Large-scale clustering through functional embedding

We present a new framework for large-scale data clustering. The main idea is to modify functional dimensionality reduction techniques to directly optimize over discrete labels using stochastic gradient descent. Compared to methods like spectral clustering our approach solves a single optimization problem, rather than an ad-hoc two-stage optimization approach, does not require a matrix inversion, can easily encode prior knowledge in the set of implementable functions, and does not have an ?out-of-sample? problem. Experimental results on both artificial and real-world datasets show the usefulness of our approach.

Functional error modeling for uncertainty quantification in hydrogeology

Approximate models (proxies) can be employed to reduce the computational costs of estimating uncertainty. The price to pay is that the approximations introduced by the proxy model can lead to a biased estimation. To avoid this problem and ensure a reliable uncertainty quantification, we propose to combine functional data analysis and machine learning to build error models that allow us to obtain an accurate prediction of the exact response without solving the exact model for all realizations. We build the relationship between proxy and exact model on a learning set of geostatistical realizations for which both exact and approximate solvers are run. Functional principal components analysis (FPCA) is used to investigate the variability in the two sets of curves and reduce the dimensionality of the problem while maximizing the retained information. Once obtained, the error model can be used to predict the exact response of any realization on the basis of the sole proxy response. This methodology is purpose-oriented as the error model is constructed directly for the quantity of interest, rather than for the state of the system. Also, the dimensionality reduction performed by FPCA allows a diagnostic of the quality of the error model to assess the informativeness of the learning set and the fidelity of the proxy to the exact model. The possibility of obtaining a prediction of the exact response for any newly generated realization suggests that the methodology can be effectively used beyond the context of uncertainty quantification, in particular for Bayesian inference and optimization.

