Yes / Age and gender classification are two important problems that recently gained popularity in the research community, due to their wide range of applications. Research has shown that both age and gender information are encoded in the face shape and texture, hence the active appearance model (AAM), a statistical model that captures shape and texture variations, has been one of the most widely used feature extraction techniques for the aforementioned problems. However, AAM suffers from some drawbacks, especially when used for classification. This is primarily because principal component analysis (PCA), which is at the core of the model, works in an unsupervised manner, i.e., PCA dimensionality reduction does not take into account how the predictor variables relate to the response (class labels). Rather, it explores only the underlying structure of the predictor variables, thus, it is no surprise if PCA discards valuable parts of the data that represent discriminatory features. Toward this end, we propose a supervised appearance model (sAM) that improves on AAM by replacing PCA with partial least-squares regression. This feature extraction technique is then used for the problems of age and gender classification. Our experiments show that sAM has better predictive power than the conventional AAM.

Individuals are considered the weakest link when it comes to securing a personal computer system. All the technological solutions can be in place, but if individuals do not make appropriate security protection decisions they introduce holes that technological solutions cannot protect. This study investigates what personal characteristics influence differences in individual security behaviors, defined as behaviors to protect against security threats, by adapting Protection Motivation Theory into an information security context. This study developed and validated an instrument to measure individual security behaviors. It then tested the differences in these behaviors using the security research model, which built from Protection Motivation Theory, and consisted of perceived security vulnerability, perceived security threat, security self-efficacy, response efficacy, and protection cost. Participants, representing a sample population of home computer users with ages ranging from 20 to 83, provided 279 valid responses to surveys. The behaviors studied include using anti-virus software, utilizing access controls, backing up data, changing passwords frequently, securing access to personal computers, running software updates, securing wireless networks, using care when storing credit card information, educating others in one's house about security behaviors, using caution when following links in emails, running spyware software, updating a computer's operating system, using firewalls, and using pop-up blocking software. Testing the security research model found different characteristics had different impacts depending on the behavior studied. Implications for information security researchers and practitioners are provided, along with ideas for future research. / Ph. D.

A continuous improvement project (CIP) is a structured improvement project using a team of people "typically representing different departments or units in the organization" working to improve a process or work area over a relatively short period of time, such as a few days or up to several months. A CIP may use different improvement methodologies and tools, and may thus be defined according to the improvement approach. For instance, an organization adopting Lean as an improvement approach is likely to have CIPs implementing Lean tools, such as 5S or value stream mapping. These projects may be referred to as Lean projects in general, although they may also represent accelerated improvement projects such as Kaizen events, Kaizen blitz, or rapid improvement projects. Alternatively, an organization utilizing Six Sigma as an improvement approach may have Six Sigma projects that use the Define-Measure-Analyze-Improve-Control (DMAIC) process and statistical tools. Some organizations adopt an integrated improvement approach, such as Lean Six Sigma, and therefore may have CIPs with an even broader set of tools from which to choose. Lastly, many organizations may have an improvement approach not characterized by any single set of improvement processes and tools, and thus, may be thought of generally as process improvement, or quality improvement, projects using a traditional methodology as plan-do-study/check-act (PDSA or PDCA). In this dissertation, all of these types of improvement projects are referred as CIPs. Since the 1980s, hospitals have been using CIPs to address some of the problems in hospitals, such as quality in healthcare delivery, internal process efficiency, communication and coordination, and the cost of services. Some hospitals have achieved significant improvements, such as reducing the turnaround time for clinical laboratory results by 60 percent and reducing instrumentation decontaminations and sterilization cycle time by 70 percent. However, as with many other companies, hospitals often experience difficulty achieving their desired level of improvements with CIPs. Therefore, the purpose of this dissertation is to identify the critical success factors (CSFs) related to CIP success. In order to achieve this goal, five objectives were achieved: creating a methodology to assess the maturity or evolution of a research field (manuscript #1), identifying a comprehensive list of CSFs for CIPs (manuscript #2), assessing the maturity of the published literature on CIPs in hospitals (manuscript #3), identifying the most important factors related to CIPs in hospitals (manuscript #4) , and conducting an empirical investigation to define the CSFs