A comparison of mean phase difference and generalized least squares for analyzing single-case data

The present study focuses on single-case data analysis and specifically on two procedures for quantifying differences between baseline and treatment measurements The first technique tested is based on generalized least squares regression analysis and is compared to a proposed non-regression technique, which allows obtaining similar information. The comparison is carried out in the context of generated data representing a variety of patterns (i.e., independent measurements, different serial dependence underlying processes, constant or phase-specific autocorrelation and data variability, different types of trend, and slope and level change). The results suggest that the two techniques perform adequately for a wide range of conditions and researchers can use both of them with certain guarantees. The regression-based procedure offers more efficient estimates, whereas the proposed non-regression procedure is more sensitive to intervention effects. Considering current and previous findings, some tentative recommendations are offered to applied researchers in order to help choosing among the plurality of single-case data analysis techniques.

Chronic subdural hematomas treated by burr hole trepanation and a subperiostal drainage system.

OBJECTIVE: Most symptomatic chronic subdural hematomas are treated by subdural drainage. However, a subperiostal (i.e., extracranial) passive closed-drainage system in combination with double burr hole trepanation is used at our institution. Therefore, we wanted to analyze our results and compare them with the alternate treatment strategies reported in the current literature. METHODS: In a retrospective single-center study, we analyzed the data of all patients undergoing double burr hole trepanation with a subperiostal passive closed-drainage system. Data analysis included general patient data, complications, postoperative seizure rate, and outcome. RESULTS: One hundred forty-seven patients underwent surgery for 183 symptomatic chronic subdural hematomas. The perioperative mortality rate was 3.4%. Hematoma persistence or recurrence occurred in 13.1% of the cases. The postoperative seizure rate was 6.6%, and the infection rate was 1.6%, including 3 cases of superficial wound infection and 1 case with deep infection. The reintervention rate was 9.3%, including trepanation in 8.2% of the patients and craniotomy in 1.1%. The overall complication rate was 10.9%. CONCLUSION: Double burr hole trepanation combined with a subperiostal passive closed-drainage system is a technically easy, highly effective, safe, and cost-efficient treatment strategy for symptomatic chronic subdural hematomas. The absence of a drain in direct contact with the hematoma capsule may moderate the risk of postoperative seizure and limit the secondary spread of infection to intracranial compartments.

Comparative modular analysis of gene expression in vertebrate development

The focus of my PhD research was the concept of modularity. In the last 15 years, modularity has become a classic term in different fields of biology. On the conceptual level, a module is a set of interacting elements that remain mostly independent from the elements outside of the module. I used modular analysis techniques to study gene expression evolution in vertebrates. In particular, I identified ``natural'' modules of gene expression in mouse and human, and I showed that expression of organ-specific and system-specific genes tends to be conserved between such distance vertebrates as mammals and fishes. Also with a modular approach, I studied patterns of developmental constraints on transcriptome evolution. I showed that none of the two commonly accepted models of the evolution of embryonic development (``evo-devo'') are exclusively valid. In particular, I found that the conservation of the sequences of regulatory regions is highest during mid-development of zebrafish, and thus it supports the ``hourglass model''. In contrast, events of gene duplication and new gene introduction are most rare in early development, which supports the ``early conservation model''. In addition to the biological insights on transcriptome evolution, I have also discussed in detail the advantages of modular approaches in large-scale data analysis. Moreover, I re-analyzed several studies (published in high-ranking journals), and showed that their conclusions do not hold out under a detailed analysis. This demonstrates that complex analysis of high-throughput data requires a co-operation between biologists, bioinformaticians, and statisticians.

Differing trends in the association between obesity and self-reported health in Portugal and Switzerland: data from national health surveys 1992-2007.

