The overwhelming trend of the usage of multimedia services has raised the consumers' awareness about quality. Both service providers and consumers are interested in the delivered level of perceptual quality. The perceptual quality of an original video signal can get degraded due to compression and due to its transmission over a lossy network. Video quality assessment (VQA) has to be performed in order to gauge the level of video quality. Generally, it can be performed by following subjective methods, where a panel of humans judges the quality of video, or by using objective methods, where a computational model yields an estimate of the quality. Objective methods and specifically No-Reference (NR) or Reduced-Reference (RR) methods are preferable because they are practical for implementation in real-time scenarios. This doctoral thesis begins with a review of existing approaches proposed in the area of NR image and video quality assessment. In the review, recently proposed methods of visual quality assessment are classified into three categories. This is followed by the chapters related to the description of studies on the development of NR and RR methods as well as on conducting subjective experiments of VQA. In the case of NR methods, the required features are extracted from the coded bitstream of a video, and in the case of RR methods additional pixel-based information is used. Specifically, NR methods are developed with the help of suitable techniques of regression using artificial neural networks and least-squares support vector machines. Subsequently, in a later study, linear regression techniques are used to elaborate the interpretability of NR and RR models with respect to the selection of perceptually significant features. The presented studies on subjective experiments are performed using laboratory based and crowdsourcing platforms. In the laboratory based experiments, the focus has been on using standardized methods in order to generate datasets that can be used to validate objective methods of VQA. The subjective experiments performed through crowdsourcing relate to the investigation of non-standard methods in order to determine perceptual preference of various adaptation scenarios in the context of adaptive streaming of high-definition videos. Lastly, the use of adaptive gain equalizer in the modulation frequency domain for speech enhancement has been examined. To this end, two methods of demodulating speech signals namely spectral center of gravity carrier estimation and convex optimization have been studied.

The multi-reference correlation consistent composite approach (MR-ccCA) was designed to reproduce the accuracy of more computationally intensive ab initio quantum mechanical methods like MR-ACPF-DK/aug-cc-pCV?Z-DK, albeit at a significantly reduced cost. In this dissertation, the development and applications of the MR-ccCA method and a variant of its single reference equivalent (the relativistic pseudopotential ccCA method) are reported. MR-ccCA is shown to predict the energetic properties of reactive intermediates, excited states species and transition states to within chemical accuracy (i.e. ±1.0 kcal mol 1) of reliable experimental values. The accuracy and versatility of MR-ccCA are also demonstrated in the prediction of the thermochemical and spectroscopic properties (such as atomization energies, enthalpies of formation and adiabatic transition energies of spin-forbidden excited states) of a series of silicon-containing compounds. The thermodynamic and kinetic feasibilities of the oxidative addition of an archetypal arylglycerol ?-aryl ether (?-O-4 linkage) substructure of lignin to Ni, Cu, Pd and Pt transition metal atoms using the efficient relativistic pseudopotential correlation consistent composite approach within an ONIOM framework (rp-ccCA-ONIOM), a multi-level multi-layer QM/QM method formulated to enhance the quantitative predictions of the chemical properties of heavy element-containing systems larger than hitherto attainable, are also reported.

En biologie clinique, disposer de mesures fiables et comparables, indépendamment du temps et du laboratoire, est primordial afin de permettre une prise en charge et un suivi adaptés des patients. Le moyen privilégié est d’établir la traçabilité métrologique des résultats de mesure aux unités du système international (SI), notamment par le biais de méthodes de référence et de matériaux de référence certifiés d’ordre supérieur. Ces travaux de thèse se sont attachés à caractériser deux méthodes de référence primaires candidates. Une méthode d’énumération absolue des lipoprotéines par ES-DMA pour l’évaluation du risque cardiovasculaire a été développée et caractérisée. Les résultats obtenus ont mis en évidence certaines limitations instrumentales rendant l’établissement de la traçabilité au SI difficile. Malgré les avantages certains de cette méthode pour l’analyse avancée de lipoprotéines, la fidélité et la robustesse des mesures ne sont pas suffisantes pour la considérer comme méthode de référence primaire. Une méthode de quantification de l’hémoglobine glyquée HbA1c, un biomarqueur du diabète sucré, par LC-MS/MS associée à la dilution isotopique a été mise en place et validée. La traçabilité au SI des résultats de dosage a été établie, toutefois, les incertitudes de mesures demeurent élevées et la comparabilité avec la méthode de référence IFCC reste à évaluer sur un nombre plus significatif d’échantillons et de laboratoires. Pour conclure, ces travaux ont mis en évidence les défis associés à la mise en place de nouvelles chaînes de traçabilité ainsi que les problématiques liées au développement de méthodes de référence primaires pour le dosage de biomarqueurs complexes. / Ensuring reliability and between-laboratory comparability of in-vitro diagnostic test results is essential for appropriate and internationally harmonized decision making and patient follow-up. To this end, a privileged way is to establish results metrological traceability to the international system of units (SI) through the development of primary reference methods and higher order reference materials. This thesis aimed at characterizing two candidate primary reference procedures.A method for lipoprotein absolute quantification by ES-DMA for cardiovascular diseases risk assessment was developed and characterized. Results evidenced some advantages of this method compared to other advanced lipoprotein testing assays, but some instrumental weaknesses make the establishment of results traceability to the SI difficult in the current state-of-the-art. Precision and robustness are additionally not sufficient for this method to be considered as a primary reference method.A method for the quantification of glycated hemoglobin HbA1c, a biomarker of diabetes mellitus, by LC/MS/MS associated to isotopic dilution was implemented and validated. Results traceability to the SI units could successfully be established. However, measurement uncertainties remain large and comparability with the IFCC reference method is still to be further assessed on a significant number of samples and laboratories.To conclude, this study highlighted the challenges associated with the implementation of new traceability chains and the difficulties related to the development and characterization of primary reference procedures for the quantification of complex biological markers.

