Experimental Study of Ice Jam Thickening under Dynamic Flow Conditions

River ice jams are a common occurrence on northern rivers, and their formation can present a severe flood risk to nearby communities. As more and more river regulation projects are developed to provide an alternative to fossil fuels for electrical power-generating capacity, our need to understand the mechanisms associated with ice jam formation under variable flow conditions becomes more vital. This is because, at present, hydropeaking operations are often severely curtailed during the ice-affected seasons due to concerns that sudden flow fluctuations might instigate ice jams and associated flooding. Here, an experimental investigation explores the effects of rapid increases in discharge on ice jam formation and evolution. It is found that the thickness of ice jams formed under highly dynamic flow conditions tend to be slightly thinner than those formed during steady carrier flows for comparable discharges. Also, despite the highly dynamic nature of these consolidation events, the resulting ice thicknesses appear reasonably well approximated by steady flow theory.     

Acquiring rule sets as a product of learning in a logical neuralarchitecture

Envisioning neural networks as systems that learn rules calls forth the verification issues already being studied in knowledge-based systems engineering, and complicates these with neural-network concepts such as nonlinear dynamics and distributed memories. We show that the issues can be clarified and the learned rules visualized symbolically by formalizing the semantics of rule-learning in the mathematical language of two-valued predicate logic. We further show that this can, at least in some cases, be done with a fairly simple logical model. We illustrate this with a combination of two example neural-network architectures, LAPART, designed to learn rules as logical inferences from binary data patterns, and the stack interval network, which converts real-valued data into binary patterns that preserve the semantics of the ordering of real values. We discuss the significance of the formal model in facilitating the analysis of the underlying logic of rule-learning and numerical data representation. We provide examples to illustrate the formal model, with the combined stack interval/LAPART networks extracting rules from numerical data.     

Passive dissolution kinetics of titanium in vitro

The dissolution of titanium in simulated interstitial electrolyte (SIE), human serum in SIE (serum/SIE) and 8.0 mM ethylenediaminetetraacetic acid (EDTA) in SIE (EDTA/SIE) was measured in vitro. Titanium fibre samples were immersed in these solutions and maintained at 37°C, 10% O₂, 5% CO₂ and 97±3% relative humidity for 0–5000 h. The concentration of titanium released was quantified using electrothermal atomic absorption spectroscopy. Changes in oxide stoichiometry were determined by Auger electron spectroscopy after processing and immersion in the test solutions. The oxide became nearly stoichiometric TiO₂ after immersion, suggesting equilibration of the surface with the solutions. Solution ligands enhanced the magnitude of dissolution, with EDTA>serum/SIE>SIE. The dissolution kinetics were empirically fitted by a two-phase logarithmic relationship. The first phase of dissolution (t<300 h) was dominated by equilibration of the oxide with the solution and the second phase (t>300 h) by mass diffusion. The dissolution kinetics were similar for the EDTA/SIE and serum/SIE solutions, indicating that the mechanisms of dissolution for each solution may be the same.     

Computational Methods for Cardiac Electromechanics

Computational modeling provides a potentially powerful way to integrate structural properties measured in vitro to physiological functions measured in vivo. Focusing on the various scales (cell-tissue-organ-system), we give an overview of the importance and applications of numerical models of ventricular anatomy, electrophysiology, mechanics, and circulatory models. The integration of these models in one multiscale model of cardiac electromechanics is discussed in the light of applications to hypothesis generation, diagnosis, surgery(planning, training, and outcome of interventions), and therapies. Special attention is paid to practical use in terms of computational demand. Because of growing computer power and the development of efficient algorithms, we expect that real-time simulations with multiscale models of cardiac electromechanics become feasible in 2008 (despite the increasing complexity of models due to data accumulation on molecular and cellular mechanisms).     

