<p>The gut and brain are involved in a bi-directional communication system, referred to as the gut-brain axis. While it has been established that antimicrobials induce dysbiosis in the gut, which further disrupts immune and metabolic homeostasis, research on brain and behaviour development is becoming a topic of interest. We propose that alterations via antibiotics at the level of the gut microbiota impacts the gut-brain axis. The primary interest of this thesis is to understand the effects that antibiotics have on brain and behaviour development in conjunction with changes in the immune system and metabolism using the antibiotic mouse model. Mice treated with antibiotics revealed behavioural differences in the open field apparatus and three-chamber social behaviour apparatus, but not in the elevated plus maze and auditory fear conditionings enclosures. Evaluation of intestinal permeability revealed that female Balb/C mice administered a combination of bacitracin, neomycin and primaricin and another group administered a combination of ampicillin, neomycin and primaricin showed reduced intestinal permeability. Furthermore, the immune system condition was evaluated using flow cytometric analysis of spleens, which revealed no effect of treatment on immune cell profiles in CD1 mice treated with ampicillin. Evaluation of serum cytokine levels showed minimal differences in Balb/C and C57Bl/6 mice treated with antibiotics. Body weight and water and food consumption were evaluated in mice administered antibiotics. Weight loss differences were observed in two groups of female Balb/C mice, with the first group administered bacitracin, neomycin and primaricin and the second group administered ampicillin , neomycin and primaricin. Antibiotic treatment dependent differences in water and food consumption were observed. Serum insulin and leptin level investigation revealed that female Balb/C mice administered ampicillin, neomycin and primaricin had reduced serum insulin levels compared to strain matched controls. These findings indicate that antibiotic treatment impact metabolic function. This pilot study using antibiotic treated mouse models provides insight on the microbiota’s effects on the gut-brain axis, which can help to potentially identify methods of preventing gut microbiota mediated pathology in humans.</p> / Master of Science (MSc)

Traditional management-leadership and organizational literature depicts the individual as conflated with their role and instructs them to handle their employees and the situations that arise daily and over the course of business cycles instrumentally and for the purposes of control and productivity. This more traditional and mainstream literature does not adequately address, if at all, the unconscious factors influencing people or the management situations in which they find themselves. Using a model based upon the theory of the Swiss psychologist Carl Gustav Jung and the relationship of consciousness to the unconscious, this dissertation looks at situational management cases and reveals the existence of the unconscious in the midst of our strongest claims to rationality. Present and active, the cases show the unconscious to be a significant factor in creating subjective meaning and ordering our world even in the most "rational" moments of our lives in public administration. They further describe how it is that the individual in the manager-leader role is implicated in and caused by the very situations they are attempting to manage and the way in which acknowledging and relating to the unconscious provides an additional resource for public managers. / Ph. D.

This thesis provides insight into state-of-the-art Modular Multilevel Converters (MMC) for medium and high voltage applications. Modular Multilevel Converters have increased in interest in many industrial applications, as they offer the following advantages: modularity, scalability, reliability, distributed location of capacitors, etc. In this study, the modeling, control and design considerations of modular based multilevel converters, with an emphasis on the reliability of the converter, is carried out. Both modular multilevel converters with half-bridge and full-bridge sub-modules are evaluated in order to provide a complete analysis of the converter. From among the family of modular based hybrid multilevel converters, the newly released Alternate Arm Converter (AAC) is considered for further assessment in this study. Thus, the modular multilevel converter with half-bridge and full-bridge power cells and the Alternate Arm Converter as a commercialized hybrid structure of this family are the main areas of study in this thesis. Finally, the DC fault analysis as one of the main issues related to conventional VSC converters is assessed for Modular Multilevel Converters (MMC) and the DC fault ride-through capability and DC fault current blocking ability is illustrated in both the