No description available.

Graduation date: 1995

Since its first use in human in 1929, the electroencephalogram (EEG) has become one of the most important diagnostic tool in clinical neurophysiology. However, their use in clinical studies is limited because the huge quantity of collected information is complicated to treat. Indeed, it is very difficult to have an overall picture of this multivariate problem. In addition to the impressive quantity of data to be treated, an intrinsic problem with electroencephalograms is that the signals are "contaminated" by body signals not directly related to cerebral activity. However, these signals do not interest us directly to evaluate treatment effect on the brain. Removing these signals known as "parasitic noise" from electroencephalograms is a difficult task. We use clinical data kindly made available by the pharmaceutical company Eli Lilly (Lilly Clinical Operations S.A., Louvain-la-Neuve, Belgium). Particular types of analyses were already carried out on these data, most based on frequency bands. They mainly confirmed the enormous potential of EEG in clinical studies without much insight in the understanding of treatment effect on the brain. The aim of this thesis is to propose and evaluate a panel of statistical techniques to clean and to analyze electroencephalograms. The first presented tool enables to align curves such as selected parts of EEGs before any further statistical treatment. Indeed, when monitoring some continuous process on similar units (like patients in a clinical study), one often notices a typical pattern common to all curves but with variation both in amplitude and dynamics across curves. In particular, typical peaks could be shifted from unit to unit. This complicates the statistical analysis of sample of curves. For example, the cross-sectional average usually does not reflect a typical curve pattern: due to shifts, the signal structure is smeared or might even disappear. Another of the presented tools is based on the preliminary linear decomposition of EEGs into statistically independent signals. This decomposition provides on the one hand an effective cleaning method and on the other hand a considerable reduction of the quantity of data to be analyzed. The technique of decomposition of our signals in statistically independent signals is a well-known technique in physics primarily used to unmix sound signals. This technique is named Independent Component Analysis or ICA. The last studied tool is functional ANOVA. The analysis of longitudinal curve data is a methodological and computational challenge for statisticians. Such data are often generated in biomedical studies. Most of the time, the statistical analysis focuses on simple summary measures, thereby discarding potentially important information. We propose to model these curves using non parametric regression techniques based on splines.

This paper develops a potential reduction algorithm for solving a linear-programming problem directly from a "warm start" initial point that is neither feasible nor optimal. The algorithm is of an "interior point" variety that seeks to reduce a single potential function which simultaneously coerces feasibility improvement (Phase I) and objective value improvement (Phase II). The key feature of the algorithm is the ability to specify beforehand the desired balance between infeasibility and nonoptimality in the following sense. Given a prespecified balancing parameter /3 > 0, the algorithm maintains the following Phase I - Phase II "/3-balancing constraint" throughout: (cTx- Z*) < /3TX, where cTx is the objective function, z* is the (unknown) optimal objective value of the linear program, and Tx measures the infeasibility of the current iterate x. This balancing constraint can be used to either emphasize rapid attainment of feasibility (set large) at the possible expense of good objective function values or to emphasize rapid attainment of good objective values (set /3 small) at the possible expense of a lower infeasibility gap. The algorithm exhibits the following advantageous features: (i) the iterate solutions monotonically decrease the infeasibility measure, (ii) the iterate solutions satisy the /3-balancing constraint, (iii) the iterate solutions achieve constant improvement in both Phase I and Phase II in O(n) iterations, (iv) there is always a possibility of finite termination of the Phase I problem, and (v) the algorithm is amenable to acceleration via linesearch of the potential function.

Electroosmotic flow is a convenient mechanism for transporting polar fluid in a microfluidic device. The flow is generated through the application of an external electric field that acts on the free charges that exists in a thin Debye layer at the channel walls. The charge on the wall is due to the chemistry of the solid-fluid interface, and it can vary along the channel, e.g. due to modification of the wall. This investigation focuses on the simulation of the electroosmotic flow (EOF) profile in a cylindrical microchannel with step change in zeta potential. The modified Navier-Stoke equation governing the velocity field and a non-linear two-dimensional Poisson-Boltzmann equation governing the electrical double-layer (EDL) field distribution are solved numerically using finite control-volume method. Continuities of flow rate and electric current are enforced resulting in a non-uniform electrical field and pressure gradient distribution along the channel. The resulting parabolic velocity distribution at the junction of the step change in zeta potential, which is more typical of a pressure-driven velocity flow profile, is obtained. / Singapore-MIT Alliance (SMA)

No description available.

