Trung Hieu Nguyen

RWTH Aachen University, Aachen, North Rhine-Westphalia, Germany

Are you Trung Hieu Nguyen?

Claim your profile

Publications (9)24.42 Total impact

  • Source
    Trung Hieu Nguyen
    [Show abstract] [Hide abstract]
    ABSTRACT: In this study, some of azomethine (Schiff base) [1] were investigated by quantum chemistry method at the level of DFT/B3LYP with the 6-31+G (d, p) base sets. Quantum chemical parameters such as: HOMO, LUMO energies, orbital density distributions, energy gap, the charge distributions, absolute electronegativity values (χ), electron affinity (A), ionization potential (I), global hardness (η), softness (), the fraction of electrons transferred from inhibitors to iron (ΔN) have been calculated. The corrosion inhibition efficiency (%IE) published in [1] was then regressed on the result parameters individually using one variable function. Then the primary parameters that play the most important role in %IE were determined.
    Full-text · Conference Paper · May 2015
  • Trung Hieu Nguyen · Eng Siong Chng · Haizhou Li
    [Show abstract] [Hide abstract]
    ABSTRACT: Speaker diarization is the task of determining “Who spoke when?”, where the objective is to annotate a continuous audio recording with appropriate speaker labels corresponding to the time regions where they spoke. The labels are not necessarily the actual speaker identities, i.e. speaker identification, as long as the same labels are assigned to the regions uttered by the same speakers. These regions may overlap as multiple speakers could talk simultaneously. Speaker diarization is thus essentially the combination of two different processes: segmentation, in which the speaker turns are detected, and unsupervised clustering, in which segments of the same speakers are grouped. The clustering process is considered as unsupervised problem since there is no prior information about the number of speakers, their identities or acoustic conditions (Meignier et al., Comput Speech Lang 20(2–3):303–330, 2006; Zhou and Hansen, IEEE Trans Speech Audio Process 13(4):467–474, 2005). This chapter presents the fundamentals of speaker diarization and the most significant works over the recent years on this topic.
    No preview · Chapter · Jan 2015
  • Source
    Trung Hieu Nguyen
    [Show abstract] [Hide abstract]
    ABSTRACT: Aujourd'hui, la reconfiguration des réseaux électriques en général ou des réseaux de distribution en particulier après le défaut est réalisée par des différentes approches de calcul hors ligne. Cependant, ces approches doivent reposer beaucoup sur l'expérience des exploitants et l'immutabilité de la topologie du réseau. Une telle approche peut facilement tomber lorsque certaines contraintes d'exploitation ne peuvent pas être respectées. Pour résoudre ces problèmes, la nécessité de développer un outil efficace et réactif est posée comme la problématique de la thèse. Actuellement, plusieurs méthodes ont été développées. Ces méthodes sont tout à fait fascinant, mais ils ne travaillent pas assez efficacement en raison de leur complexité de calcul et de l'incapacité de produire des résultats optimaux. En utilisant une approche raisonnable, le travail présenté cette thèse vise à créer un outil qui aide les exploitants à reconfigurer dynamiquement et automatiquement le réseau de distribution après un défaut. Ici, nous présenterons l'approche retenue pour la reconfiguration des systèmes à événements discrets (DESs), et expliquer son adaptation, les progrès de mise en œuvre dans le domaine des réseaux électriques.
    Preview · Article · Mar 2010
  • Source
    Arnold Reusken · Trung Hieu Nguyen
    [Show abstract] [Hide abstract]
    ABSTRACT: We consider a parabolic interface problem which models the transport of a dissolved species in two-phase incompressible flow problems. Due to the so-called Henry interface condition the solution is discontinuous across the interface. We use an extended finite element space combined with a method due to Nitsche for the spatial discretization of this problem and derive optimal discretization error bounds for this method. For the time discretization a standard θ-scheme is applied. Results of numerical experiments are given that illustrate the convergence properties of this discretization.
