Search

This paper, first, analyzes the problem about alternative models in parallel process; Secondly, present a neural network architecture and learning algorithm; Finally, discuss parallel software design method based on multi-layered neural network model and object-oriented techniques. An object-oriented approach to the initial analysis and design of a solution model provided the benefits of data abstraction and encapsulation which aided in the identification and modeling of the parallelism inherent in the problem domain.

Our objective was to examine factors that affect the accuracy of energy expenditure measurements (EE), when using flow-through indirect calorimetry (IC), to determine the minimum length of time needed to measure 24- and 48-h EE and to compare cross-over and parallel designs as methods of investigation during energy balance collections (EB) in preterm infants. A baby doll manikin was used to determine equilibration times and to compare VCO(2) and VO(2) as measured by flow meter and indirect calorimetry under different study conditions, one of which simulated an EB. "Continuous" EE was measured to determine the minimum length of time needed to accurately reflect 24- and 48-h EE and to compare parallel and cross-over studies as methods of study design in a group of "normal" enterally fed preterm infants. The mean (+/-SD) errors between flow meter and indirect calorimetry determinations for VCO(2) and VO(2) were -1.9 +/- 2.5 and -1.8 +/- 4.3% under conditions that simulated an EB. Cumulative 6-h EE accurately predicted 24- and 48-h EE. Expressed in absolute terms (kcal/d), EE did not change on a day-to-day basis but did increase over the 2-wk study period. Expressed on body weight basis (kcal/kg/d), EE did not change on a day-to-day or week-to-week basis. The variance in EE due to biologic variability; i.e. the parallel design, was approximately 6 times greater than that due to age, weight, and weight gain; i.e. the cross-over design. Indirect calorimetry, therefore, accurately measures EE in conditions simulating an energy balance collection. Six-hour EE determinations are valid estimate of EE during a 48-h balance collection, while cross-over studies may be the preferred method of study design during short-term studies of EE in preterm infants.

The following topics are summarized for the 2nd quarter of 2000: (1) We amended the structural design of the buoy, moving the horizontal bulkhead from the bottom to the top of the buoy. (2) We designed the main bearing using Hilman Rollers. We discarded a parallel design using Lubron slide bearings due to concerns about keeping dirt out of the sliding surfaces. (3) We performed a preliminary failure tree risk analysis for the system as required by ABS. (4) We made various drawings of the system and sub-components.

Now-a-days, companies have to face individual customer demands and rapid technological changes. Beside the fulfilment of quality and cost requirements companies are forced to realise a shortened time-to-market to maintain and increase competitiveness. One approach to meet these requirements is Simultaneous Engineering, the integrated and time parallel design of products and processes.

Parallelisation of the watershed segmentation method is studied in this paper. Starting with a successful parallel watershed design solution, extensive tests on various parallel machines are presented to prove its portability and performance. Next, the watershed algorithm has been re-formulated as a modified connected component problem. Consequently, we present a scalable parallel implementation of the connected component problem, which is the key for the future improving of the parallel design for the watershed algorithm.

In spite of rather negative publicity on the crossover/self-controlled design for clinical trials in the early 80’s the Lancet and the New England Journal of Medicine published 41 such studies soon afterwards. Using these studies asexamples I try to give an overview of major advantages and disadvantages of crossover and parallel group studies. Strengths of the crossover versus parallel design include: (1) elimination of between-subject variability of symptoms, (2) no need for large samples, (3) fewer ethical problems, (4) the test-persons are able to express their preference for one of the compounds being given. Weaknesses include: (1) carryover effect from one treatment period into the other, (2) time effect due to spontaneously evolving symptoms in a lengthy trial. The crossover/self-controlled design although routinely used for all types of therapies in phase I/II studies cannot be used in phase III/IV studies otherwise than for symptomatic treatments of stable disease. Treatments of chronic diseases are directed primarily to the relief of persistent symptoms rather than the cure of a rapidly evolving symptomatology. These very aspects make them particularly suitable for crossover/self-controlled studies. All of the advantages are involved while the disadvantages are not relevant. Recent statistical reports support these findings. Clinical practitioners will find it useful to consider the suitability of the trial design for the condition under study when assessing the practical importance of published reports of clinical trials.