BACKGROUND: The escalating prevalence of obesity might prompt obese subjects to consider themselves as normal, as this condition is gradually becoming as frequent as normal weight. In this study, we aimed to assess the trends in the associations between obesity and self-rated health in two countries. METHODS: Data from the Portuguese (years 1995-6, 1998-6 and 2005-6) and Swiss (1992-3, 1997, 2002 and 2007) National Health Surveys were used, corresponding to more than 130,000 adults (64,793 for Portugal and 65,829 for Switzerland). Body mass index and self-rated health were derived from self-reported data. RESULTS: Obesity levels were higher in Portugal (17.5% in 2005-6 vs. 8.9% in 2007 in Switzerland, p < 0.001) and increased in both countries. The prevalence of participants rating their health as "bad" or "very bad" was higher in Portugal than in Switzerland (21.8% in 2005-6 vs 3.9% in 2007, p < 0.001). In both countries, obese participants rated more frequently their health as "bad" or "very bad" than participants with regular weight. In Switzerland, the prevalence of "bad" or "very bad" rates among obese participants, increased from 6.5% in 1992-3 to 9.8% in 2007, while in Portugal it decreased from 41.3% to 32.3%. After multivariate adjustment, the odds ratio (OR) of stating one self's health as "bad" or "very bad" among obese relative to normal weight participants, almost doubled in Switzerland: from 1.38 (95% confidence interval, CI: 1.01-1.87) in 1992-3 to 2.64 (95% CI: 2.14-3.26) in 2007, and similar findings were obtained after sample weighting. Conversely, no such trend was found in Portugal: 1.35 (95% CI: 1.23-1.48) in 1995-6 and 1.52 (95% CI: 1.37-1.70) in 2005-6. CONCLUSION: Obesity is increasing in Switzerland and Portugal. Obesity is increasingly associated with poorer self-health ratings in Switzerland but not in Portugal.

Kernel-based Mapping of Meteorological Fields in Complex Orography

Due to the advances in sensor networks and remote sensing technologies, the acquisition and storage rates of meteorological and climatological data increases every day and ask for novel and efficient processing algorithms. A fundamental problem of data analysis and modeling is the spatial prediction of meteorological variables in complex orography, which serves among others to extended climatological analyses, for the assimilation of data into numerical weather prediction models, for preparing inputs to hydrological models and for real time monitoring and short-term forecasting of weather.In this thesis, a new framework for spatial estimation is proposed by taking advantage of a class of algorithms emerging from the statistical learning theory. Nonparametric kernel-based methods for nonlinear data classification, regression and target detection, known as support vector machines (SVM), are adapted for mapping of meteorological variables in complex orography.With the advent of high resolution digital elevation models, the field of spatial prediction met new horizons. In fact, by exploiting image processing tools along with physical heuristics, an incredible number of terrain features which account for the topographic conditions at multiple spatial scales can be extracted. Such features are highly relevant for the mapping of meteorological variables because they control a considerable part of the spatial variability of meteorological fields in the complex Alpine orography. For instance, patterns of orographic rainfall, wind speed and cold air pools are known to be correlated with particular terrain forms, e.g. convex/concave surfaces and upwind sides of mountain slopes.Kernel-based methods are employed to learn the nonlinear statistical dependence which links the multidimensional space of geographical and topographic explanatory variables to the variable of interest, that is the wind speed as measured at the weather stations or the occurrence of orographic rainfall patterns as extracted from sequences of radar images. Compared to low dimensional models integrating only the geographical coordinates, the proposed framework opens a way to regionalize meteorological variables which are multidimensional in nature and rarely show spatial auto-correlation in the original space making the use of classical geostatistics tangled.The challenges which are explored during the thesis are manifolds. First, the complexity of models is optimized to impose appropriate smoothness properties and reduce the impact of noisy measurements. Secondly, a multiple kernel extension of SVM is considered to select the multiscale features which explain most of the spatial variability of wind speed. Then, SVM target detection methods are implemented to describe the orographic conditions which cause persistent and stationary rainfall patterns. Finally, the optimal splitting of the data is studied to estimate realistic performances and confidence intervals characterizing the uncertainty of predictions.The resulting maps of average wind speeds find applications within renewable resources assessment and opens a route to decrease the temporal scale of analysis to meet hydrological requirements. Furthermore, the maps depicting the susceptibility to orographic rainfall enhancement can be used to improve current radar-based quantitative precipitation estimation and forecasting systems and to generate stochastic ensembles of precipitation fields conditioned upon the orography.

Corporate responsibility in the postnational constellation : a multiple-case study

