One way of reducing the examination time in magnetic resonance imaging (MRI) is to reduce the amount of raw data acquired, by performing so-called undersampling. Conventionally, MRI data is acquired line-by-line on a Cartesian grid. In the field of Cardiovascular Magnetic Resonance (CMR), however, radial k-space sampling is seen as a promising emerging technique for rapid image acquisitions, mainly due to its robustness against motion disturbances occurring from the beating heart. Whereas Cartesian undersampling will result in image aliasing, radial undersampling will introduce streak artifacts. The objective of this work was to train the deep learning architecture, CycleGAN, to reduce streak artifacts in radially undersampled CMR images, and to evaluate the model performance. A benefit of using CycleGAN over other deep learning techniques for this application is that it can be trained on unpaired data. In this work, CycleGAN network was trained on 3060 radial and 2775 Cartesian unpaired CMR images acquired in human subjects to learn a mapping between the two image domains. The model was evaluated in comparison to images reconstructed using another emerging technique called GRASP. Whereas more investigation is warranted, the results are promising, suggesting that CycleGAN could be a viable method for effective streak-reduction in clinical applications.

Interessen werden meist aus einer Trait- oder State-Perspektive beforscht. Erst kürzlich wurden die beiden Perspektiven in einem theoretischen Modell integriert (Su et al., 2019), das postuliert, dass sich Interesse-Traits als intraindividuell variierende States manifestieren. Aufbauend auf dieses Modell untersucht die vorgelegte Dissertation Interessen anhand der wichtigsten Taxonomie für Interesse-Traits—Hollands (1997) Taxonomie beruflicher Interessen. Das Ziel dieser Dissertation ist zu untersuchen, ob und wie sich Berufsinteresse-Traits als States im Alltag manifestieren. Kernstück dieser Arbeit bilden zwei präregistrierte Experience Sampling Studien. Sie werden mit einem Literaturüberblick zu Interesse-Traits und -States eingeleitet sowie der Identifizierung von zwei Forschungszielen: Die Untersuchung (1) der Natur alltäglicher Manifestationen von Interesse-Traits und (2) der Korrelate von Interesse-States. Beide Studien verfolgten diese zwei Forschungsziele. Zentrale Ergebnisse waren, dass (1) Personen systematisch in ihren Interesse-States im Alltag variieren; (2) ein spezifisches Variabilitätsmuster in Interesse-States die Natur von Interesse-States von derjenigen von Persönlichkeit-States zu unterscheiden scheint; (3) Interesse-States systematisch mit spezifischen situations- und personenbezogenen Variablen assoziiert sind. Die Dissertation schließt mit einer Zusammenfassung, wie die beiden Studien zu den identifizierten Forschungszielen beitragen und einer Diskussion der allgemeinen Stärken, Limitationen und Anregungen für künftige Forschung. Theoretische Implikationen werden vorgestellt und in das integrative Modell (Su, et al., 2019) eingebettet. Insgesamt beleuchtet die vorgestellte Forschung die Natur und Korrelate momentaner Manifestationen von Berufsinteressen im täglichen Leben und kann künftige Forschung dazu anregen, stärker eine State-Perspektive auf Interessen zu berücksichtigen. / Research on interests is typically conducted with a trait or state perspective. Only recently, the two perspectives have been integrated within a theoretical model (Su et al., 2019), proposing that interest traits manifest as intraindividually varying states. Based on this model, this dissertation examines interests as conceptualized with the most prominent interest trait taxonomy—Holland’s (1997) taxonomy of vocational interests. With this dissertation, I sought to explore whether and how vocational interest traits manifest as states in daily life. Two preregistered experience sampling studies are this dissertation’s core. They are introduced with a literature overview on interest traits and states and an identification of two research goals: Exploring (1) the nature of interest trait manifestations in daily life and (2) correlates of interest states. While the first study provided first empirical answers on the nature and correlates of vocational interest states in daily life, the second study sought to refine these insights. The key findings of the two studies are that (1) persons systematically vary in their vocational interest states in daily life; (2) a distinctive pattern of interest state variability seems to set the nature of vocational interest states apart from the nature of Big Five personality states; (3) vocational interest states are systematically associated with specific situation- and person-related variables. The dissertation concludes with a summary of how the two studies contribute to the identified research goals and a discussion of the studies’ general strengths, limitations, and suggestions for further research. Theoretical implications are presented and associated with the integrative model of interests (Su, et al., 2019). Altogether, shedding light on the nature and correlates of vocational interest states in daily life, the presented research may encourage future work to consider a state perspective in research on vocational interests.