The paper presents the new computational model of Herbrand engines which combines finite-state control with uninterpreted data and function registers, thus yielding a finite representation of infinite-state machines. Herbrand engines are used to provide a high-level model of out-of-order execution in the design of micro-processors. The problem of verifying that a highly parallel design for out-of-order execution correctly implements the Instruction Set Architecture is reduced to establishing the equivalence of two Herbrand engines. We show that, for a reasonably restricted class of such engines, the equivalence problem is decidable, and present two algorithms for solving this problem.

This multi-center single-blind study compared clinical outcomes following guided tissue regeneration (GTR) treating human Class II furcation defects with a new polylactic-acid-based bioabsorbable barrier (test treatment) or a non-absorbable ePTFE barrier (control treatment). Clinical parameters evaluated were change in vertical attachment level (VAL), horizontal attachment level (HAL), probing depth (PD), and gingival margin location (REC). Surgical treatment resulted in clinically and statistically equivalent changes when comparisons were made between test and control treatments. VAL gain was 2.0 mm for test and 1.6 mm for control groups; HAL gain was 2.1 mm for both test and control groups. PD reduction was 2.3 mm for the test group and 2.1 mm for the control group. Test sites experienced an additional 0.3 mm of recession beyond baseline; control sites, 0.5 mm. Within-group comparisons showed that the amount of recession was not significantly different from baseline in the test group. Recession in the control group was significantly different from baseline. All other parameters in both the test and control groups were significantly different from baseline. Evaluation of safety data indicated no significant differences between test and control treatments, although there was a strong trend for the control group to have more postoperative abscess or suppuration than test sites (control = 11; test = 4; P = 0.06).

Aim: To investigate whether the use of an oxidizing mouth rinse as an adjunct to chlorhexidine is efficacious in reducing stains and plaque. Materials and Methods: This study had a single-blind, three-group (n = 35 each) parallel design, including a 21 days experimental period during which group I rinsed with chlorhexidine (CHX) 0.2% alone, group II used chlorhexidine (CHX) followed by hydrogen peroxide (H 2 O 2 ) 1.5%. Group III rinsed with the same mouthwashes in reverse order. Patients were randomly assigned to one of the three groups. The examination for plaque, and stains was done after 1, 2, and 3 weeks of rinsing. Results: Group II showed significantly less stain intensity in comparison with group I after 14 and 21 days (P values 0.025 and 0.005, respectively). The proportion of stained surfaces was less in the group II than in the group I and was significant at the end of 1 week. The plaque formation was significantly less in groups II and III than group I at 7, 14, and 21 days. Conclusion: The adjunctive use of hydrogen peroxide to chlorhexidine proved to be superior to chlorhexidine alone with regard to the inhibition of plaque and development of stains.

PERCH is the name of a software tool specifically designed to capture the requirements of hybrid systems. The aim is to provide an integrated environment from requirement capture through to implementation of real applications that are complex and hybrid in nature. The tool encourages a parallel view of the systems being modelled and acts as a high level interface to further parallel design and analysis environments.

This report provides calculational results from the updated Lagrangian structural finite-element programs SPECTROM-32 and SPECTROM-333 for the purpose of qualifying these codes to perform analyses of structural situations in the Waste Isolation Pilot Plant (WIPP). Results are presented for the Second WIPP Benchmark (Benchmark II) Problems and for a simplified heated room problem used in a parallel design calculation study. The Benchmark II problems consist of an isothermal room problem and a heated room problem. The stratigraphy involves 27 distinct geologic layers including ten clay seams of which four are modeled as frictionless sliding interfaces. The analyses of the Benchmark II problems consider a 10-year simulation period. The evaluation of nine structural codes used in the Benchmark II problems shows that inclusion of finite-strain effects is not as significant as observed for the simplified heated room problem, and a variety of finite-strain and small-strain formulations produced similar results. The simplified heated room problem provides stratigraphic complexity equivalent to the Benchmark II problems but neglects sliding along the clay seams. The simplified heated problem does, however, provide a calculational check case where the small strain-formulation produced room closures about 20 percent greater than those obtained using finite-strain formulations. A discussion is given of each of the solved problems, and the computational results are compared with available published results. In general, the results of the two SPECTROM large strain codes compare favorably with results from other codes used to solve the problems.

In parallel design, several interface designers work independently on a design problem, and produce a set of ideas that can be studied further. In a panel at INTERCHI’93 three expert designers presented their first designs of a system for selecting and playing songs from a remote CD database. We have worked on the same exercise with novice designers. We show that also inexperienced interface designers can produce designs with many features worth investigating further. By increasing the number of novice designers also the number of new design ideas increases, and (in this exercise) finally exceeds the number of ideas produced by expert designers.