In my thesis I present the findings of a multiple-case study on the CSR approach of three multinational companies, applying Basu and Palazzo's (2008) CSR-character as a process model of sensemaking, Suchman's (1995) framework on legitimation strategies, and Habermas (1996) concept of deliberative democracy. The theoretical framework is based on the assumption of a postnational constellation (Habermas, 2001) which sends multinational companies onto a process of sensemaking (Weick, 1995) with regards to their responsibilities in a globalizing world. The major reason is that mainstream CSR-concepts are based on the assumption of a liberal market economy embedded in a nation state that do not fit the changing conditions for legitimation of corporate behavior in a globalizing world. For the purpose of this study, I primarily looked at two research questions: (i) How can the CSR approach of a multinational corporation be systematized empirically? (ii) What is the impact of the changing conditions in the postnational constellation on the CSR approach of the studied multinational corporations? For the analysis, I adopted a holistic approach (Patton, 1980), combining elements of a deductive and inductive theory building methodology (Eisenhardt, 1989b; Eisenhardt & Graebner, 2007; Glaser & Strauss, 1967; Van de Ven, 1992) and rigorous qualitative data analysis. Primary data was collected through 90 semi-structured interviews in two rounds with executives and managers in three multinational companies and their respective stakeholders. Raw data originating from interview tapes, field notes, and contact sheets was processed, stored, and managed using the software program QSR NVIVO 7. In the analysis, I applied qualitative methods to strengthen the interpretative part as well as quantitative methods to identify dominating dimensions and patterns. I found three different coping behaviors that provide insights into the corporate mindset. The results suggest that multinational corporations increasingly turn towards relational approaches of CSR to achieve moral legitimacy in formalized dialogical exchanges with their stakeholders since legitimacy can no longer be derived only from a national framework. I also looked at the degree to which they have reacted to the postnational constellation by the assumption of former state duties and the underlying reasoning. The findings indicate that CSR approaches become increasingly comprehensive through integrating political strategies that reflect the growing (self-) perception of multinational companies as political actors. Based on the results, I developed a model which relates the different dimensions of corporate responsibility to the discussion on deliberative democracy, global governance and social innovation to provide guidance for multinational companies in a postnational world. With my thesis, I contribute to management research by (i) delivering a comprehensive critique of the mainstream CSR-literature and (ii) filling the gap of thorough qualitative research on CSR in a globalizing world using the CSR-character as an empirical device, and (iii) to organizational studies by further advancing a deliberative view of the firm proposed by Scherer and Palazzo (2008).

Chemotherapy in patients with advanced pancreatic cancer: too close to death?

PURPOSE: We evaluated the attitude in using chemotherapy near the end of life in advanced pancreatic adenocarcinoma (PAC). Clinical and laboratory parameters recorded at last chemotherapy administration were analyzed, in order to identify risk factors for imminent death. METHODS: Retrospective analysis of patients who underwent at least one line of palliative chemotherapy was made. Data concerning chemotherapy (regimens, lines, and date of last administration) were collected. Clinical and laboratory factors recorded at last chemotherapy administration were: performance status, presence of ascites, hemoglobin, white blood cell (WBC), platelets, total bilirubin, albumin, LDH, C-reactive protein (C-rp), and Ca 19.9. RESULTS: We analyzed 231 patients: males/females, 53/47 %; metastatic/locally advanced disease, 80/20 %; and median age, 66 years (range 32-85). All patients died due to disease progression. Median overall survival was 6.1 months (95 % CI 5.1-7.2). At the last chemotherapy delivery, performance status was 0-1 in 37 % and 2 in 63 %. Fifty-nine percent of patients received one chemotherapy line, while 32, 8, and 1 % had second-, third-, and fourth line, respectively. The interval between last chemotherapy administration and death was <4 weeks in 24 %, ≥4-12 in 47 %, and >12 in 29 %. Median survival from last chemotherapy to death was 7.5 weeks (95 % CI 6.7-8.4). In a univariate analysis, ascites, elevated WBC, bilirubin, LDH, C-rp and Ca 19.9, and reduced albumin were found to predict shorter survival; however, none of them remained significant in a multivariate analysis. CONCLUSIONS: A significant proportion of patients with advanced PAC received chemotherapy within the last month of life. The clinical and laboratory parameters recorded at last chemotherapy delivery did not predict shorter survival.

An epidemiologic study to determine the prevalence of the HLA-B*5701 allele among HIV-positive patients in Europe.

OBJECTIVES: HLA-B*5701 is a major histocompatibility complex class I allele associated with an immunologically-mediated hypersensitivity reaction to abacavir. The objectives of this study were to evaluate HLA-B*5701 prevalence among European, HIV-1-infected patients and to compare the local and central laboratory screening results. METHODS: Data were combined from six multicentre, prospective studies involving 10 European countries in which HIV-1-infected patients (irrespective of treatment experience or previous HLA-B*5701 screening), >or=18 years of age, were evaluated for HLA-B*5701 carriage, determined by the central and local laboratory methods. RESULTS: A total of 9720 patients from 272 centres were included in the analysis. The overall estimate of HLA-B*5701 prevalence in Europe was 4.98%, with country-specific estimates ranging from 1.53 to 7.75%. HLA-B*5701 prevalence was highest in the self-reported white population (6.49%) and lowest in the black population (0.39%). Local laboratory results had a high specificity (99.9%) and sensitivity (99.2%) when compared with the central laboratory results. CONCLUSION: This study supports data from previous studies regarding the prevalence of HLA-B*5701 in the HIV population and the variation of HLA-B*5701 prevalence between different racial groups. The high specificity and sensitivity of local laboratory results, suggests that clinicians can be confident in using local laboratories for pretreatment HLA-B*5701 screening. However, it is essential that local laboratories participate in HLA-B*5701-specific quality assurance programs to maintain 100% sensitivity. In HIV-infected patients, pretreatment HLA-B*5701 screening may allow more informed decisions regarding abacavir use and has the potential to significantly reduce the frequency of abacavir-related hypersensitivity reactions and costs associated with managing these reactions.