    Full-text · Article · Oct 2009 · Journal of Fourier Analysis and Applications
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Coronary in-stent restenosis cannot be directly assessed by magnetic resonance angiography (MRA) because of the local signal void of currently used stainless steel stents. The aim of this study was to investigate the potential of a new, dedicated, coronary MR imaging (MRI) stent for artifact-free, coronary MRA and in-stent lumen and vessel wall visualization. Fifteen prototype stents were deployed in coronary arteries of 15 healthy swine and investigated with a double-oblique, navigator-gated, free-breathing, T2-prepared, 3D cartesian gradient-echo sequence; a T2-prepared, 3D spiral gradient-echo sequence; and a T2-prepared, 3D steady-state, free-precession coronary MRA sequence. Furthermore, black-blood vessel wall imaging by a dual-inversion-recovery, turbo spin-echo sequence was performed. Artifacts of the stented vessel segment and signal intensities of the coronary vessel lumen inside and outside the stent were assessed. With all investigated sequences, the vessel lumen and wall could be visualized without artifacts, including the stented vessel segment. No signal intensity alterations inside the stent when compared with the vessel lumen outside the stent were found. The new, coronary MRI stent allows for completely artifact-free coronary MRA and vessel wall imaging.
    Preview · Article · Apr 2005 · Circulation
  • [Show abstract] [Hide abstract]
    ABSTRACT: The objective of this study was to investigate the potential for artifact-free coronary magnetic resonance angiography (cMRA) in the presence of dedicated metallic MR stents in vitro and in a swine model. All investigations were performed at 1.5 T, applying a standard cMRA gradient echo sequence with a T2 preparation pulse. Two prototypes of each hand-woven, mechanically woven, and lasered Aachen Resonance Coronary MR Stents made out of an MR-compatible metallic alloy and dilated to 2.5 mm and 4 mm were examined in a water bath. Artifact behavior was judged independently by 2 radiologists as showing "no artifacts" for all tested stent types. Signal-to-noise ratios inside and outside of the stents were measured yielding a Pearson correlation coefficient of 0.98 (y = 1.22 + 0.92x). Nineteen stents (8 hand woven, 3 mechanically woven, 8 lasered) were deployed in coronary arteries of 19 domestic pigs and were examined by cMRA. Artifact behavior of the stents was analyzed by measuring the signal-to-noise ratio at the stent positions and compared with signal-to-noise ratio measurements outside of the stents, yielding a Pearson correlation coefficient of 0.90 (y = -0.75 + 1.06x). All 3 prototypes of coronary MR stents allowed complete visualization of the stent lumen and consequently determination of stent patency by cMRA.
    No preview · Article · Jun 2004 · Investigative Radiology
  • [Show abstract] [Hide abstract]
    ABSTRACT: Recent developments of magnetic resonance imaging enabled free-breathing coronary MRA (cMRA) using steady-state-free-precession (SSFP) for endogenous contrast. The purpose of this study was a systematic comparison of SSFP cMRA with standard T2-prepared gradient-echo and spiral cMRA. Navigator-gated free-breathing T2-prepared SSFP-, T2-prepared gradient-echo- and T2-prepared spiral cMRA was performed in 18 healthy swine (45-68 kg body-weight). Image quality was investigated subjectively and signal-to-noise ratio (SNR), contrast-to-noise ratio (CNR) and vessel sharpness were compared. SSFP cMRA allowed for high quality cMRA during free breathing with substantial improvements in SNR, CNR and vessel sharpness when compared with standard T2-prepared gradient-echo imaging. Spiral imaging demonstrated the highest SNR while image quality score and vessel definition was best for SSFP imaging. Navigator-gated free-breathing T2-prepared SSFP cMRA is a promising new imaging approach for high signal and high contrast imaging of the coronary arteries with improved vessel border definition.