In a randomized, parallel design, 19 patients with winter depression were treated with either a week of a white 1.5-hr dawn simulation peaking at 250 lux or a week of a red, 1.5-hr dawn signal peaking at 2 lux. The subjects were told that they would receive either a white or red dawn reaching in intensity that would be dimmer than standard bright light treatment. At the end of both the baseline week and the treatment week subjects were blindly assessed with the Hamilton Rating Scale for Depression (HDRS). Analysis of covariance was used to compare the two dawn treatments. The white, 1.5-hr, 250 lux dawn simulation resulted in significantly (p0.05) lower HDRS scores compared to the red, 1.5-hr, 2 lux dawn. This is the second controlled study which indicates that dawn simulation is an effective treatment for winter depression.

Thirty healthy, male subjects (age I8 to 75 yrs) were used in a two-group, double-blinded, parallel design to evaluate the safety and tolerance of a potential probiotic organism, Lactobacillus reuteri. Subjects ( 15/treatment) consumed two gelatin capsules daily for 21 d that contained either a freeze dried L. reuteri preparation in a cryoprotectant, or a placebo (cryoprotectant). The concentration of L. reuteri was such that subjects consumed 1 x 1011 CFU per day. The study was 28 d in length with daily documentation of the presence of gastrointestinal symptoms (nausea, diarrhoea, cramping, distention, flatulence, vomiting, constipation, burping and reflux). In addition, serum chemistries, haematology, urinalysis, urinary indican excretion, and faecal microbiota (L. reuteri and total Lactobacillus spp. enumeration) were measured weekly (day 0, 7, 14, 21 and 28). A physical exam was given on day 0. 21, and 28. An additional faecal sample was obtained on day 77 for microbial enumeration. Subjects could consume their regular diets; however, alcohol was not allowed. Physical exam and urinalysis parameters were not clinically direrent between treatments. Supplemental L. reuteri reduced (P0.05) on subsequent urine collections. Although significant differences were observed for a few of the serum chemistry and haematology variables, all of the d u e s remained within the expected normal range for healthy adult males. Subjects consuming supplemental L. reuteri ad increased (P

Three user interface designers were asked to design interfaces for a given problem. These designs were made available to a group of usability specialists for heuristic evaluation. The reviewers will lead off the panel with specific questions to the designers regarding the usability aspects of their designs. The panel will feature a lively discussion of the designers' various approaches and solutions.

Diagnostic criteria, random parallel placebo-controlled study design, and appropriate clinical assessment for both safety and efficacy are all among the essential requirements for the evaluation of a new antidepressant agent. Paroxetine and imipramine were compared for efficacy and safety in a large multicentre, randomized, placebo-controlled, double-blind, parallel design study in the USA. The study involved 717 outpatients with major depressive disorder; after a one-week washout period they were treated for 6 weeks, being assessed at weekly intervals. The results from all six participant centres combined showed that both active drugs were statistically superior to placebo by week 2 and that the antidepressant response was significant for both. Paroxetine was better tolerated than imipramine with fewer dropouts from side effects. This combined study clearly indicated that paroxetine was an effective and well-tolerated antidepressant.

Registration or alignment of medical images in clinical applications requires cost-effective high-performance computing. In this paper, we present a parallel design and implementation of a mutualinformation-based multiresolution nonrigid registration algorithm that takes advantage of the Cell Broadband Engine® (Cell/B.E.) Architecture by exploiting the different levels of parallelism and optimization strategies. The new method was tested with a dualprocessor Cell/B.E. processor-based system. The experiments show an average performance of 1.09 µs per voxel and excellent scalability, demonstrating real-time or near-real-time performance for the computationally demanding task of nonrigid image registration.

3594 Background: PTK787/ZK 222584 (PTK/ZK) is a novel, oral, anti-angiogenesis compound that blocks tyrosine kinase activity from all known vascular endothelial growth factor (VEGF) receptors. Prel...