Error models in hydrogeology applications

Notre consommation en eau souterraine, en particulier comme eau potable ou pour l'irrigation, a considérablement augmenté au cours des années. De nombreux problèmes font alors leur apparition, allant de la prospection de nouvelles ressources à la remédiation des aquifères pollués. Indépendamment du problème hydrogéologique considéré, le principal défi reste la caractérisation des propriétés du sous-sol. Une approche stochastique est alors nécessaire afin de représenter cette incertitude en considérant de multiples scénarios géologiques et en générant un grand nombre de réalisations géostatistiques. Nous rencontrons alors la principale limitation de ces approches qui est le coût de calcul dû à la simulation des processus d'écoulements complexes pour chacune de ces réalisations. Dans la première partie de la thèse, ce problème est investigué dans le contexte de propagation de l'incertitude, oú un ensemble de réalisations est identifié comme représentant les propriétés du sous-sol. Afin de propager cette incertitude à la quantité d'intérêt tout en limitant le coût de calcul, les méthodes actuelles font appel à des modèles d'écoulement approximés. Cela permet l'identification d'un sous-ensemble de réalisations représentant la variabilité de l'ensemble initial. Le modèle complexe d'écoulement est alors évalué uniquement pour ce sousensemble, et, sur la base de ces réponses complexes, l'inférence est faite. Notre objectif est d'améliorer la performance de cette approche en utilisant toute l'information à disposition. Pour cela, le sous-ensemble de réponses approximées et exactes est utilisé afin de construire un modèle d'erreur, qui sert ensuite à corriger le reste des réponses approximées et prédire la réponse du modèle complexe. Cette méthode permet de maximiser l'utilisation de l'information à disposition sans augmentation perceptible du temps de calcul. La propagation de l'incertitude est alors plus précise et plus robuste. La stratégie explorée dans le premier chapitre consiste à apprendre d'un sous-ensemble de réalisations la relation entre les modèles d'écoulement approximé et complexe. Dans la seconde partie de la thèse, cette méthodologie est formalisée mathématiquement en introduisant un modèle de régression entre les réponses fonctionnelles. Comme ce problème est mal posé, il est nécessaire d'en réduire la dimensionnalité. Dans cette optique, l'innovation du travail présenté provient de l'utilisation de l'analyse en composantes principales fonctionnelles (ACPF), qui non seulement effectue la réduction de dimensionnalités tout en maximisant l'information retenue, mais permet aussi de diagnostiquer la qualité du modèle d'erreur dans cet espace fonctionnel. La méthodologie proposée est appliquée à un problème de pollution par une phase liquide nonaqueuse et les résultats obtenus montrent que le modèle d'erreur permet une forte réduction du temps de calcul tout en estimant correctement l'incertitude. De plus, pour chaque réponse approximée, une prédiction de la réponse complexe est fournie par le modèle d'erreur. Le concept de modèle d'erreur fonctionnel est donc pertinent pour la propagation de l'incertitude, mais aussi pour les problèmes d'inférence bayésienne. Les méthodes de Monte Carlo par chaîne de Markov (MCMC) sont les algorithmes les plus communément utilisés afin de générer des réalisations géostatistiques en accord avec les observations. Cependant, ces méthodes souffrent d'un taux d'acceptation très bas pour les problèmes de grande dimensionnalité, résultant en un grand nombre de simulations d'écoulement gaspillées. Une approche en deux temps, le "MCMC en deux étapes", a été introduite afin d'éviter les simulations du modèle complexe inutiles par une évaluation préliminaire de la réalisation. Dans la troisième partie de la thèse, le modèle d'écoulement approximé couplé à un modèle d'erreur sert d'évaluation préliminaire pour le "MCMC en deux étapes". Nous démontrons une augmentation du taux d'acceptation par un facteur de 1.5 à 3 en comparaison avec une implémentation classique de MCMC. Une question reste sans réponse : comment choisir la taille de l'ensemble d'entrainement et comment identifier les réalisations permettant d'optimiser la construction du modèle d'erreur. Cela requiert une stratégie itérative afin que, à chaque nouvelle simulation d'écoulement, le modèle d'erreur soit amélioré en incorporant les nouvelles informations. Ceci est développé dans la quatrième partie de la thèse, oú cette méthodologie est appliquée à un problème d'intrusion saline dans un aquifère côtier. -- Our consumption of groundwater, in particular as drinking water and for irrigation, has considerably increased over the years and groundwater is becoming an increasingly scarce and endangered resource. Nofadays, we are facing many problems ranging from water prospection to sustainable management and remediation of polluted aquifers. Independently of the hydrogeological problem, the main challenge remains dealing with the incomplete knofledge of the underground properties. Stochastic approaches have been developed to represent this uncertainty by considering multiple geological scenarios and generating a large number of realizations. The main limitation of this approach is the computational cost associated with performing complex of simulations in each realization. In the first part of the thesis, we explore this issue in the context of uncertainty propagation, where an ensemble of geostatistical realizations is identified as representative of the subsurface uncertainty. To propagate this lack of knofledge to the quantity of interest (e.g., the concentration of pollutant in extracted water), it is necessary to evaluate the of response of each realization. Due to computational constraints, state-of-the-art methods make use of approximate of simulation, to identify a subset of realizations that represents the variability of the ensemble. The complex and computationally heavy of model is then run for this subset based on which inference is made. Our objective is to increase the performance of this approach by using all of the available information and not solely the subset of exact responses. Two error models are proposed to correct the approximate responses follofing a machine learning approach. For the subset identified by a classical approach (here the distance kernel method) both the approximate and the exact responses are knofn. This information is used to construct an error model and correct the ensemble of approximate responses to predict the "expected" responses of the exact model. The proposed methodology makes use of all the available information without perceptible additional computational costs and leads to an increase in accuracy and robustness of the uncertainty propagation. The strategy explored in the first chapter consists in learning from a subset of realizations the relationship between proxy and exact curves. In the second part of this thesis, the strategy is formalized in a rigorous mathematical framework by defining a regression model between functions. As this problem is ill-posed, it is necessary to reduce its dimensionality. The novelty of the work comes from the use of functional principal component analysis (FPCA), which not only performs the dimensionality reduction while maximizing the retained information, but also allofs a diagnostic of the quality of the error model in the functional space. The proposed methodology is applied to a pollution problem by a non-aqueous phase-liquid. The error model allofs a strong reduction of the computational cost while providing a good estimate of the uncertainty. The individual correction of the proxy response by the error model leads to an excellent prediction of the exact response, opening the door to many applications. The concept of functional error model is useful not only in the context of uncertainty propagation, but also, and maybe even more so, to perform Bayesian inference. Monte Carlo Markov Chain (MCMC) algorithms are the most common choice to ensure that the generated realizations are sampled in accordance with the observations. Hofever, this approach suffers from lof acceptance rate in high dimensional problems, resulting in a large number of wasted of simulations. This led to the introduction of two-stage MCMC, where the computational cost is decreased by avoiding unnecessary simulation of the exact of thanks to a preliminary evaluation of the proposal. In the third part of the thesis, a proxy is coupled to an error model to provide an approximate response for the two-stage MCMC set-up. We demonstrate an increase in acceptance rate by a factor three with respect to one-stage MCMC results. An open question remains: hof do we choose the size of the learning set and identify the realizations to optimize the construction of the error model. This requires devising an iterative strategy to construct the error model, such that, as new of simulations are performed, the error model is iteratively improved by incorporating the new information. This is discussed in the fourth part of the thesis, in which we apply this methodology to a problem of saline intrusion in a coastal aquifer.