for CIPs in hospital settings (manuscript #5 and #6). This investigation was conducted in three phases: research framing, variable reduction, and model development and testing. During these phases, the researcher used the following methodologies and data collection tools: systematic literature review, maturity framework (developed as part of this dissertation), expert study, retrospective survey questionnaire, exploratory factor analysis, partial-least squares structural equation modeling, and regression modeling. A maturity framework with nine dimensions was created (manuscript #1) and applied in order to identify a list of 53 factors related to CIP in general, involving any organization (manuscript #2). Additionally, the maturity framework was used to assess the literature available on CIPs in hospitals, considering only the authorship characteristic dimension (manuscript #3). Considering the frequency of new authors per year, the relative new integration of research groups, and the limited set of predominant authors, the research field, or area, of CIPs in hospitals is one with opportunities for improving maturity. Using the systematic literature review from manuscript #3, the list of 53 factors, and the list of predominant authors, a review of the literature was conducted, along with an expert study to more fully characterize the importance of various factors (manuscript #4). A conclusion from this particular work was that it is not possible to reduce the list of 53 factors based on these results, thus, a field study using the complete comprehensive list of factors was determined to have stronger practical implications. A field study was conducted to identify factors most related to CIP perceived success (manuscript #5) and CIP goal achievement (manuscript #6). The final results and practical implications of this dissertation consist in the identification of the following CSFs for CIP success in hospitals: Goal Characteristics, Organizational Processes, Improvement Processes, and Team Operation. These CSFs include several specific factors that, to the researcher's knowledge, have not been previously studied in empirical investigations: goal development process, organizational policies and procedures, CIP progress reporting, and CIP technical documentation. Practitioners involved with CIPs, such as CIP leaders, facilitators, stakeholders/customers, and continuous improvement managers/leaders, can utilize these results to increase the likelihood of success by considering these factors in planning and conducting CIPs. / Ph. D.

El fitoplancton es uno de los compuestos orgánicos de las aguas naturales y su diagnóstico es importante para evaluar el estado ecológico de los ecosistemas acuáticos, entre ellos las aguas costeras y de transición. El enriquecimiento de nutrientes antropogénicos y las alteraciones en la cadena de alimentación, incluyendo la reducción de consumidores de fitoplancton, produce un aumento espectacular de las existencias de fitoplancton. Esto ha causado cambios significativos en los ciclos de nutrientes de las áreas costeras, en la calidad del agua, en la biodiversidad y en el estado global del ecosistema. La caracterización de las comunidades fitoplanctónicas en ecosistemas acuáticos mediante el método de los recuentos microscópicos por epifluorescencia, es una tarea costosa en tiempo, material y recursos humanos altamente cualificados. El objetivo de este trabajo es, sin pretender sustituir a los recuentos con el microscopio sino complementarlos, poner a punto una técnica mediante espectrofotometría que disminuya estos costes, realizando medidas de espectros de absorción en el rango del visible en las muestras. Para llevar a cabo este trabajo se han tomado muestras en cinco zonas de la costa mediterránea de España. Estas zonas corresponden a ecosistemas acuáticos en los que influyen tanto las aguas continentales como las del mar Mediterráneo, es decir, zonas costeras influidas por aguas continentales (plumas continentales) y zonas continentales influidas por aguas marinas (estuarios). Las muestras tomadas presentan un gradiente de salinidad, en función de una mayor o menor influencia continental y también en función de la capa superficial de menor salinidad que yace sobre las aguas salinas más densas. En estas muestras con distintas salinidades también existen unas diferencias cualitativas y cuantitativas de la composición fitoplanctónica. / Martínez Guijarro, MR. (2010). DETERMINACIÓN DE COMUNIDADES FITOPLACTÓNICAS MEDIANTE ESPECTROSCOPÍA VISIBLE Y SU RELACIÓN CON LOS RECUENTOS POR MICROSCOPIA DE EPIFLUORESCENCIA [Tesis doctoral]. Universitat Politècnica de València. https://doi.org/10.4995/Thesis/10251/7106

No / Research in facial age estimation has been active for over a decade. This is due to its numerous applications. Recently, convolutional neural networks (CNNs) have been used in an attempt to solve this age old problem. For this purpose, researchers have proposed various CNN architectures. Unfortunately, most of the proposed techniques have been based on relatively ‘shallow’ networks. In this work, we leverage the capability of an off-the-shelf deep CNN model, namely the VGG-Face model, which has been trained on millions of face images. Interestingly, despite being a simple approach, features extracted from the VGG-Face model, when reduced and fed into linear regressors, outperform most of the state-of-the-art CNNs. e.g. on both FGNET-AD and Morph II benchmark databases. Furthermore, contrary to using the last fully connected (FC) layer of the trained model, we evaluate the activations from different layers of the architecture. In fact, our experiments show that generic features learnt from intermediate layer activations carry more ageing information than the FC layers.