Seasonality of congenital anomalies in Europe.

BACKGROUND: This study describes seasonality of congenital anomalies in Europe to provide a baseline against which to assess the impact of specific time varying exposures such as the H1N1 pandemic influenza, and to provide a comprehensive and recent picture of seasonality and its possible relation to etiologic factors. METHODS: Data on births conceived in 2000 to 2008 were extracted from 20 European Surveillance for Congenital Anomalies population-based congenital anomaly registries in 14 European countries. We performed Poisson regression analysis encompassing sine and cosine terms to investigate seasonality of 65,764 nonchromosomal and 12,682 chromosomal congenital anomalies covering 3.3 million births. Analysis was performed by estimated month of conception. Analyses were performed for 86 congenital anomaly subgroups, including a combined subgroup of congenital anomalies previously associated with influenza. RESULTS: We detected statistically significant seasonality in prevalence of anomalies previously associated with influenza, but the conception peak was in June (2.4% excess). We also detected seasonality in congenital cataract (April conceptions, 27%), hip dislocation and/or dysplasia (April, 12%), congenital hydronephrosis (July, 12%), urinary defects (July, 5%), and situs inversus (December, 36%), but not for nonchromosomal anomalies combined, chromosomal anomalies combined, or other anomalies analyzed. CONCLUSION: We have confirmed previously described seasonality for congenital cataract and hip dislocation and/or dysplasia, and found seasonality for congenital hydronephrosis and situs inversus which have not previously been studied. We did not find evidence of seasonality for several anomalies which had previously been found to be seasonal. Influenza does not appear to be an important factor in the seasonality of congenital anomalies.

The impact of insight in a first-episode mania with psychosis population on outcome at 18 months.

BACKGROUND: To explore whether poor initial insight during a first episode of mania with psychotic features was predictive of poor psychosocial and clinical outcomes at 18 months. METHODS: Secondary analysis was performed on data collected during an 8-week RCT comparing the efficacy of olanzapine versus chlorpromazine as an adjunct to lithium, and at 18-month follow-up. 74 participants were divided into three groups (no insight, partial insight, and full insight) according to the insight item from the Young Mania Rating Scale (YMRS). Differences between these three groups were examined at baseline and at 18 months on measures of symptoms (YMRS, HAMD-21, and CGI-S), and social and occupational functioning (SOFAS). Baseline differences between the three groups were determined using general linear models and chi-squared analyses. Group differences from baseline to 18-month follow-up were determined using repeated measures general linear models. RESULTS: At baseline there were significant differences between the three insight groups in terms of mania and functioning, but at 18 months all groups had improved significantly in terms of psychopathology, mania, depression and social and occupational functioning. There were no significant differences between the three groups at study completion with respect to these domains. LIMITATIONS: The study was limited by the lack of availability of a more detailed rating scale for insight, and it did not account for the duration of untreated psychosis (DUI). CONCLUSIONS: Poor initial insight during a first episode of mania with psychotic features does not predict poor clinical and psychosocial outcome at 18 months.

Quantify this! Report on a round table discussion on quantitative mass spectrometry in proteomics.