Identification and visualization of multidimensional antigen-specific T-cell populations in polychromatic cytometry data.

An important aspect of immune monitoring for vaccine development, clinical trials, and research is the detection, measurement, and comparison of antigen-specific T-cells from subject samples under different conditions. Antigen-specific T-cells compose a very small fraction of total T-cells. Developments in cytometry technology over the past five years have enabled the measurement of single-cells in a multivariate and high-throughput manner. This growth in both dimensionality and quantity of data continues to pose a challenge for effective identification and visualization of rare cell subsets, such as antigen-specific T-cells. Dimension reduction and feature extraction play pivotal role in both identifying and visualizing cell populations of interest in large, multi-dimensional cytometry datasets. However, the automated identification and visualization of rare, high-dimensional cell subsets remains challenging. Here we demonstrate how a systematic and integrated approach combining targeted feature extraction with dimension reduction can be used to identify and visualize biological differences in rare, antigen-specific cell populations. By using OpenCyto to perform semi-automated gating and features extraction of flow cytometry data, followed by dimensionality reduction with t-SNE we are able to identify polyfunctional subpopulations of antigen-specific T-cells and visualize treatment-specific differences between them.

Fatigue et réduction de la performance motrice chez le sportif, syndrome de surentraînement [Fatigue and reduction in motor performance in sportspeople or overtraining syndrome].

The main goal of training activities is to improve motor performance. After strenuous workouts, it is physiological to experience fatigue, which relieves within two weeks, and then induce an improvement in motor capacities. An overtraining syndrome is diagnosed when fatigue is postponed beyond two weeks, and affects mainly endurance athletes. It is a condition of chronic fatigue, underperformance and an increased vulnerability to infection leading to recurrent infections. The whole observed spectrum of symptoms is physiological, psychological, endocrinogical and immunological. All play a role in the failure to recover. Monitoring of athletes activities helps to prevent the syndrome with days with no sports. Rest, patience and empathy are the only ways of treatment options.

Dimensionality of the Rosenberg self-esteem scale and its relationships with the three- and five-factor personality models

The authors investigated the dimensionality of the French version of the Rosenberg Self-Esteem Scale (RSES; Rosenberg, 1965) using confirmatory factor analysis. We tested models of 1 or 2 factors. Results suggest the RSES is a 1-dimensional scale with 3 highly correlated items. Comparison with the Revised NEO-Personality Inventory (NEO-PI-R; Costa, McCrae, & Rolland, 1998) demonstrated that Neuroticism correlated strongly and Extraversion and Conscientiousness moderately with the RSES. Depression accounted for 47% of the variance of the RSES. Other NEO-PI-R facets were also moderately related with self-esteem.

Reduction of small infrarenal abdominal aortic aneurysm expansion rate by statins.

BACKGROUND: To evaluate the effect of statins on the annual expansion rate (ER) of small infrarenal abdominal aortic aneurysms (AAA). PATIENTS AND METHODS: All patients under regular surveillance for small AAA between January 2000 and September 2007, in the Department of Angiology, Lausanne University Hospital, were included. Inclusion criteria were baseline abdominal aortic diameter between 25 and 55 mm, at least two measurements of AAA diameter and a minimum follow up of 6 months. Patients with Marfan disease, infectious or inflammatory AAA, and patients with prior AAA repair were excluded. The influence of statin use and other factors on ER were examined by bivariate and multivariate analysis. RESULTS: Among 589 patients who underwent an abdominal aorta evaluation, 94 patients (89 % men, mean age 69.1 years) were finally included in the analysis. Baseline AAA size was 39.9 ± 7.7 mm (mean±SE) and 48.7 ± 8.4 mm at end of follow-up. Patients had a regular aneurysm size assessment during 38.5 ± 27.7 months. Mean ER was 3.59 mm/y (± 2.81). The 50 patients who were treated with statin during the study period had a lower ER compared to the 44 controls (2.91 vs 4.37 mm/year, p = 0.01). CONCLUSIONS: This study confirms the considerable individual variations in the AAA expansion rate, and emphasizes the need for regular aortic diameter assessments. In this study, patients treated with statin demonstrate a significant decrease in the ER compared to controls. This finding need to be evaluated in prospective interventional studies powered to demonstrate the potential benefit of statin treatment.