To the Editor: I was intrigued by the expedited publication of Harati et al., who reported that tramadol was safe and effective in the treatment of painful diabetic neuropathy.1 This was a randomized controlled parallel design study that used an inert placebo as the control drug. The side effect profile of tramadol relative to placebo was striking and raises the issue of the efficacy of blinding in this study. In the tramadol group, …

Context.—Although standards of reporting randomized controlled trials are well established internationally, essential study elements continue to be omitted, which hampers interpretation and systematic review of randomized controlled trials.Objective.—To identify deficiencies in the quality of reporting of placebo-controlled randomized trials published in German or English.Design.—Observational study comparing 32 German- and 89 English-language reports of placebo-controlled trials with parallel design, published by the same group of authors between 1985 and 1994.Main Outcome Measure.—High reporting quality, defined as adherence to published standards and measured by an 18-item scale based on 2 standard guidelines.Results.—The mean quality score was 8.4 (SD, 3.0; range, 1-16) of 18. The difference of the mean quality scores between English-language reports compared with German-language reports was small (0.27; 95% confidence interval, −0.97 to 1.52). More articles reported clinical aspects than trial methods or statistics.Conclusion.—There is room for improvement in the reporting of placebo-controlled randomized trials for both English and German reports.

SynopsisThis study examines the influence of 40 mg of propranolol on agoraphobics throughout 5-hour periods of exposure in vivo on 3 alternate days. Twenty-three patients were studied using a double-blind parallel design and 19 followed up for 3 months. The propranolol group spent significantly less time than the placebo group travelling alone in the month after treatment, and had improved significantly less on a measure of general symptoms at 3 months. The adverse influence of propranolol on treatment outcome appeared mainly due to a waning effect in the last hour of exposure. Attempts to measure coping with panics as an independent variable were largely unsuccessful.

The bioavailability of two altretamine preparations was studied in a randomized cross-over design. The two preparations were compared with a third in a parallel design. Dissolution differences between the preparations were observed, which could give rise to differences in bioavailability caused by the extensive first-pass effect of altretamine. The in vivo data showed a trend to differences in bioavailability.

The rapid progress in computer hardware technology that has been made in the past will be matched by a continuing increase in maximum computing speed. However, much of the future speed increase will be gained not through basic hardware improvements, but through entirely new types of machine organization, such as the parallel design, as is used in the ILLIAC IV, the pipeline design, as would be used in the proposed STAR, and other more speculative designs. These designs place rather severe restrictions upon the formulation of problems that can take advantage of the designs. The theoretical structure of the limitations will be developed from the viewpoint of the user, and the results of a comparative programming of several problems for the ILLIAC IV, the STAR, and other more conventional machines will be discussed.

SYNOPSIS In a double-blind study comparing crossover and parallel group designs, 55 patients with recurrent throbbing headache were given either belladonna-ergotamine-phenobarbital (BEP) or placebo for headache prophylaxis. Analysis of the crossover design showed BEP to be significantly more effective than placebo in reduction of headache severity and overall effectiveness. During BEP treatment, patients had less days during which medications for relief of headache symptoms were required. Although there were period effects favoring BEP, no significant differences between treatments were observed in the parallel design. The greater variability in the parallel portion of the study precluded demonstration of statistical significance with this number of patients. The crossover design indicates that BEP offers benefits in the prophylaxis of recurrent throbbing headache, without serious or frequent side effects.

Background: Split thickness skin graft (STSG) is one of the modalities used to close a defect. The donor site can be healed secondarily with tulle grass and moist gauze after 14 days and with transparent dressing will take about 10-13 days. Recently the wound healing by using honey application has been used. Whether or not using honey application with transparent dressing will hasten the epithelialization rate of the STSG donor site is now in question. Method: This research is an open, non-randomized clinical trial with a parallel design and intervention using honey application with transparent dressing. A total of 19 patients (7 female and 8 male) were included in this study. The reapplication of honey was done every two days. The patients was followed up every day, with clinical photograph taken and complaints such as pain, odor and infection noted. Results: The mean epithelialization rate of the donor site treated with honey and transparent dressing was 9,74 (+0,24) days compared to 10,79 (+1,23) days in the transparent-dressing-only group (p=0,00). Conclusion: The application of honey with transparent dressing to cover for STSG donor site led to a faster epithelialization rate, less odor and less pain. Commercial honey was used, and readily available.

In this study, we were interested in how approaches to teaching high school physics in New Zealand influenced students' perceptions of physics and their consequent desire to continue with physics. We also investigated the reasons participants became physics teachers to inform how more teachers might be attracted into the profession. The convergent parallel design of this study used mixed methods including a national survey as well as classroom observations and interviews with teachers and students. The study has identified how a focus on content knowledge and more "traditional" teaching approaches tends to discourage students to progress with physics.