Modular Multilevel Converter with Full-Bridge (FB) power cells and in the Alternate Arm Converter (AAC). Accordingly, the DC fault control scheme employed in the converter and the operation of the converter under the fault control scheme are explained. The main contributions of this study are as follows: The new D-Q model for the MMC is proposed for use in the design of the inner and outer loop control. The extended control scheme from the modular multilevel converter is employed to control the Alternate Arm Converters. A practical reliability-oriented sub-module capacitor bank design is described based on different reliability modeling tools. A Zero Current Switching (ZCS) scheme of the Alternate Arm Converter is presented in order to reduce the switching losses of the Director Switches (DS) and, accordingly, to implement the ZCS, a design procedure for the Arm inductor in the AAC is proposed. The capacitor voltage waveform is extracted analytically in different load power factors and the waveforms are verified by simulation results. A reliability-oriented switching frequency analysis for the modular multilevel converters is carried out to evaluate the effect of the switching frequency on the MMC's operation. For the latter, a DC fault analysis for the MMC with Full-Bridge (FB) power cells and the AAC is performed and a DC fault control scheme is employed to provide the capacitor voltage control and DC fault current limit, and is illustrated herein. / Master of Science

In the field of video surveillance, vast amounts of data are gathered each day. To be able to identify what occurred during a recorded session, a human annotator has to go through the footage and annotate the different events. This is a tedious and expensive process that takes up a large amount of time. With the rise of machine learning and in particular deep learning, the field of both image and video captioning has seen large improvements. Contrastive Language-Image Pretraining is capable of efficiently learning a multimodal space, thus able to merge the understanding of text and images. This enables visual features to be extracted and processed into text describing the visual content. This thesis presents a system for extracting and ranking important events from surveillance videos as well as a way of automatically generating a description of the event. By utilizing the pre-trained models X-CLIP and GPT-2 to extract visual information from the videos and process it into text, a video captioning model was created that requires very little training. Additionally, the ranking system was implemented to extract important parts in video, utilizing anomaly detection as well as polynomial regression. Captions were evaluated using the metrics BLEU, METEOR, ROUGE and CIDEr, and the model receives scores comparable to other video captioning models. Additionally, captions were evaluated by experts in the field of video surveillance, who rated them on accuracy, reaching up to 62.9%, and semantic quality, reaching 99.2%. Furthermore the ranking system was also evaluated by the experts, where they agree with the ranking system 78% of the time. / Inom videoövervakning samlas stora mängder data in varje dag. För att kunna identifiera vad som händer i en inspelad övervakningsvideo så måste en människa gå igenom och annotera de olika händelserna. Detta är en långsam och dyr process som tar upp mycket tid. Under de senaste åren har det setts en enorm ökning av användandet av olika maskininlärningsmodeller. Djupinlärningsmodeller har fått stor framgång när det kommer till att generera korrekt och trovärdig text. De har också använts för att generera beskrivningar för både bilder och video. Contrastive Language-Image Pre-training har gjort det möjligt att träna en multimodal rymd som kombinerar förståelsen av text och bild. Detta gör det möjligt att extrahera visuell information och skapa textbeskrivningar. Denna master uppsatts beskriver ett system som kan extrahera och ranka viktiga händelser i en övervakningsvideo samt ett automatiskt sätt att generera beskrivningar till dessa. Genom att använda de förtränade modellerna X-CLIP och GPT-2 för att extrahera visuell information och textgenerering, har en videobeskrivningsmodell skapats som endast behöver en liten mängd träning. Dessutom har ett rankingsystem implementerats för att extrahera de viktiga delarna i en video genom att använda anomalidetektion och polynomregression. Video beskrivningarna utvärderades med måtten BLEU, METOER, ROUGE och CIDEr, där modellerna får resultat i klass med andra videobeskrivningsmodeller. Fortsättningsvis utvärderades beskrivningarna också av experter inom videoövervakningsområdet där de fick besvara hur bra beskrivningarna var i måtten: beskrivningsprecision som uppnådde 62.9% och semantisk kvalité som uppnådde 99.2%. Ranknignssystemet utvärderades också av experterna. Deras åsikter överensstämde till 78% med rankningssystemet.