Die Dissertation hat eine umfassende Betrachtung der aktuellen urheberrechtlichen Probleme des Tonträgersamplings zum Ziel. Sie teilt sich in einen zivil- und einen strafrechtlichen Teil. Der zivilrechtliche Teil der Untersuchung zeigt, inwieweit durch unautorisiertes Tonträgersampling die Vervielfältigungsrechte des Urhebers, des Interpreten sowie des Tonträgerherstellers verletzt werden können. Im Bereich der Rechte des Urhebers liegen die Schwerpunkte der Bearbeitung in der Frage der Schutzfähigkeit von Werkteilen, in der Anwendung der Regelung zur freien Benutzung gemäß § 24 Abs. 1 UrhG sowie in der Erörterung der für das Sampling relevanten urheberrechtlichen Schranken, insbesondere des Zitatrechts. Im Bereich der Rechte des Tonträgerherstellers sowie des ausübenden Künstlers werden schwerpunktmäßig die Schutzfähigkeit von Teilen der jeweiligen Leistung behandelt sowie die Frage einer möglichen analogen Anwendung der Regelung zur freien Benutzung gemäß § 24 Abs. 2 UrhG erörtert. Hierbei wird ein Überblick über die in der Lehre und der Rechtsprechung vertretenen Ansichten gegeben und ausführlich auf das zum Fall „Metall auf Metall“ im Jahr 2016 ergangene Urteil des Bundesverfassungsgerichts eingegangen. Der strafrechtliche Teil der Arbeit fragt danach, inwieweit unautorisiertes Sampling auch strafrechtliche Folgen haben kann. Schwerpunkte des strafrechtlichen Teils bilden die Erörterung der für das Sampling relevanten Straftatbestände des UrhG sowie die Behandlung der für das Sampling relevanten Irrtumskonstellationen nach der Irrtumsdogmatik des Strafgesetzbuches. / The purpose of this dissertation is to conduct a comprehensive investigation into current copyright issues associated with the sampling of audio recordings. The study is divided into two sections, one dealing with civil law and the other with criminal law. The section dedicated to civil law seeks to demonstrate to what extent unauthorised sound recording sampling represents a violation of the reproduction rights of the creator, the practising musical artist and the sound recording manufacturer. In terms of the rights of the creator, the main focus of the processing lies in the question of the protectability of work pieces used, the application of regulations determining free use pursuant to Section 24 paragraph 1 of the German Copyright Law, and the consideration of copyright limitations relevant to the sampling process, in particular quotation rights. In terms of the rights of the sound recording manufacturer as well as those of the practising artist, principal considerations are the protectability of parts of the respective performance and the question of a possible analogous application of regulations determining free use pursuant to Section 24 paragraph 2 of the German Copyright Law. In this regard, an overview of the views held in academia and jurisprudence will be provided as well as an extensive analysis of the judgment reached by the Bundesverfassungsgericht (German Federal Constitutional Court) in the 2016 case of “Metall auf Metall” (“Metal on Metal”). The section on criminal law addresses the question as to what extent sampling may result in penal consequences. The focus of the section on criminal law comprises the discussion of the respective copyright offences relevant to sampling, as well as the handling of the respective error framework relevant to sampling and pursuant to the doctrine of error in the German Criminal Code.

Doctor of Philosophy / Department of Biochemistry and Molecular Biophysics / Jianhan Chen / Intrinsically disordered proteins (IDPs) are functional proteins that lack stable tertiary structures under physiological conditions. IDPs are key components of regulatory networks that dictate various aspects of cellular decision-making, and are over-represented in major disease pathways. For example, about 30% of eukaryotic proteins contain intrinsic disordered regions, and over 70% of cancer-associated proteins have been identified as IDPs. The highly heterogeneous nature of IDPs has presented significant challenge for experimental characterization using NMR, X-ray crystallography, or FRET. These challenges represent a unique opportunity for molecular mod- eling to make critical contributions. In this study, computer simulations at multiple scales were utilized to characterize the structural properties of unbound IDPs as well as to obtain a mechanistic understanding of IDP interactions. These studies of IDPs also reveal significant limitations in the current simulation methodology. In particular, successful simulations of biomolecules not only require accurate molecular models, but also depend on the ability to sufficiently sample the com- plex conformational space. By designing a realistic yet computationally tractable coarse-grained protein model, we demonstrated that the popular temperature replica exchange enhanced sampling is ineffective in driving faster reversible folding transitions for proteins. The second original contribution of this dissertation is the development of novel simulation methods for enhanced sampling of protein conformations, specifically, replica exchange with guided-annealing (RE-GA) method and multiscale enhanced sampling (MSES) method. We expect these methods to be highly useful in generating converged conformational ensembles.