En biologie clinique, il est indispensable de disposer de mesures fiables et comparables dans le temps et entre les laboratoires afin de permettre un dépistage et un suivi appropriés des patients. Pour cela, il est indispensable d’établir leur traçabilité métrologique aux unités du système international notamment par des méthodes de référence primaires ou des matériaux de référence certifiés (MRC) d’ordre supérieur. Ces travaux de thèse ont consisté à développer et valider des méthodes de référence pour le cholestérol total, les triglycérides, le HDL-cholestérol et le LDL-cholestérol. Leur valeur ajoutée par rapport à une valeur consensuelle a été démontrée lors d’évaluations externes de la qualité. Elles ont également permis de certifier un MRC qui pourra être utilisé pour le contrôle qualité et/ou l’étalonnage des méthodes de routine. Nous avons montré que le MRC était commutable pour la plupart des méthodes de routine pour les différents biomarqueurs, ce qui a permis de l’utiliser pour évaluer leur justesse. Les méthodes de routine avaient généralement tendance à sous-estimer la concentration en triglycérides (en particulier aux valeurs basses) et à surestimer nettement la concentration de cholestérol total et de LDL-cholestérol (en particulier aux concentrations proches du seuil de décision clinique), ce qui se traduit par une augmentation du nombre de faux-positifs (patients traités à tort). Une approche de correction de non commutabilité a également été proposée afin de permettre l’utilisation de matériaux non commutables pour évaluer la justesse. Pour conclure, ces travaux ont démontré l’importance de disposer de méthodes de référence ainsi que de MRC commutables / Reliable measurements in medical biology are essential for early screening and appropriate follow-up of patients. Ensuring metrological traceability of clinical measurements to higher order reference methods or certified reference materials enables to obtain comparable results over time and between different laboratories that could use different methods to quantify the same biomarker.In this study, reference methods were developed and validated for lipid profile biomarkers (total cholesterol, triglycerides, HDL-C, and LDL-C). Their value added in proficiency testing schemes was demonstrated against consensus mean. They were also used to characterize a certified reference material (CRM) that may be used both as quality control and/or calibrator of field methods. The CRM was shown to be commutable for most field methods and lipid profile biomarkers, which proved it was suitable to assess trueness. Results of our multicenter study showed that field methods tend to underestimate triglycerides (particularly at low concentrations) and overestimate total cholesterol and LDL-C (especially around the clinical threshold), resulting in false positives and significant patient misclassifications. An approach of non-commutanility correction was also presented to allow trueness assessment with non-commutable samples. In conclusion, this work highlights the importance of using reference methods and also commutable CRM to rigorously assess accuracy of field methods used in clinical laboratories

Anhand von theoretischen Untersuchungen wird eine umfassende Beschreibung von kleinen Metalloxidclustern gegeben. Bei den untersuchten Systemen handelt es sich um Aluminium- und Eisenoxid-Ionen sowie entsprechende Oxid-Cluster, die beide Metalle enthalten. Neben der Bestimmung der geometrischen Struktur der Cluster werden auch die allgemeinen elektronischen Eigenschaften der eisenhaltigen Verbindungen untersucht. Alle Vorhersagen werden durch Vergleich mit verfügbaren experimentellen Ergebnissen -- hauptsächlich aus der Infrarot-Photodissoziations- und der Photoelektronenspektroskopie -- überprüft und bewertet. Soweit möglich werden die Bewegungen von Atomen oder kleinen Gruppen innerhalb der Cluster einzelnen experimentellen Signalen zugeordnet. Besondere Aufmerksamkeit wird dem Eisendioxidmolekül und seinem Anion gewidmet. Es wird mit spezialisierten Wellenfunktionsmethoden untersucht, mit denen ab initio Franck-Condon-Simulationen einschließlich nicht-adiabatischer und Spin-Orbit-Kopplungen für die Photoionisation des Anions erstellt werden. Sie liefern Erklärungen für die komplizierte Schwingungsstruktur des experimentellen hochauflösenden Photoelektronenspektrums. / By means of theoretical investigations, a comprehensive description of small metal oxide clusters is given. The studied systems are aluminum and iron oxide ions as well as respective bi-metallic oxide clusters. Besides the determination of the geometrical structure of the clusters, the general electronic properties of the iron-containing compounds are investigated. All predictions are checked and assessed by comparison with available experimental results, mainly infrared photodissociation and photoelectron spectroscopy measurements. As far as possible, motions of atoms or small groups within the clusters are assigned to distinct experimental vibrational features. Particular attention is paid to the iron dioxide molecule and its anion. It is studied with sophisticated wave function methods based on which ab initio Franck-Condon simulations for the photodetachment from the anion, including non-adiabatic and spin-orbit couplings, are generated. They provide explanations for the complicated vibrational structure of the experimental high-resolution photoelectron spectrum.