A randomized controlled clinical trial evaluating the performance and safety of platelets treated with MIRASOL pathogen reduction technology.

BACKGROUND: Pathogen reduction of platelets (PRT-PLTs) using riboflavin and ultraviolet light treatment has undergone Phase 1 and 2 studies examining efficacy and safety. This randomized controlled clinical trial (RCT) assessed the efficacy and safety of PRT-PLTs using the 1-hour corrected count increment (CCI(1hour) ) as the primary outcome. STUDY DESIGN AND METHODS: A noninferiority RCT was performed where patients with chemotherapy-induced thrombocytopenia (six centers) were randomly allocated to receive PRT-PLTs (Mirasol PRT, CaridianBCT Biotechnologies) or reference platelet (PLT) products. The treatment period was 28 days followed by a 28-day follow-up (safety) period. The primary outcome was the CCI(1hour) determined using up to the first eight on-protocol PLT transfusions given during the treatment period. RESULTS: A total of 118 patients were randomly assigned (60 to PRT-PLTs; 58 to reference). Four patients per group did not require PLT transfusions leaving 110 patients in the analysis (56 PRT-PLTs; 54 reference). A total of 541 on-protocol PLT transfusions were given (303 PRT-PLTs; 238 reference). The least square mean CCI was 11,725 (standard error [SE], 1.140) for PRT-PLTs and 16,939 (SE, 1.149) for the reference group (difference, -5214; 95% confidence interval, -7542 to -2887; p<0.0001 for a test of the null hypothesis of no difference between the two groups). CONCLUSION: The study failed to show noninferiority of PRT-PLTs based on predefined CCI criteria. PLT and red blood cell utilization in the two groups was not significantly different suggesting that the slightly lower CCIs (PRT-PLTs) did not increase blood product utilization. Safety data showed similar findings in the two groups. Further studies are required to determine if the lower CCI observed with PRT-PLTs translates into an increased risk of bleeding.

Genotype-level interactions determine the degree of reduction of leaf rust on wheat by seed application of beneficial Pseudomonas spp.

Some bacteria have the capacity to reduce incidence and severity of plant diseases either by inhibiting the pathogen or by modulating the resistance response of the plant. Plants dispose of different resistance mechanisms that are influenced by the biotic and abiotic environment. The present experiments explored the effects of biocontrol strains of Pseudomonas fluorescens on the resistance of wheat varieties against brown rust disease caused by Puccinia triticina. Root inoculation with biocontrol pseudomonads reduced the disease severity on the leaves. The plant response depended on the genotype of both the microbes and the wheat varieties, suggesting a straight interaction at the molecular level.

Withdrawal, "serosorting" and "strategic positioning": use of risk reduction strategies with casual sex partners in men who have sex with men (MSM) in Switzerland, 2007

Introduction: 1) Withdrawal before ejaculation, "serosorting" (to choose a partner of same serostatus) and "strategic positioning" (only insertive vs. only receptive role in anal sex according to serostatus) are known to be used by MSM as alternatives to condom use. 2) Despite their questionable levels of effectiveness they are collectively labelled as "risk reduction strategies" (RRS). Objectives: The aim of this study is to estimate the prevalence and factors related to RRS in men who report unprotected anal intercourse (UAI) with occasional partners in the last 12 months. Methods: 1) In 2007, a module on RRS was included in a repeated national survey conducted among readers of gay newspapers, members of gay organizations and visitors of gay websites (N=2953). 2) Using an anonymous self-completed questionnaire, participants were asked whether, with the aim of avoiding HIV infection, RRS were used with occasional partners. Analysis: 1) Prevalences were calculated in participants who reported UAI with occasional partners in the last 12 months (n=416). 2) A logistic regression was performed, using "at least one RRS" as dependent variable. Number of partners in the last 12 months, HIV-status and usual socio-demographic characteristics were used as independent factors. Result : 1) 70% (292/416) of the participants reporting UAI used at least one RRS when they had unprotected sex with casual partners in the last 12 months (Table 1). 2) Withrawal before ejaculation was the most frequently reported strategy, followed by serosorting and strategic positioning (Table 1). 3) Participants who reported at least one RRS were more likely to be over 30 years and to belong to a gay organisation. HIV-positive and non-tested participants were less likely to report RRS than HIV-negative participants (Table 2). Conclusions: 1) The majority of MSM who reported UAI in the last 12 months tried to reduce risk of HIV transmission by using specific strategies (withdrawal, serosorting, strategic positioning). It is not known, however, to what extent the use of these strategies was systematic. 2) It is necessary to provide MSM with balanced information on these strategies and their respective level of effectiveness. 3) It is important to monitor the use of RRS in HIV behavioural surveillance surveys in MSM.