Yes / The use of near infra red spectroscopy to predict the concentration of two pharmaceutical co-crystals; 1:1 ibuprofen – nicotinamide (IBU-NIC) and 1:1 carbamazepine – nicotinamide (CBZ-NIC) has been evaluated. A Partial Least Squares (PLS) regression model was developed for both co-crystal pairs using sets of standard samples to create calibration and validation data sets with which to build and validate the models. Parameters such as the root mean square error of calibration (RMSEC), root mean square error of prediction (RMSEP) and correlation coefficient were used to assess the accuracy and linearity of the models. Accurate PLS regression models were created for both co-crystal pairs which can be used to predict the co-crystal concentration in a powder mixture of the co-crystal and the active pharmaceutical ingredient (API). The IBU-NIC model had smaller errors than the CBZ-NIC model, possibly due to the complex CBZ-NIC spectra which could reflect the different arrangement of hydrogen bonding associated with the co-crystal compared to the IBU-NIC co-crystal. These results suggest that NIR spectroscopy can be used as a PAT tool during a variety of pharmaceutical co-crystal manufacturing methods and the presented data will facilitate future offline and in-line NIR studies involving pharmaceutical co-crystals.

L'imagerie cérébrale connaît un intérêt grandissant, en tant que phénotype intermédiaire, dans la compréhension du chemin complexe qui relie les gènes à un phénotype comportemental ou clinique. Dans ce contexte, un premier objectif est de proposer des méthodes capables d'identifier la part de variabilité génétique qui explique une certaine part de la variabilité observée en neuroimagerie. Les approches univariées classiques ignorent les effets conjoints qui peuvent exister entre plusieurs gènes ou les covariations potentielles entre régions cérébrales.Notre première contribution a été de chercher à améliorer la sensibilité de l'approche univariée en tirant avantage de la nature multivariée des données génétiques, au niveau local. En effet, nous adaptons l'inférence au niveau du cluster en neuroimagerie à des données de polymorphismes d'un seul nucléotide (SNP), en cherchant des clusters 1D de SNPs adjacents associés à un même phénotype d'imagerie. Ensuite, nous prolongeons cette idée et combinons les clusters de voxels avec les clusters de SNPs, en utilisant un test simple au niveau du "cluster 4D", qui détecte conjointement des régions cérébrale et génomique fortement associées. Nous obtenons des résultats préliminaires prometteurs, tant sur données simulées que sur données réelles.Notre deuxième contribution a été d'utiliser des méthodes multivariées exploratoires pour améliorer la puissance de détection des études d'imagerie génétique, en modélisant la nature multivariée potentielle des associations, à plus longue échelle, tant du point de vue de l'imagerie que de la génétique. La régression Partial Least Squares et l'analyse canonique ont été récemment proposées pour l'analyse de données génétiques et transcriptomiques. Nous proposons ici de transposer cette idée à l'analyse de données de génétique et d'imagerie. De plus, nous étudions différentes stratégies de régularisation et de réduction de dimension, combinées avec la PLS ou l'analyse canonique, afin de faire face au phénomène de sur-apprentissage dû aux très grandes dimensions des données. Nous proposons une étude comparative de ces différentes stratégies, sur des données simulées et des données réelles d'IRM fonctionnelle et de SNPs. Le filtrage univarié semble nécessaire. Cependant, c'est la combinaison du filtrage univarié et de la PLS régularisée L1 qui permet de détecter une association généralisable et significative sur les données réelles, ce qui suggère que la découverte d'associations en imagerie génétique nécessite une approche multivariée. / Brain imaging is increasingly recognised as an interesting intermediate phenotype to understand the complex path between genetics and behavioural or clinical phenotypes. In this context, a first goal is to propose methods to identify the part of genetic variability that explains some neuroimaging variability. Classical univariate approaches often ignore the potential joint effects that may exist between genes or the potential covariations between brain regions. Our first contribution is to improve the sensitivity of the univariate approach by taking advantage of the multivariate nature of the genetic data in a local way. Indeed, we adapt cluster-inference techniques from neuroimaging to Single Nucleotide Polymorphism (SNP) data, by looking for 1D clusters of adjacent SNPs associated with the same imaging phenotype. Then, we push further