    No preview · Article · Jun 2003 · Investigative Radiology
  • Source
    Trung Hieu Nguyen
    [Show abstract] [Hide abstract]
    ABSTRACT: In this thesis, we presented numerical methods for discretizing and solving the mass transport problem in two-phase flows. The level set method is used for capturing the time-dependent interface. The motion of the fluid is described by the two-phase Navier-Stokes equations. For the spatial discretization of these equations we use the known methods in the literature, namely the emph{improved} Laplace-Beltrami discretization for the surface force and the extended finite element (XFEM) for the pressure approximation. The combination of these methods delivers optimal error bounds when the surface tension coefficient is constant. For the general case with a variable surface tension coefficient, we introduce a new discretization of the localized surface force term. The solution of the mass transport equation must satisfy certain interface conditions, which imply that in general both the concentration and its derivatives are discontinuous across the interface. A simple transformation is often used in the literature to eliminate the discontinuity of the solution, which, however, results in a suboptimal approximation error bound O(h^{1/2}) in the L^2 norm for the finite element discretization. We use the Nitsche-XFEM method to handle the Henry condition and obtain an optimal error estimate O(h^2) in the L^2-norm for the spatial discretization in the case of a emph{stationary} interface. The semi-discretization resulting from the Nitsche-XFEM method is combined with the standard $theta$-scheme and an optimal time discretization error bound is also obtained. This method can also be applied for problem with moving interface but a full error analysis is not available. Finally, we performed numerical simulations of the coupled two-phase Navier-Stokes and mass transport equations for rising droplet problems for both cases of constant and concentration-dependent surface tension coefficients. For the latter case, different phenomena were observed, such as the occurrence of the so-called stagnant cap in the velocity field and a significant change in the droplet rising velocity. Due to the absence of a stabilization method for the discretization of the mass transport problem, we restrict ourselves to the case of medium diffusivity instead of the physically correct (much smaller) diffusivity. Effects of the initial concentration and the size of the convection (relative to the diffusion) on the droplet rising velocity and the droplet concentration at steady state are investigated. In der vorliegenden Doktorarbeit präsentieren wir numerische Methoden, um das Stofftransportproblems bei Zweiphasenströmungen zu diskretisieren und zu lösen. Zur Beschreibung der zeitabhängigen Grenzfläche wird die Levelsetmethode verwendet. Die Bewegung des Fluids wird mit Hilfe der zweiphasigen Navier-Stokes-Gleichungen beschrieben. Für die räumliche Diskretisierung benutzen wir in der Literatur bekannte Methoden aus verbesserter Laplace-Beltrami-Diskretisierung der Grenzflächenspannung und erweiterten finiten Elementen (XFEM) für den Druck. Diese Kombination liefert im Fall eines konstanten Grenzflächenspannungkoeffizienten optimale Fehlerschranken. Im allgemeinen Fall eines variablen Grenzflächenspannungkoeffizienten stellen wir eine neue Diskretisierung des lokalen Grenzflächenkraftterms vor. Die Lösung der Stofftransportgleichung genügt zwei Bedingungen, die dazu führen, dass im allgemeinen sowohl die Konzentration als auch ihre Ableitung an der Grenzfläche unstetig sind. In Methoden aus der Literatur wird häufig eine einfache Transformation verwendet, um die Unstetigkeit zu eliminieren. Diese Transformation führt bei der Finite-Elemente-Diskretisierung zu einer suboptimalen Approximationsfehlerschranke O(h^{1/2}) in der L^2-Norm. Wir setzen die Nitsche-XFEM-Methode ein, um die Henry-Bedingung zu