Call number: LD2668 .T4 1985 S246 / Master of Science

Kriging is proposed as a tool for Wi-Fi signal strength estimation for complex indoor environments. This proposal is based on two studies suggesting that kriging might be suitable for this application. Both of these studies have shortcomings in supporting this proposal, but their results encourage a more in depth investigation into this. Even though kriging is a geostatistical method developed for geographical interpolation, it has been used successfully in a wide range of other applications as well. This further suggests that kriging might be a versatile method to overcome some of the difficul- ties of existing signal strength estimation methods. Two main types of signal strength estimation are deterministic methods and empirical methods. Deterministic methods are generally very complex and requires input parameters that are difficult to obtain. Empirical methods are known to have low accuracy which makes them unreliable for practical use. Three main investigations are presented in order to evaluate the use of kriging for this application. A sampling plan is proposed as part of a generic application protocol for the practical use of kriging for Wi-Fi signal strength. It is concluded that kriging can be conffidently used as an estimation technique for Wi-Fi signal strength in complex indoor environments. Kriging is recommended for practical applications, especially where in- sufficient information is available about a building or where time consuming site surveys are not feasible. / MIng (Computer and Electronic Engineering), North-West University, Potchefstroom Campus, 2015

Kriging is proposed as a tool for Wi-Fi signal strength estimation for complex indoor environments. This proposal is based on two studies suggesting that kriging might be suitable for this application. Both of these studies have shortcomings in supporting this proposal, but their results encourage a more in depth investigation into this. Even though kriging is a geostatistical method developed for geographical interpolation, it has been used successfully in a wide range of other applications as well. This further suggests that kriging might be a versatile method to overcome some of the difficul- ties of existing signal strength estimation methods. Two main types of signal strength estimation are deterministic methods and empirical methods. Deterministic methods are generally very complex and requires input parameters that are difficult to obtain. Empirical methods are known to have low accuracy which makes them unreliable for practical use. Three main investigations are presented in order to evaluate the use of kriging for this application. A sampling plan is proposed as part of a generic application protocol for the practical use of kriging for Wi-Fi signal strength. It is concluded that kriging can be conffidently used as an estimation technique for Wi-Fi signal strength in complex indoor environments. Kriging is recommended for practical applications, especially where in- sufficient information is available about a building or where time consuming site surveys are not feasible. / MIng (Computer and Electronic Engineering), North-West University, Potchefstroom Campus, 2015

Master of Science / Food Science Institute / Daniel Y.C. Fung / There are many different environmental sampling methods that are currently used in the industry. They include swab, sponge, flocked swab, direct agar contact, and M-Vac. Several studies have been conducted to determine the benefits and drawbacks of each method. Sampling methods utilized in this study were the swab, flocked swab, and M-Vac. Three surfaces were utilized in this study: ultra high density polypropylene, 304 stainless steel with a 2B finish, and 304 stainless steel with a 2B finish and a buffed surface. Surfaces sampled were 100 cm2. Prior to inoculation, surfaces were autoclaved for 15 min at 121 °C for sterilization. Surfaces were inoculated by either Listeria monocytogenes or Escherichia coli O157:H7 at a concentration of 9 log10 CFU/ml by painting the inoculum onto the surface with a sterilized paintbrush. Brushes were dipped in inoculum for 2 sec before painting from left to right once and then from up to down once. Brushes were redipped for 2 sec and the painting step was repeated. The same brush was used for all E. coli O157:H7 samples and a different brush was used for all L. monocytogenes samples. Then, the surfaces were allowed to dry for 30 min before sampling took place. Listeria monocytogenes samples were appropriately diluted and plated in duplicate onto Tryptic Soy Agar (TSA) and Modified Oxford Media (MOX). Escherichia coli O157:H7 samples were properly diluted and plated in duplicate onto TSA and MacConkey Sorbital Agar (MSA). After plating, dry surfaces were stained using LIVE/DEAD® BacLight™ Bacterial Viability Kit. The Zeiss LSM 5 Pascal confocal laser scanning electron microscope was used for microscopy images and photographs. Six 1 mm by 1 mm random and representative images were taken of each surface. Viable cell count results show that the sponge sampling method, in general, recovered a higher number of microorganisms. The swab was normally shown to recover the least number of microorganisms. When examining the microscopy images it can be concluded that biofilms are more easily formed with L. monocytogenes than E. coli O157:H7. Imaging also allowed for a visual representation of the remaining organisms that made it appear as if there was actually more bacteria recovery when the M-Vac sampling method was employed than when the sponge method was utilized.

