automatically segment and label brain magnetic resonance (MR) images to show normal and abnormal brain tissues using self-organizing maps (SOM) and knowledge-based expert systems. Elements of a feature vector are formed by image intensities, first-order features, texture features extracted from gray-level co-occurrence matrix and multiscale features. This feature vector is used as an input to the SOM. SOM is used to over segment images and a knowledge-based expert system is used to join and label the segments. Spatial distributions of segments extracted from the SOM are also considered as well as gray level properties. Segments are labeled as background, skull, white matter, gray matter, cerebrospinal fluid (CSF) and suspicious regions.     

Photosensitized free radical polymerization using pyridinium salts

Summary The polymerization of methyl methacrylate (MMA), ethyl acrylate (EA), styrene (St) and 2-vinyl pyridine (VP) is initiated upon irradiation at >350 nm of dichloromethane solutions containing N-ethoxy-2-methylpyridinium hexafluorophosphate (EMP⁺PF₆ ^-) and anthracene or thioxanthone. Initiation mechanisms involving the formation of ethoxyl radicals during the decomposition of EMP⁺ ions via electron transfer are proposed.     

Evaluation of hydrogen bonding ability of liquids and solids by C-13 NMR. Silica gel as a strong hydrogen bond donor

The chemical shift difference between signals of C() and C() of unsaturated ketones, , which we used before to measure acid strengths, has now been used to evaluate the hydrogen bond donor ability of solvents which are not acidic enough to hydronate the indicator. For such solvents there is no general correlation between H-bond donor ability and acid strength: hexa-fluoroisopropanol is a much weaker acid than acetic acid, but it is a stronger H-bond donor. The method can be applied to evaluate the H-bonding properties of solid surfaces, and it was thus found that silica gel has a much stronger H-bond donor ability than methanol or acetic acid.     

Investigating the influence of tool shape during FSW of aluminum alloy via CFD analysis

This paper describes the application of the CFD code, Comsol Multiphysics, to modeling the 3-D metal flow in friction stir welding of AA 2024-T3 aluminum alloy in order to investigate the influence of tool shape over the metal flow. Heat transfer and non-Newtonian flow equations were solved simultaneously. The results from the benchmark experiments found in the literature were used for validation purposes. Scrolled shoulders and threaded pins were given as kinematic boundary conditions. This made the computational problem an easy one. A welding engineer can predict the metal flow around the tool with different scrolls and threads under any welding conditions without making expensive experiments. Investigation of the velocity field before actual welding can save a lot of engineering hours.     

Dynamic material properties of wood subjected to low-velocity impact

This article studied the effects of low-velocity impact on the failure stresses and stiffness using a pendulum test. The specimens were of variable depth (20, 30, and 40 mm), a width of 50 mm, length of 650 mm, and span-length of 480 mm. The smallest specimen depth was similar to specimen sizes tested in the literature used to create the duration of load curve, while the largest specimen depth are considered structural size specimens. The impact was predicted using a numerical approach with Euler–Bernoulli beam, as well as Timoshenko beam theory, with a plastic contact law. The models were validated for impact from a low release-angle (where the beam remained elastic), but could use improvement for the force prediction at a high incidence velocity. The measured force signals were used as forcing functions to obtain the dynamic failure stresses for all of the evaluated specimens, and the Timoshenko–Goens–Hearmon Method to derive the dynamic E. The resulting strain rates ranged from 9.11?×?10^?5 s^?1 for the quasi-static specimens up to 25 s^?1 for the greatest incidence velocity. The results from this study suggest different duration of load factors than the Madison Curve, influencing the design of structures subjected to dynamic loading.     

The axisymmetric crack problem in a non-homogeneous interfacial region between homogeneous half-spaces

In this paper, the axisymmetric crack problem in a non-homogeneous interfacial region between two homogeneous half-spaces is considered. It is assumed that the shear modulus varies continuously between that of the two half-spaces; and the shear modulus for the interface region is approximated by = ₀ e^mz. By using Hankel transform technique the problem is reduced to a pair of singular integral equations. The solutions of the problem are obtained for different material combinations and loading conditions; and modes I and II stress intensity factors, and the direction of a probable crack growth are calculated.     

Efficient monitoring of parametric context-free patterns

Recent developments in runtime verification and monitoring show that parametric regular and temporal logic specifications can be efficiently monitored against large programs. However, these logics reduce to ordinary finite automata, limiting their expressivity. For example, neither can specify structured properties that refer to the call stack of the program. While context-free grammars (CFGs) are expressive and well-understood, existing techniques for monitoring CFGs generate large runtime overhead in real-life applications. This paper demonstrates that monitoring parametric CFGs is practical (with overhead on the order of 12% or lower in most cases). We present a monitor synthesis algorithm for CFGs based on an LR(1) parsing algorithm, modified to account for good prefix matching. In addition, a logic-independent mechanism is introduced to support matching against the suffixes of execution traces.     

A new perspective on data homogeneity in software cost estimation: a study in the embedded systems domain

Cost estimation and effort allocation are the key challenges for successful project planning and management in software development. Therefore, both industry and the research community have been working on various models and techniques to accurately predict the cost of projects. Recently, researchers have started debating whether the prediction performance depends on the structure of data rather than the models used. In this article, we focus on a new aspect of data homogeneity, “cross- versus within-application domain”, and investigate what kind of training data should be used for software cost estimation in the embedded systems domain. In addition, we try to find out the effect of training dataset size on the prediction performance. Based on our empirical results, we conclude that it is better to use cross-domain data for embedded software cost estimation and the optimum training data size depends on the method used.     

Prediction of compressive strength of heavyweight concrete by ANN and FL models

The compressive strength of heavyweight concrete which is produced using baryte aggregates has been predicted by artificial neural network (ANN) and fuzzy logic (FL) models. For these models 45 experimental results were used and trained. Cement rate, water rate, periods (7–28–90 days) and baryte (BaSO₄) rate (%) were used as inputs and compressive strength (MPa) was used as output while developing both ANN and FL models. In the models, training and testing results have shown that ANN and FL systems have strong potential for predicting compressive strength of concretes containing baryte (BaSO₄).