the concept of clusters and we combined voxel clusters and SNP clusters, by using a simple 4D cluster test that detects conjointly brain and genome regions with high associations. We obtain promising preliminary results on both simulated and real datasets .Our second contribution is to investigate exploratory multivariate methods to increase the detection power of imaging genetics studies, by accounting for the potential multivariate nature of the associations, at a longer range, on both the imaging and the genetics sides. Recently, Partial Least Squares (PLS) regression or Canonical Correlation Analysis (CCA) have been proposed to analyse genetic and transcriptomic data. Here, we propose to transpose this idea to the genetics vs. imaging context. Moreover, we investigate the use of different strategies of regularisation and dimension reduction techniques combined with PLS or CCA, to face the overfitting issues due to the very high dimensionality of the data. We propose a comparison study of the different strategies on both a simulated dataset and a real fMRI and SNP dataset. Univariate selection appears to be necessary to reduce the dimensionality. However, the generalisable and significant association uncovered on the real dataset by the two-step approach combining univariate filtering and L1-regularised PLS suggests that discovering meaningful imaging genetics associations calls for a multivariate approach.

Varje studie av en verklig process eller ett verkligt system är baserat på mätdata. Förr var den tillgängliga datamängden vid undersökningar ytterst begränsad, men med dagens teknik är mätdata betydligt mer lättillgängligt. Från att tidigare enbart haft få och ofta osammanhängande mätningar för någon enstaka variabel, till att ha många och så gott som kontinuerliga mätningar på ett större antal variabler. Detta förändrar möjligheterna att förstå och beskriva processer avsevärt. Multivariat analys används ofta när stora datamängder med många variabler utvärderas. I det här projektet har de multivariata analysmetoderna PCA (principalkomponentanalys) och PLS (partial least squares projection to latent structures) använts på data över avloppsvatten insamlat på Hammarby Sjöstadsverk. På reningsverken ställs idag allt hårdare krav från samhället för att de ska minska sin miljöpåverkan. Med bland annat bättre processkunskaper kan systemen övervakas och styras så att resursförbrukningen minskas utan att försämra reningsgraden. Vissa variabler är lätta att mäta direkt i vattnet medan andra kräver mer omfattande laboratorieanalyser. Några parametrar i den senare kategorin som är viktiga för reningsgraden är avloppsvattnets innehåll av fosfor och kväve, vilka bland annat kräver resurser i form av kemikalier till fosforfällning och energi till luftning av det biologiska reningssteget. Halterna av dessa ämnen i inkommande vatten varierar under dygnet och är svåra att övervaka. Syftet med den här studien var att undersöka om det är möjligt att utifrån lättmätbara variabler erhålla information om de mer svårmätbara variablerna i avloppsvattnet genom att utnyttja multivariata analysmetoder för att skapa modeller över variablerna. Modellerna kallas ofta för mjukvarusensorer (soft sensors) eftersom de inte utgörs av fysiska sensorer. Mätningar på avloppsvattnet i Linje 1 gjordes under tidsperioden 11 – 15 mars 2013 på flera ställen i processen. Därefter skapades flera multivariata modeller för att försöka förklara de svårmätbara variablerna. Resultatet visar att det går att erhålla information om variablerna med PLS-modeller som bygger på mer lättillgänglig data. De framtagna modellerna fungerade bäst för att förklara inkommande kväve, men för att verkligen säkerställa modellernas riktighet bör ytterligare validering ske. / Studies of real processes are based on measured data. In the past, the amount of available data was very limited. However, with modern technology, the information which is possible to obtain from measurements is more available, which considerably alters the possibility to understand and describe processes. Multivariate analysis is often used when large datasets which contains many variables are evaluated. In this thesis, the multivariate analysis methods PCA (principal component analysis) and PLS (partial least squares projection to latent structures) has been applied to wastewater data collected at Hammarby Sjöstadsverk WWTP (wastewater treatment plant). Wastewater treatment plants are required to monitor and control their systems in order to reduce their environmental impact. With improved knowledge of the processes involved, the impact can be significantly decreased without affecting the plant efficiency. Several variables are easy to