Serotonin gilt als Hauptangriffsstelle gängiger Antidepressiva bei schweren Depressionen, wie bspw. selektive Serotonin-Wiederaufnahmehemmer (SSRI), und -Enhancer (SSRE). Es bleibt offen, ob SSRI / E ausschließlich über die Manipulation des Serotoninspiegels wirken, oder ob alternative Signalwege daran beteiligt sind. Ansatzpunkte hierfür sind beispielsweise die neurotrophen Signalwege (spez. Brain derived neurotophic factor, BDNF) oder die Hypothalamus-Hypophysen-Nebennieren- (HPA) – Signalwege des Stressachsensystems. Ebenfalls wurde in Nagetiermodellen beobachtet, dass mit der Dysregulation des zentralen Serotoninsystems bei schweren Depressionen, ein Rückgang der Neurogenese im Gyrus dentatus des Hippocampus einhergeht. Ziel dieser Arbeit war, das Zusammenspiel von Serotonin, BDNF, adulter Neurogenese und der Stressachse zu untersuchen. Zentrum der Studien ist ein Mausmodell, mit einer genetischen Depletion des zentralen Serotonin-synthetisierenden Enzyms Tryptophanhydroxylase 2 (sog. Tph2-/- Mäuse). Es wurden die physiologische Reaktionen auf die Behandlung mit gängigen Antidepressiva abhängig von der Abwesenheit von Serotonin untersucht, um mögliche alternative Signalwege aufzeigen zu können. Die bekannte Zunahme der Neurogenese nach SSRI/SSRE-Behandlung wurde in Wildtyptieren beobachtet, während die Therapie in Tph2-/- Mäusen keine direkte kausale Wirkung zeigte. Im Gegensatz dazu waren die BDNF-Spiegel in depressionsrelevanten Hirnregionen in Tph2-/- Mäusen nach SSRI, signifikant verringert. Auch zeigen die Studien eine neurobiologische Relevanz von Serotonin im ZNS, bei den antidepressiven Mechanismen einer Elektrokonvulsiven Krampftherapie. Ebenfalls deuten erhöhte Neurogeneseraten bei lebenslanger Abwesenheit von Serotonin im ZNS, Therapiemethoden-unabhängig, möglicherweise auf eine modulierte Stressreaktion hin. Untersuchungen der Parameter des HPA-Stressachsensystems, wiesen auf einen grundlegend veränderten Stresshormonspiegel in Tph2-/- Mäusen hin. / Serotonin, the "molecule of happiness" is an important target for antidepressants. The mainly prescribed drugs in major depression are selective serotonin re-uptake inhibitors (SSRI); but recently, SSR-enhancer (SSRE) have also attracted clinical attention. However, only a quarter of patients responds to treatment. It needs to be determined, whether SSRI/E act solely via manipulating serotonin levels or whether other pathways are involved, e.g. neurotrophic signaling (brain-derived neurotrophic factor, BDNF) or the hypothalamus-pituitary-adrenal (HPA)-axis. Furthermore, in major depression, dysregulation of central serotonin signaling is accompanied with a decline in hippocampal neurogenesis, as has been observed in rodent models. At the center of this thesis is a mouse model deficient in the central serotonin-synthesizing enzyme, tryptophan hydroxylase 2 (Tph2-/- mice). I have investigated physiological responses to antidepressant treatment in the absence of brain serotonin, and the possible role of alternative pathways. I observed the typical increase in neurogenesis upon SSRI treatment in WT mice, while it had no effect in Tph2-/- mice. In contrast, BDNF levels were significantly decreased in Tph2-/- mice after treatment with no effect in WT control mice. Furthermore, my results show a critical role of brain serotonin in the neurobiological effects of electroconvulsive seizure. Surprisingly, in animals lacking central serotonin, increased neurogenesis was observed independently of the treatment. The gathered data indicated an altered stress response; therefore, parameters of the HPA-axis have been studied, indicating a downregulated HPA system in Tph2-/-animals in baseline state, but showed no difference in treatment or feedback control. This thesis gives insight into the mechanisms of antidepressant action and reveals ideas for novel pathways involved in the process that could be used as targets in therapeutic approaches and further research in major depression.

The aim of this work was to design the construction of transport equipment under industrial robot, pallet holder or other equipment. This manipulation device can be moved in one axis on a rigid guide. The device is designed for use in a linear manufacturing system for handling workpieces and tools. It is capable of operating two rows of machines when it is placed in the middle. The main function of this device is to extend the workspace to the carried devices. This type of construction can be used in combination with an industrial robot for other technological tasks such as welding or light machining. The main parts of this device include statically mounted beds that can be stacked behind each other to form optionally long conveyor paths and a positioning plate that moves along the path formed by these beds. An industrial robot, pallet holder or other device is attached to this positioning plate. In addition to the design itself, the thesis also includes a thorough research of the issue, further possibilities of solution in the form of concepts, verification of the structure by means of calculations and simulations and implementation of the solution into a specific production cell.