Published Article / This paper analyses work sampling and time study as work measurement methods with the view of employing them in an automated labour performance measurement system. These are compared with respect to Hawthorn effect, labour intensiveness, cost, tediousness and knowledge extensiveness. The analysis proves that work sampling is a better option for developing an automated labour performance measurement system that employs computer vision. Web cameras are used to feed real-time images to a central computer via USB extenders. The computer runs a standalone C++ application that uses a random function to establish when measurements are to be taken. The developed video camera footage is converted into a pixel matrix using OpenCV. This matrix is then filtered and analysed, enabling the tracking of a worker. The data generated is stored in text files. After the work sampling period has elapsed, the data is transferred into Microsoft Excel for analysis. Finally a report of the labour utilisation is generated in Microsoft Excel and then send to the analyst for review.

Pricing different financial derivatives is an essential part of the financial industry. For some derivatives there exists a closed form solution, however the pricing of high-dimensional American-style derivatives is still today a challenging problem. This project focuses on the derivative called option and especially pricing of American-style basket options, i.e. options with both an early exercise feature and multiple underlying assets. In high-dimensional problems, which is definitely the case for American-style options, Monte Carlo methods is advantageous. Therefore, in this thesis, regression-based Monte Carlo has been used to determine early exercise strategies for the option. The well known Least Squares Monte Carlo (LSM) algorithm of Longstaff and Schwartz (2001) has been implemented and compared to Robust Regression Monte Carlo (RRM) by C.Jonen (2011). The difference between these methods is that robust regression is used instead of least square regression to calculate continuation values of American style options. Since robust regression is more stable against outliers the result using this approach is claimed by C.Jonen to give better estimations of the option price. It was hard to compare the techniques without the duality approach of Andersen and Broadie (2004) therefore this method was added. The numerical tests then indicate that the exercise strategy determined using RRM produces a higher lower bound and a tighter upper bound compared to LSM. The difference between upper and lower bound could be up to 4 times smaller using RRM. Importance sampling and Quasi Monte Carlo have also been used to reduce the variance in the estimation of the option price and to speed up the convergence rate. / Prissättning av olika finansiella derivat är en viktig del av den finansiella sektorn. För vissa derivat existerar en sluten lösning, men prissättningen av derivat med hög dimensionalitet och av amerikansk stil är fortfarande ett utmanande problem. Detta projekt fokuserar på derivatet som kallas option och särskilt prissättningen av amerikanska korg optioner, dvs optioner som både kan avslutas i förtid och som bygger på flera underliggande tillgångar. För problem med hög dimensionalitet, vilket definitivt är fallet för optioner av amerikansk stil, är Monte Carlo metoder fördelaktiga. I detta examensarbete har därför regressions baserad Monte Carlo använts för att bestämma avslutningsstrategier för optionen. Den välkända minsta kvadrat Monte Carlo (LSM) algoritmen av Longstaff och Schwartz (2001) har implementerats och jämförts med Robust Regression Monte Carlo (RRM) av C.Jonen (2011). Skillnaden mellan metoderna är att robust regression används istället för minsta kvadratmetoden för att beräkna fortsättningsvärden för optioner av amerikansk stil. Eftersom robust regression är mer stabil mot avvikande värden påstår C.Jonen att denna metod ger bättre skattingar av optionspriset. Det var svårt att jämföra teknikerna utan tillvägagångssättet med dualitet av Andersen och Broadie (2004) därför lades denna metod till. De numeriska testerna indikerar då att avslutningsstrategin som bestämts med RRM producerar en högre undre gräns och en snävare övre gräns jämfört med LSM. Skillnaden mellan övre och undre gränsen kunde vara upp till 4 gånger mindre med RRM. Importance sampling och Quasi Monte Carlo har också använts för att reducera variansen i skattningen av optionspriset och för att påskynda konvergenshastigheten.