measure directly in the water, while other require extensive laboratory analysis. Some of the parameters from the latter category are the contents of phosphorus and nitrogen in the water, both of which are important for the wastewater treatment results. The concentrations of these substances in the inlet water vary during the day and are difficult to monitor properly. The purpose of this study was to investigate whether it is possible, from the more easily measured variables, to obtain information on those which require more extensive analysis. This was done by using multivariate analysis to create models attempting to explain the variation in these variables. The models are commonly referred to as soft sensors, since they don’t actually make use of any physical sensors to measure the relevant variable. Data were collected during the period of March 11 to March 15, 2013 in the wastewater at different stages of the treatment process and a number of multivariate models were created. The result shows that it is possible to obtain information about the variables with PLS models based on easy-to-measure variables. The best created model was the one explaining the concentration of nitrogen in the inlet water.

L'analyse statistique de données de séquençage à haut débit (NGS) pose des questions computationnelles concernant la modélisation et l'inférence, en particulier à cause de la grande dimension des données. Le travail de recherche dans ce manuscrit porte sur des méthodes de réductions de dimension hybrides, basées sur des approches de compression (représentation dans un espace de faible dimension) et de sélection de variables. Des développements sont menés concernant la régression "Partial Least Squares" parcimonieuse (supervisée) et les méthodes de factorisation parcimonieuse de matrices (non supervisée). Dans les deux cas, notre objectif sera la reconstruction et la visualisation des données. Nous présenterons une nouvelle approche de type PLS parcimonieuse, basée sur une pénalité adaptative, pour la régression logistique. Cette approche sera utilisée pour des problèmes de prédiction (devenir de patients ou type cellulaire) à partir de l'expression des gènes. La principale problématique sera de prendre en compte la réponse pour écarter les variables non pertinentes. Nous mettrons en avant le lien entre la construction des algorithmes et la fiabilité des résultats.Dans une seconde partie, motivés par des questions relatives à l'analyse de données "single-cell", nous proposons une approche probabiliste pour la factorisation de matrices de comptage, laquelle prend en compte la sur-dispersion et l'amplification des zéros (caractéristiques des données single-cell). Nous développerons une procédure d'estimation basée sur l'inférence variationnelle. Nous introduirons également une procédure de sélection de variables probabiliste basée sur un modèle "spike-and-slab". L'intérêt de notre méthode pour la reconstruction, la visualisation et le clustering de données sera illustré par des simulations et par des résultats préliminaires concernant une analyse de données "single-cell". Toutes les méthodes proposées sont implémentées dans deux packages R: plsgenomics et CMF / The statistical analysis of Next-Generation Sequencing data raises many computational challenges regarding modeling and inference, especially because of the high dimensionality of genomic data. The research work in this manuscript concerns hybrid dimension reduction methods that rely on both compression (representation of the data into a lower dimensional space) and variable selection. Developments are made concerning: the sparse Partial Least Squares (PLS) regression framework for supervised classification, and the sparse matrix factorization framework for unsupervised exploration. In both situations, our main purpose will be to focus on the reconstruction and visualization of the data. First, we will present a new sparse PLS approach, based on an adaptive sparsity-inducing penalty, that is suitable for logistic regression to predict the label of a discrete outcome. For instance, such a method will be used for prediction (fate of patients or specific type of unidentified single cells) based on gene expression profiles. The main issue in such framework is to account for the response to discard irrelevant variables. We will highlight the direct link between the derivation of the algorithms and the reliability of the results. Then, motivated by questions regarding single-cell data analysis, we propose a flexible model-based approach for the factorization of count matrices, that accounts for over-dispersion as well as zero-inflation (both characteristic of single-cell data), for which we derive an estimation procedure based on variational inference. In this scheme, we consider probabilistic variable selection based on a spike-and-slab model suitable for count data. The interest of our procedure for data reconstruction, visualization and clustering will be illustrated by simulation experiments and by preliminary results on single-cell data analysis. All proposed methods were implemented into two R-packages "plsgenomics" and "CMF" based on high performance computing