Dimensionality of drinking consequences - cross-cultural comparability and stability over time

Despite the long tradition for asking about the negative social and health consequences of alcohol consumption in surveys, little is known about the dimensionality of these consequences. Analysing cross-sectional and longitudinal data from the Nordic Taxation Study collected for Sweden, Finland, and Denmark in two waves in 2003 and 2004 by means of an explorative principal component analysis for categorical data (CATPCA), it is tested whether consequences have a single underlying dimension across cultures. It further tests the reliability, replicability, concurrent and predictive validity of the consequence scales. A one-dimensional solution was commonly preferable. Whereas the two-dimensional solution was unable to distinguish clearly between different concepts of consequences, the one-dimensional solution resulted in interpretable, generally very stable scales within countries across different samples and time.

The contribution of aging to the understanding of the dimensionality of executive functions.

It has been reported in the literature that executive functions may be fractioned into updating, shifting, and inhibition. The present study aimed to explore whether these executive sub-components can be identified in a more age-heterogeneous sample and see if they are prone to an age-related decline. We tested the performances of 81 individuals aged from 18 to 88 years old in each executive sub-component, working memory, fluid intelligence and processing speed. Correlation analysis revealed only a slight positive relationship between the two updating measures. A linear decrement with age was observed only for two complex executive tests. Tasks indexing working memory, processing speed and fluid intelligence showed a stronger linear decline with age than executive tasks. In conclusion, our results did not replicate the executive structure known from the literature, and revealed that decrement in executive function is not an unavoidable concomitant of aging but rather concerns specific executive tasks.

Paclitaxel-eluting biodegradable synthetic vascular prostheses: a step towards reduction of neointima formation?

BACKGROUND: Clinical small-caliber vascular prostheses are unsatisfactory. Reasons for failure are early thrombosis and late intimal hyperplasia. We thus prepared biodegradable small-caliber vascular prostheses using electrospun polycaprolactone (PCL) with slow-releasing paclitaxel (PTX), an antiproliferative drug. METHODS AND RESULTS: PCL solutions containing PTX were used to prepare nonwoven nanofibre-based 2-mm ID prostheses. Mechanical morphological properties and drug loading, distribution, and release were studied in vitro. Infrarenal abdominal aortic replacement was carried out with nondrug-loaded and drug-loaded prostheses in 18 rats and followed for 6 months. Patency, stenosis, tissue reaction, and drug effect on endothelialization, vascular remodeling, and neointima formation were studied in vivo. In vitro prostheses showed controlled morphology mimicking extracellular matrix with mechanical properties similar to those of native vessels. PTX-loaded grafts with suitable mechanical properties and controlled drug-release were obtained by factorial design. In vivo, both groups showed 100% patency, no stenosis, and no aneurysmal dilatation. Endothelial coverage and cell ingrowth were significantly reduced at 3 weeks and delayed at 12 and 24 weeks in PTX grafts, but as envisioned, neointima formation was significantly reduced in these grafts at 12 weeks and delayed at 6 months. CONCLUSIONS: Biodegradable, electrospun, nanofibre, polycaprolactone prostheses are promising because in vitro they maintain their mechanical properties (regardless of PTX loading), and in vivo show good patency, reendothelialize, and remodel with autologous cells. PTX loading delays endothelialization and cellular ingrowth. Conversely, it reduces neointima formation until the end point of our study and thus may be an interesting option for small caliber vascular grafts.