Im Bereich der Werkzeugmaschinen können Identifikationsmethoden zur Realisierung einer ressourceneffizienten Produktion durch Überwachung prozessrelevanter Parameter beitragen. Aufgrund der hierarchischen Trennung moderner Steuerungssysteme bildet deren konzeptionelle Gestaltung und Implementierung am Beispiel einer nichtinvasiven Identifikationsroutine an einer Werkzeugmaschine den Kern der vorliegenden Arbeit. Das Ziel besteht darin, eine autonome Anwendung zur Integration unterschiedlicher Überwachungsmechanismen zu entwickeln und durch exemplarische Integration einer Parameteridentifikation zu validieren. Im Anschluss an die theoretischen Vorbetrachtungen werden deshalb zunächst verschiedene konzeptionelle Entwürfe zur Integration des Identifikationsverfahrens an einer Werkzeugmaschine diskutiert und anhand definierter Bewertungskriterien klassifiziert. Die Auswahl eines geeigneten Konzeptes unter Berücksichtigung eines Bewertungsindex sowie eine beispielhafte Implementierung bilden die Grundlage für den experimentellen Funktionsnachweis. / In the field of machine tools, identification methods can contribute to the realization of a resource-efficient production by monitoring of process-relevant parameters. Due to the hierarchical separation of modern control systems, their conceptual design and implementation using the example of a non-invasive identification routine on a machine tool forms the main aspect of the master thesis. The goal is to develop an autonomous application for different monitoring mechanisms and its validation through an exemplary parameter identification. Following the theoretical previews different conceptual designs are discussed and classified according to defined evaluation criteria. The selection of an appropriate concept under consideration of an evaluation index and its exemplary implementation are the basis for the experimental functional verification.

La cardiomyopathie ischémique et l’insuffisance cardiaque (IC) sont deux des principales causes de morbidité et de mortalité dans les pays industrialisés. L’IC représente la condition finale résultant de plusieurs pathologies affectant le myocarde. Au Canada, plus de 400 000 personnes souffrent d’IC. Malgré la grande variété de traitements disponibles pour prendre en charge ces patients à haut risque de mortalité, l’évolution et le pronostic clinique de cette population demeurent sombres. Les thérapies de régénération par transplantation cellulaire représentent de nouvelles approches pour traiter les patients souffrant d’IC. L’impact de cette approche cellulaire et les mécanismes qui sous-tendent l’application de ce nouveau mode de traitement demeurent obscurs. Les hypothèses proposées dans cette thèse sont les suivantes : 1) l’évolution à long terme des patients qui se présentent en IC grave est nettement défavorable malgré les techniques actuelles de revascularisation chirurgicale à cœur battant; 2) la thérapie cellulaire et, plus spécifiquement, l’injection intracoronaire précoce de milieu de culture cellulaire, permet d’améliorer la récupération fonctionnelle du ventricule gauche suite à un infarctus aigu du myocarde; et 3) la mobilisation de l’axe cœur-moelle osseuse constitue un mécanisme de réponse important lors de la survenue d’un événement ischémique chronique affectant le myocarde. / Congestive heart failure (CHF) remains a leading cause of mortality in the developed world. There are more than 400,000 diagnosed cases of this pathology in Canada. Despite the numerous treatment options available for patients presenting with left ventricular dysfunction, the evolution of this population is still dismal. Stem cell transplantation is a potential approach to repopulate the injured myocardium, to treat heart failure, and to restore cardiac function. However, the exact mechanisms underlying the beneficial effects of this approach remain to be elucidated. The hypotheses of this thesis are the following: 1) the long-term evolution of patients undergoing coronary artery bypass graft surgery is still poor, even when considering the use of new innovative surgical strategies such as off-pump coronary revascularization; 2) the intracoronary injection of concentrated biologically active factors secreted by stem cells can achieve early protection of the ischemic myocardium and preserve heart function; and 3) the bone marrow/heart interaction in a critical axis is involved in chronic myocardial repair following persistent ischemic injury.