Computer simulations play a profound and fundamental role inmodern theoretical physics, chemistry and materials science. Tounderstand the complex physics of metally liquids, metals,quasicrystals and metally glasses a working model imposing thelocal and global order is needed. Experiments and theory havepredicted the local order in liquid metals to beicosahedral. The current work has been done using molecular dynamicscomputer simulations of a monatomic system using a simplepair-potential for the interactions. Two new pair-potentialshas been developed, the Zetterling-1(Z1) and Zetterling-2(Z2)potentials. They are specifically modeled to impose icosahedralorder. The basis for the development of the potentials was theold Dzugutov potential which is known to freeze into adodecagonal quasicrystal. The new Zetterling potentials have alonger interaction range and a narrower first minimum. The morenarrow first minimum will enhance the local icosahedralordering and the longer interaction range was introduced toincorporate a second maximum in the potential mimicing theFriedel oscillations found in metallic systems. These Friedeloscillations are due to the singularity which arises at theFermi surface due to the screening of the positive charge bythe electron gas. Five papers are included in the study. The first two papersare studies of icosahedral clustering in the liquid andsupercooled liquid. The simulations in Paper I was done usingthe old Dzugutov potential while the new potentials were usedin Paper II using both molecular dynamics and the Basin Hoppingalgorithm presented in Chapter 5. Paper III considers theconcept of dynamical ergodicity in the context of thesuper-cooled liquid behaviour. The simulations were made usingthe old Dzugutov potential. Paper IVr eports a moleculardynamics simulation using the Dzugutov potential undersuper-cooling. A formation of icosahedrally structured domainswith distinctly slow diffusion which grows with cooling in alow-dimensional manner and percolate around Tc, the criticaltemperature of the mode-coupling theory. A sharp slowing downof the structural relaxation relative to diffusion is observed.It is concluded that this effect cannot be accounted for by thespatial variation in atomic mobility. The low-dimensionalclustering is discussed as a possible mechanism of fragility.Paper Vin vestigates the crystallization of a simple monatomicliquid model which utilizes the Zetterling-1 potential. Thesystem forms a thermodynamically stable solid phase exhibitingcubic symmetry. Its diffraction pattern is identified as thatof γ-brass, a tetrahedrally packed crystalline structurewith 52 atoms in the unit cell. <b>Keywords:</b>simple liquids, molecular dynamics, pairpotential, icosahedral cluster.

The demand for customised products that meet different markets and different customers is steadily increasing. Also, the demand for shorter lead times for the delivery of these customised products puts strains on design departments whose work tends to become increasingly repetitive. At the same time, designing variants takes time from innovative, original design, and/or problem-solving tasks. A powerful tool in the endeavour to cut lead times, workloads, and ultimately costs in order to become more competitive in an increasingly globalised market is Design Automation. Automating tedious and repetitive design tasks will free the designers to focus on the tasks that require skill, creativity, intuition, and cooperation to be solved. Consequently, seeing a need for design automation systems is not difficult. What becomes a lot more difficult is identifying the type, scope, and format of the system implementation, as well as the actual design tasks and activities to support or automate. Therefore, there is a need for structured and systematic approaches for the realisation and implementation of design automation systems. This research work is aimed at presenting such approaches, methods, and aids. It also addresses the importance of identifying the exact tasks to be automated. This has to be done in order to find the method and implementations best suited for solving the tasks, something that is especially important for companies whose human and financial resources might not allow them to invest in a system with functionality that vastly exceeds their actual needs. The contribution of this work is a structured method for planning for design automation implementation. First, the design process is discussed from an automation perspective. Following this is a presentation of a framework of design automation. This framework has the purpose of serving as a common base for consensual discussions about design automation. In addition, it supports the setting-up of system specifications. The framework is followed by the introduction of a set of identifiers of system needs and potentials, focusing on the existing processes that need to be broken down and identified in order to specify the tasks to be automated. Following this is a set of criteria of system characteristics, focusing on properties of the intended system implementation. Finally, some realisation and implementation issues are addressed and exemplified through a number of pilot system implementations. The presented method for planning design automation, together with the presented framework of design automation, provides implementers with issues to address regarding potential, need, scope, and format of system implementations. Further, it supports the weighing of desired system characteristics in order to find the right balance between system complexity and functionality.

Tidigare forskning har visat på att konkurrensen kring talangfulla medarbetare aldrig förr varithårdare och att ha strategier för att behålla dessa blir därför viktigt. En strategi som syftar till att görajust detta är talent management. Men hur arbetar man med detta och vilka föreställningar kringfenomenet finns? Vilka faktorer kan finnas som hindrar och/eller möjliggör arbetet med talentmanagement? För att ta reda på detta har det i studien valts att göra en fallstudie där data samlats ingenom semistrukturerade intervjuer. Det framkommer i studien att talent management upplevspositivt då det ger erkänsla och uppskattning. Dock är det viktigt att det finns en tydligkommunikation så att tanken kring vad satsningen ska leda till och att förväntningarna är tydliga.HR-avdelning, chefer och talanger har alla ett ansvar för att talent management ska bli en strategi föratt få talangfulla medarbetare att stanna kvar i företaget.

This bachelor thesis aims to estimate Sweden’s export potential towards South-Korea since initial data indicates that Sweden has from 1997 up until 2005 been exporting less to South-Korea when compared to, in general, OECD. Furthermore, South-Korea seems to be a low prioritized market for Swedish firms in the East-Asian region. As many before us, we have used a basic gravity model, including GDP and distance in kilometer has been used as explanatory variables for the observed trade value. The dummy variable land-locked, to estimate trade potential for 15 commodity groups. Sweden was set to be the exporting country, South-Korea the importing country together with all the other OECD members, which were used as points of reference. The outcome of the gravity regression shows that distance and the dummy variable landlocked (if a country does not have access to open water) have a very strong relationship to the observed export data. However, GDP was proven to have a very weak relationship to the observed export data thus making the estimation process of trade potential for all, except one, commodity group biased. The gravity model has been widely criticized for inflating export potential due to misspecification a problem that we experienced when running our regression. Thus, from this study no strong conclusions can be drawn concerning the trade potential from Sweden to South-Korea.