Femoral artery pseudoaneurysms: risk factors, prevalence, and treatment options

Femoral artery pseudoaneurysms (PSA or false aneurysm) develop in about 1% of patients undergoing procedures that require femoral artery cannulation. PSA prolong hospitalization, consuming health-care resources, and result in significant morbidity. We designed a study to review the prevalence, risk factors, and treatment of PSA. We performed a retrospective, case-controlled study of patients diagnosed with PSA at a large, urban, tertiary-care teaching hospital. We reviewed 48 patients with PSA for our study. The estimated prevalence of PSA was 0.28%, with identical rates found for procedures done in the interventional radiology department and in the cardiac catheterization suite. Logistic regression identified three independent risk factors for developing PSA: being female (odds ratio 2.62), having an intervention performed (odds ratio 3.22), and not having a closure device used (odds ratio 10.2). Patients with PSA had longer hospital length of stay than that of patients without PSA. We found no statistically significant difference in failure or complication rates for the four treatment options and that spontaneous resolution of PSA does not correlate with its size. Thrombin injection and observation are effective, low-complication treatment options.     

Assessment of abdominal lymph nodes in a normal paediatric population: an ultrasound study

During a 9 month period, 50 consecutive children were evaluated by ultrasound to determine the size, location (mesenteric vs para-aortic), number, shape and texture of abdominal lymph nodes in a normal paediatric population. High resolution linear array transducers were used with graded compression. Nodes ranging from 10 to 20 mm were recorded in the majority of subjects. In all cases mesenteric lymph nodes were larger and more numerous than para-aortic nodes. Para-aortic lymph nodes were not seen in isolation.     

Electrosynthesis in systems of two immiscible liquids and a phase transfer catalyst. V. The anodic chlorination of naphthalene

The anodic chlorination of naphthalene in a water/methylene chloride emulsion and using tetrabutylammonium ion as the phase transfer catalyst is demonstrated; in conditions where the aqueous phase is saturated NaCl, the organic yield of 1-chloronaphthalene is 56% and the current yield is 33% after the passage of 2.33 F mol^–1 of naphthalene. It is shown, however, that when the aqueous phase also contains zinc chloride so that the species transferred is [(C₄H₉)₄N⁺]₂ZnCl ₄ ^–– , the yields can be increased to 92% and 49% respectively. The mechanism of these chlorinations is discussed.     

Development of a Cloud Trust Label: A Delphi Approach

The purpose of this research is to identify the potential information components of an online, real-time trust label, which is proposed as a communication mechanism to encourage trust in cloud service providers and cloud computing products. An online Delphi process was used with 28 cloud computing experts (including vendors, software providers, and legal and business representatives). The proposed label contains 81 information components, covering the cloud service provider (e.g. physical location, legal jurisdiction), the cloud service itself (e.g. data location, security, backup, certification), and a historical service-level summary (e.g. uptime data, support response times). The potential benefits of such a label to encourage trustworthiness perceptions and trust behaviors in the cloud computing environment are explored. Limitations of the study are highlighted, and further research studies are suggested to test the concept of the label and to refine the components of the label itself.     

Deriving engineering software from requirements

With the advent of intelligent computer aided design systems, companies such as Boeing are embarking on an era in which core competitive engineering knowledge and design rationale is being encoded in software systems. The promise of this technology is that this knowledge can be leveraged across many different designs, product families, and even different uses (e.g., generative process planning for manufacturing). The current state of the practice attempts to achieve this goal through the reuse of software components. A fundamental problem with this approach to knowledge sharing and reuse is that what we are trying to reuse is software—the end artifact in a long and complicated process that goes from requirement specifications, through a process of design, to implementations. Knowledge sharing and reuse can not easily and uniformly occur at the software level. So what can be done as an alternative? This paper describes a theory, methodology, language, and tool for the semi-automatic development and maintenance of engineering software from requirement specifications. In essence, this paradigm for software development and maintenance is one that explicitly captures requirement specifications, designs, implementations, and the refinement processes that lead from requirements all the way down to software. By recording this entire refinement history, we stand a better chance of leveraging knowledge for different uses.