erfüllen. Deshalb erhalten wir im Fall einer stationären Grenzfläche für die räumliche Diskretisierung eine optimale Fehlerschranke O(h^2) in der L^2-Norm. Die aus der Nitsche-XFEM-Methode resultierende Semidiskretisierung wird mit einem Standard-$theta$-Schema kombiniert. Wir zeigen eine optimale Fehlerschranke für die Zeitdiskretisierung. Die Methode kann auf Probleme mit beweglicher Grenzfläche angewendet werden, jedoch ist in diesem Fall keine vollständige Fehleranalyse verfügbar. Abschließend führen wir numerische Simulationen mit den gekoppelten zweiphasigen Navier-Stokes- und Stofftransportgleichungen durch. Ein aufsteigender Tropfen wird sowohl mit konstantem als auch mit konzentrationsabhängigem Grenzflächenspannungskoeffizienten simuliert. Im letzteren Fall treten im Vergleich zum ersteren neue Phänomene auf, zum Beispiel eine starre Kappe im Geschwindigkeitsfeld und eine signifikante Änderung der Aufstiegsgeschwindigkeit des Tropfens. Da keine Stabilisierung für die Diskretisierung des Stofftransportproblems implementiert wurde, beschränken wir uns auf den Fall mittlerer Diffusivität. Die physikalisch korrekte Diffusivität ist viel kleiner. In weiteren Simulationen wird der Effekt der Anfangskonzentration und der Konvektionsstärke (relativ zur Diffusion) auf die Aufstiegsgeschwindigkeit des Tropfens und auf die Stoffkonzentration im Tropfen im Gleichgewichtszustand untersucht.
    Preview · Article ·
  • Source
    Trung Hieu Nguyen
    [Show abstract] [Hide abstract]
    ABSTRACT: Derzeit befindet sich die Magnetresonanz-Koronarangiographie (MRKA) noch in der Entwicklungsphase. Um neue Erkenntnisse über das technische Potenzial dieser nicht invasiven diagnostischen Methode zu erlangen, wurde in dieser Arbeit ein Vergleich zwischen vier MRKA-Sequenzen vorgenommen. Als Vergleich diente jene Sequenz (Standard T2 präparierte 3D Gradienten Echo Sequenz), welche bisher als einzige in einer validierten Studie auf die Nutzung als MRKA-Sequenz untersucht worden ist. Ihr wurden drei verschiedene Sequenzen gegenübergestellt, welche in anderen Studien eine erfolgreiche technische Implementation erzielt hatten. Es zeigte sich, dass keine der vier Sequenzen bei den üblichen Vergleichskriterien als insgesamt überragende Sequenz hervorging. Die Ergebnisse erlaubten jedoch die Schlussfolgerung, dass zwei Sequenzen (T2 präpariert 3D steady-state free precession Sequenz mit kartesischer und radialer k-Raumabtastung) gegenüber der Vergleichssequenz bei differenzierter Interpretation der Vergleichskriterien überlegen waren. Insbesondere die radiale k-Raumabtastung könnte das derzeit größte Problem in der MRKA, die Bewegungsartefaktunterdrückung, sehr günstig beeinflussen. Daher sind zukünftige Studien mit der SSFP-Technik zur Verbesserung dieser viel versprechenden Sequenzen zu erwarten. At the moment coronary magnetic resonance angiography (corMRA) is still in research. To get more information about the potential of this non invasive method of diagnosis we compared four corMRA sequences. The sequence we used for reference sequence is the only one which has been investigated in a validated study so far. The three sequences we used for comparison have already been successfully technically implemented in former studies. With respect to the usual comparison criteria this study showed that none of the four sequences turned out as a generally outstanding sequence. However, differentiated interpretation of the comparison criteria allowed the conclusion that the reference sequence (Standard T2 Prepared 3D Gradient-Echo Sequence) could not reach the performance of two other sequences (two T2 Prepared 3D SSFP Sequences, one with cartesian and the other one with radial k-space sampling). In particular, radial k-space sampling could solve the currently largest problem in corMRA, the suppression of motion artifacts. Therefore we can surely expect new studies with improvement of these promising sequences.
    Preview · Article ·