Les contrevenants de la conduite avec capacités affaiblies (CCA) n’entrent pas tous dans les registres de la sécurité routière avec le même risque de récidive. Pour pallier cette hétérogénéité, cette thèse propose de modéliser les interrelations entre les traits de personnalité et les comportements à risque associés à la récidive et de détecter un sous-groupe de contrevenants au risque de récidive élevé à l’aide de l’axe hypothalamo-hypophyso-surrénalien (HHS). Plus particulièrement, les trois articles de cette thèse s’intéressent au cortisol, l’hormone du stress. Le premier article élabore un modèle théorique réconciliant les connaissances sur l’axe HHS issues du domaine de la CCA et de domaines connexes. Lors de précédentes études, le nombre de condamnations antérieures pour CCA a été associé négativement à la réactivité du cortisol à la suite d’une situation stressante. Chez les récidivistes, cette faible réactivité s’explique partiellement par la recherche d’expériences, une dimension de la recherche de sensations. Au-delà ce trait de personnalité désinhibiteur, une faible activité de l’axe HHS a été associée à d’autres traits (c.-à-d. impulsivité et tendances antisociales) et d’autres comportements à risque (c.-à-d. infractions routières, arrestations criminelles et consommation problématique de substances psychoactives). Ce modèle intégrant la réactivité du cortisol permet une conceptualisation approfondie des diverses caractéristiques des contrevenants de la CCA et explique hypothétiquement la répétition des comportements à risque. Les deux articles suivants se penchent sur l’intérêt empirique d’utiliser l’axe HHS pour déterminer un sous-groupe de contrevenants à risque élevé de récidive. Plus précisément, le deuxième article émet l’hypothèse que les récidivistes (n = 30) ayant une faible activité de leur cortisol (c.-à-d. médiane de la surface sous la courbe relative au niveau de base et relative à la réactivité) ont davantage de traits de personnalité désinhibiteurs et de comportements à risque que les récidivistes ayant une forte activité. L’hypothèse n’a pas été confirmée. Au contraire, les récidivistes présentant une faible réactivité commettent moins d’infractions routières et d’arrestations criminelles que ceux ayant une forte réactivité. Quant à lui, le troisième article investigue une hypothèse similaire auprès des contrevenants primaires (n = 139). Les contrevenants manifestant une faible réactivité du cortisol (c.-à-d. différence entre prélèvements post-stress et pré-stress) ont davantage d’impulsivité attentionnelle, de non-planification, d’arrestations criminelles et de cigarettes fumées par jour que les contrevenants ayant une forte réactivité. Lors d’analyses exploratoires, la présence d’une variété de traits de personnalité désinhibiteurs et de comportements à risque chez les contrevenants primaires présentant une faible réactivité lorsque comparé au groupe contrôle (n = 31) suggère encore une fois leur risque élevé de récidive. L’intérêt d’ajouter un mécanisme neurobiologique pour modéliser les interrelations entre les traits de personnalité désinhibiteurs et les comportements à risque des contrevenants de la CCA a été exploré dans cette thèse. La détermination d’un sous-groupe de contrevenants présentant un risque élevé de récidive, à l’aide de l’axe HHS, semble davantage profitable auprès de l’hétérogène population des contrevenants primaires. En contrepartie, l’axe HHS ne permet pas de déterminer un sous-groupe ayant une problématique sévère auprès des récidivistes à l’extrême du continuum du risque. / Among driving while impaired (DWI) offenders, the risk of recidivism varies greatly. In order to overcome the heterogeneity among this population, the present thesis proposed a renewed conceptualization of DWI. More specifically, a model integrating disinhibitory personality traits and high-risk behaviours associated with DWI recidivism and the categorization of a high-risk subgroup of offenders based on hypothalamic-pituitary-adrenal (HPA) axis activity has been proposed. Three manuscripts aimed to answer these general objectives using salivary cortisol, that is, the stress hormone. In the first manuscript, the current HPA axis literature of DWI offenders and other high-risk populations have been merged into a comprehensive theoretical model. Previous studies have demonstrated an inverse correlation between DWI convictions frequency and cortisol reactivity to stress. Among recidivists, cortisol reactivity was partially explained by experience seeking, a sensation seeking dimension. Beyond this disinhibitory personality trait, low HPA axis activity has been linked to various traits (e.g., impulsivity and antisocial tendencies) and high-risk behaviours (e.g., traffic infractions, criminal arrests, and psychoactive drugs). By incorporating cortisol reactivity, this model stimulates a thorough conceptualization of several DWI offenders’ characteristics and as such, explains hypothetically repetition of high-risk behaviours. The following manuscripts are less theoretical and more empirical. The detection of a high-risk recidivism subgroup delineated with HPA axis activity has been investigated among two DWI offender populations. More specifically, the second manuscript hypothesized that among DWI recidivists (n = 30), low cortisol responders have more characteristics linked to recidivism such as disinhibitory personality traits and high-risk behaviours than high cortisol responders (as defined by the area under the curve sensitive to total hormonal release and to response to stimulation). This hypothesis was not supported. On the contrary, low cortisol DWI recidivists have less traffic infractions and criminal arrests than high cortisol recidivists. Finally, the third manuscript investigated a similar hypothesis but this time, among first-time DWI offenders (n = 139). Results demonstrated that low cortisol offenders have more attentional impulsiveness, non-planning impulsiveness, criminal arrests and cigarettes consumed per day than high cortisol offenders (as defined by post-stress minus pre-test episode). An exploratory analysis showed a more prevalent variety of disinhibitory personality trait and high-risk behaviours among low cortisol offenders compared to a non-DWI comparator group (n = 31). These results reinforce the potentially higher recidivism risk of this first-time DWI offenders subgroup. Neurobiological mechanism usefulness in modelling disinhibitory personality trait and high-risk behaviours of DWI offenders has been investigated in the present thesis. Additionally, detection of a high-risk recidivism subgroup seems more relevant among heterogeneous first-time DWI offenders. Instead, HPA axis activity is not as useful in detecting a problematic subgroup among the most severe offenders, namely recidivists.