De nombreux constats contemporains s’alarment du malaise récurrent des ressources humaines hospitalières, particulièrement à l’endroit des médecins et des soignants, et par conséquent du risque de mauvaise qualité de prise en charge des patients. Adoptant une approche plus optimiste, des chercheurs américains en soins infirmiers ont mis en évidence depuis le début des années 1980 l’existence d’hôpitaux dits magnétiques, parce qu’attractifs et fidélisateurs, et où il ferait bon travailler et se faire soigner. Cette thèse vise à approfondir le concept de Magnet Hospital, à éclairer sa définition et sa portée pour la gestion des ressources humaines hospitalières en France. Suivant une démarche hypothético-déductive, la conceptualisation, fondée sur un état de l’art, débute par une appropriation du modèle synthétique du Magnet Hospital. Empruntant une perspective psychosociale, notre modèle original de recherche se focalise sur la perception, à l’échelle des unités de soins, des attributs managériaux du magnétisme hospitalier (leadership transformationnel, empowerment perçu de la participation et climat relationnel collégial entre médecins et soignants) et ses conséquences attitudinales positives (satisfaction, implication, intention de rester, équilibre émotionnel travail/hors travail et efficacité collective perçue). Une méthodologie quantitative interroge au moyen de 8 échelles ad hoc un échantillon représentatif de 133 médecins, 361 infirmières et 362 aides-soignantes de 36 services de médecine polyvalente français. Une série de modélisations par équations structurelles, selon l’algorithme Partial Least Squares, teste la nature et l’intensité des relations directes et indirectes du magnétisme managérial perçu. Les résultats statistiques indiquent une bonne qualité des construits et d’ajustement des modèles. Un contexte managérial magnétique produit son principal effet positif sur l’efficacité collective perçue. Des différences catégorielles existent quant à la perception de sa composition et à la transmission de ses effets par la médiation de l’efficacité collective perçue, signalant le caractère contingent du magnétisme. Ces résultats ouvrent des perspectives managériales et scientifiques, en soulignant l’intérêt des approches positives de l’organisation hospitalière. / Many contemporary findings are alarmed of the recurring discomfort of hospital human resources, especially against doctors and nurses, and consequently against risk of poor quality of care for patients. Adopting a more optimistic approach, American nursing scholars have highlighted since the 1980s, some magnet hospitals, able to attract and retain, and with good working and care conditions. This thesis aims to explore Magnet Hospital concept, to inform its definition and scope for hospital human resource management in France. According to a hypothetico-deductive approach, based on a review of the literature, the conceptualization begins with appropriation of synthetic Magnet Hospital model. Under a psychosocial perspective, our original research model focuses on perception of managerial magnetic attributes (transformational leadership, perceived empowerment of participation, collegial climate between doctors and nurses) and their consequences on positive job attitudes (satisfaction, commitment, intent to rest, emotional equilibrium work/family, perceived collective efficacy), at wards level. A quantitative methodology proceeds by a questionnaire of 8 ad hoc scales and interviews 133 doctors, 361 nurses, 362, auxiliary nurses, in 36 French medicine units. A set of structural equations modeling, according to Partial Least Squares, tests nature and intensity of direct and indirect relationships of perceived managerial magnetism. The statistical results show a good validity of constructs and a good fit of models. The major positive effect of magnetic managerial context is on perceived collective efficacy. Some professional differences exist about perceptions of composition and transmission of magnetic effects (via mediation of perceived collective efficacy), indicating the contingency of magnetism. These findings open managerial and scientific opportunities, emphasizing the interest for positive organizational approach of hospital.