This paper presents the OpenMP parallel implementation of the Distinct Lattice Spring Model (DLSM). The motivation of this study is to reduce the computing time and increase the computational capacity of the DLSM code. Firstly, the basic theory of DLSM is introduced. As an explicit method, the DLSM is very suitable for parallelization implementation. Only a few code changes are needed for the OpenMP implementation. Then; the detail parallel design of the OpenMP implementation is presented. Finally, numerical examples are evaluated on quad-core PCs to test the speedup of the parallel DLSM. It is found that a maximum speedup of more than four times is achieved. This means the implementation is successful.

This article presents information on how to improve aeration efficiency without effecting additional capital and operating costs. Two different configurations of aerator modules were studied using water test. The aerator modules consisted of a number of venturi air injectors connected either in series or in parallel aimed at achieving better aeration efficiencies. The results obtained indicated that the parallel design generally performed better than the series design in terms of transferring oxygen into water, with the oxygen transfer coefficients found to be 9.67 and 5.93 h-1 for two- and three-aerator parallel modules, as opposed to 4.54, 3.79, and 3.63 h-1 for one-, two-, and three-aerator series modules. Similarly, higher aeration efficiencies were also observed for the parallel design of the two modules (0.14 and 0.10 kgO2/kWh) as compared to those in the series design (0.07, 0.06, and 0.06 kgO2/kWh). The parallel module with two air injectors achieved the highest aeration efficiency (0.14 kgO2/kWh), which was two times the highest efficiency reached by the series modules (0.07 kgO2/kWh). In the same way, the oxygenation capacities were also significantly higher for parallel modules (0.22 and 0.15 kgO2/h) than for series modules (0.10, 0.09, and 0.09 kgO2/h). The data suggest that more research effort should be invested in exploring the possibility of using inexpensive aerators through better designs of the air injector complex in order to develop inexpensive aeration systems.

MASGORET, M.S., BOTHA, C.J., MYBURGH, J.G., NAUDE, T.W., PROZESKY, L., NAIDOO, V., VAN WYK, J.H., POOL, E.J. & SWAN, G.E. 2009. Molasses as a possible cause of an “endocrine disruptive syndrome” in calves. Onderstepoort Journal of Veterinary Research, 76:209–225 During the mid 1990s a potentially serious, chronic syndrome was reported in well-managed beef and dairy herds from unrelated parts of South Africa. Farmers reported that it manifested as various combinations of decreased production, decreased weaning masses, apparent immune breakdown in previously immunocompetent animals, increased reproductive disorders, various mineral imbalances in non-deficient areas and goitre, noticeable as enlarged thyroid glands. The farmers associated this syndrome with certain batches of sugar cane molasses and molasses-based products. The syndrome was reminiscent of an “endocrine disruptive syndrome”. The objective of this study was to evaluate the suspected endocrine disruptive effect of molasses included in cattle feed. Using existing in vitro assays, four batches of molasses syrup were screened for possible inclusion in a calf feeding trial. Two batches were selected for the trial. Thirty-two, 4- to 6-week-old, weaned Holstein bull calves were included in the single phase, three treatment, parallel design experiment. In two of the groups of calves, two different batches of molasses were included in their rations respectively. The control group was fed a ration to which no molasses was added, but which was balanced for energy and mineral content. The mass gain of the calves was recorded over the 6-month study period. The calves were clinically examined every week and clinical pathology parameters, immune responses and endocrine effects were regularly evaluated. Even though endocrine disrupting effects were detected with the in vitro screening assays, these could not be reproduced in the calves in the experiment. The two batches of molasses utilized in the calf feeding trial did not induce major differences in any of the parameters measured, with the exception of a lower mass gain in one of the molasses-fed groups (Group 1), which tended towards significance. The results of the study indicate that the two batches of molasses had no endocrine disruptive or immunosuppressive effects in calves.

The use of the two-period crossover trial for comparing two noncurative treatments in patients with a chronic disease is appealing since it eliminates the between-subject variability. However, the possibility of the existence of carryover effect leads many authors to advise that a parallel design be used whenever carryover is suspected. We examine this advice and quantify the degree of carryover required to make the parallel design preferable in terms of the power of the test of treatment effect and precision of the estimate of treatment difference. We conclude that in many situations this amount of carryover is substantial and unlikely to exist.

Parental responses to a Child Behavior Checklist were compared before and after changing antiepileptic therapy in an open, parallel design study in the School of Pharmacy and Departments of Neurology and Pediatrics, School of Medicine, University of North Carolina, Chapel Hill, North Carolina.