De nombreux travaux dans la littérature se sont concentrés sur la modélisation vibro-acoustique de coques cylindriques raidies immergées, du fait des nombreuses applications industrielles, en particulier dans le domaine aéronautique ou naval. Cependant, peu d'entre elles prennent en compte des structures internes non-axisymétriques telles que des supports moteurs, des planchers ou des carlingages, qui peuvent avoir une influence importante sur le comportement vibro-acoustique du système. C'est pourquoi une méthode de sous-structuration baptisée CTF est présentée dans cette thèse. Elle est développée dans le cas général de deux structures minces couplées le long d'une ligne. Un ensemble de fonctions orthonormées, baptisées fonctions de condensation, est défini afin d'approximer les forces et déplacements à la jonction entre les sous-systèmes. Des fonctions de transfert condensées sont définies pour chaque sous-système découplé. L'utilisation du principe de superposition, de l'équilibre des forces et de la continuité des déplacements permet de déduire le comportement des sous-systèmes couplés. La méthode est d'abord développée et validée dans le cas de plaques, puis ensuite appliquée au cas d'une coque cylindrique raidie immergée couplée à des structures internes non-axisymétriques. Le système est dans ce cas décomposé en 3 familles de sous-systèmes : la coque cylindrique immergée décrite par une méthode semi-analytique basée sur la résolution des équations de Flügge dans le domaine des nombres d’onde, les structures internes axisymétriques (raidisseurs, cloisons) décrites par éléments finis axisymétriques et les structures non-axisymétriques décrites pas des modèles éléments finis. La méthode CTF est appliquée à différents cas tests afin de montrer l'influence des structures internes non-axisymétriques sur le comportement vibro-acoustique d'une coque cylindrique pour différents types d'excitations pertinents dans le domaine naval : une force ponctuelle, une onde plane acoustique et un champ de pression aléatoire (tel qu'un champ acoustique diffus ou une couche limite turbulente). / Many works can be found in the literature concerning the vibroacoustic modelling of submerged stiffened cylindrical shells, because of high interest in the industrial domain, in particular for aeronautical or naval applications. However, only a few of them take into account non-axisymmetric internal frames, as for instance engine foundations or floor partitions, that can play a role on the vibroacoustic behavior of the system. That is why a substructuring approach called the Condensed Transfer Function (CTF) approach is proposed in the first part of this thesis. The aim is to take advantage of both analytical models and element-based models, in order to be able to deal with the geometrical complexity, and to calculate at higher frequencies than with element-based methods only. The substructuring method is developed in the general case of thin mechanical structures coupled along curves. A set of orthonormal functions called condensation functions, which depend on the curvilinear abscissa along the coupling line, is considered. This set is then used as a basis for approximating and decomposing the displacements and the applied forces at the line junctions. Thanks to the definition and calculation of condensed transfer functions for each uncoupled subsystem and by using the superposition principle for passive linear systems, the behavior of the coupled subsystems can be obtained. The method is first developed and validated for plates and convergence criteria are defined in relation with the size of the basis of condensation functions. The CTF method is then applied to the case of a submerged stiffened cylindrical shell with non-axisymmetric internal frames. The system is partitioned in 3 types of subsystems: the submerged shell, the axisymmetric frames (stiffeners, bulkheads) and the non-axisymmetric frames. The submerged shell is described by a semi-analytical method based on the Flügge equations in the spectral domain. The axisymmetric frames are described by axisymmetric Finite Element models and the non-axisymmetric frames by Finite Element models. The CTF method is applied to different test cases in order to highlight the influence of non-axisymmetric internal frames on the vibroacoustic behavior of a submerged stiffened cylindrical shell, for different excitations particularly relevant in the naval domain: a point force, an acoustic plane wave, and a random pressure field (such as a diffuse sound field or a turbulent boundary layer for instance).