Following the success of the first round table in 2001, the Swiss Proteomic Society has organized two additional specific events during its last two meetings: a proteomic application exercise in 2002 and a round table in 2003. Such events have as their main objective to bring together, around a challenging topic in mass spectrometry, two groups of specialists, those who develop and commercialize mass spectrometry equipment and software, and expert MS users for peptidomics and proteomics studies. The first round table (Geneva, 2001) entitled "Challenges in Mass Spectrometry" was supported by brief oral presentations that stressed critical questions in the field of MS development or applications (Stöcklin and Binz, Proteomics 2002, 2, 825-827). Topics such as (i) direct analysis of complex biological samples, (ii) status and perspectives for MS investigations of noncovalent peptide-ligant interactions; (iii) is it more appropriate to have complementary instruments rather than a universal equipment, (iv) standardization and improvement of the MS signals for protein identification, (v) what would be the new generation of equipment and finally (vi) how to keep hardware and software adapted to MS up-to-date and accessible to all. For the SPS'02 meeting (Lausanne, 2002), a full session alternative event "Proteomic Application Exercise" was proposed. Two different samples were prepared and sent to the different participants: 100 micro g of snake venom (a complex mixture of peptides and proteins) and 10-20 micro g of almost pure recombinant polypeptide derived from the shrimp Penaeus vannamei carrying an heterogeneous post-translational modification (PTM). Among the 15 participants that received the samples blind, eight returned results and most of them were asked to present their results emphasizing the strategy, the manpower and the instrumentation used during the congress (Binz et. al., Proteomics 2003, 3, 1562-1566). It appeared that for the snake venom extract, the quality of the results was not particularly dependant on the strategy used, as all approaches allowed Lication of identification of a certain number of protein families. The genus of the snake was identified in most cases, but the species was ambiguous. Surprisingly, the precise identification of the recombinant almost pure polypeptides appeared to be much more complicated than expected as only one group reported the full sequence. Finally the SPS'03 meeting reported here included a round table on the difficult and challenging task of "Quantification by Mass Spectrometry", a discussion sustained by four selected oral presentations on the use of stable isotopes, electrospray ionization versus matrix-assisted laser desorption/ionization approaches to quantify peptides and proteins in biological fluids, the handling of differential two-dimensional liquid chromatography tandem mass spectrometry data resulting from high throughput experiments, and the quantitative analysis of PTMs. During these three events at the SPS meetings, the impressive quality and quantity of exchanges between the developers and providers of mass spectrometry equipment and software, expert users and the audience, were a key element for the success of these fruitful events and will have definitively paved the way for future round tables and challenging exercises at SPS meetings.

Using Digital Video Analysis to Monitor Driver Behavior at Intersections

Commercially available instruments for road-side data collection take highly limited measurements, require extensive manual input, or are too expensive for widespread use. However, inexpensive computer vision techniques for digital video analysis can be applied to automate the monitoring of driver, vehicle, and pedestrian behaviors. These techniques can measure safety-related variables that cannot be easily measured using existing sensors. The use of these techniques will lead to an improved understanding of the decisions made by drivers at intersections. These automated techniques allow the collection of large amounts of safety-related data in a relatively short amount of time. There is a need to develop an easily deployable system to utilize these new techniques. This project implemented and tested a digital video analysis system for use at intersections. A prototype video recording system was developed for field deployment. A computer interface was implemented and served to simplify and automate the data analysis and the data review process. Driver behavior was measured at urban and rural non-signalized intersections. Recorded digital video was analyzed and used to test the system.

Machine learning for geospatial data : algorithms, software tools and case studies

Résumé Cette thèse est consacrée à l'analyse, la modélisation et la visualisation de données environnementales à référence spatiale à l'aide d'algorithmes d'apprentissage automatique (Machine Learning). L'apprentissage automatique peut être considéré au sens large comme une sous-catégorie de l'intelligence artificielle qui concerne particulièrement le développement de techniques et d'algorithmes permettant à une machine d'apprendre à partir de données. Dans cette thèse, les algorithmes d'apprentissage automatique sont adaptés pour être appliqués à des données environnementales et à la prédiction spatiale. Pourquoi l'apprentissage automatique ? Parce que la majorité des algorithmes d'apprentissage automatiques sont universels, adaptatifs, non-linéaires, robustes et efficaces pour la modélisation. Ils peuvent résoudre des problèmes de classification, de régression et de modélisation de densité de probabilités dans des espaces à haute dimension, composés de variables informatives spatialisées (« géo-features ») en plus des coordonnées géographiques. De plus, ils sont idéaux pour être implémentés en tant qu'outils d'aide à la décision pour des questions environnementales allant de la reconnaissance de pattern à la modélisation et la prédiction en passant par la cartographie automatique. Leur efficacité est comparable au modèles géostatistiques dans l'espace des coordonnées géographiques, mais ils sont indispensables pour des données à hautes dimensions incluant des géo-features. Les algorithmes d'apprentissage automatique les plus importants et les plus populaires sont présentés théoriquement et implémentés sous forme de logiciels pour les sciences environnementales. Les principaux algorithmes décrits sont le Perceptron multicouches (MultiLayer Perceptron, MLP) - l'algorithme le plus connu dans l'intelligence artificielle, le réseau de neurones de régression généralisée (General Regression Neural Networks, GRNN), le réseau de neurones probabiliste (Probabilistic Neural Networks, PNN), les cartes auto-organisées (SelfOrganized Maps, SOM), les modèles à mixture Gaussiennes (Gaussian Mixture Models, GMM), les réseaux à fonctions de base radiales (Radial Basis Functions Networks, RBF) et les réseaux à mixture de densité (Mixture Density Networks, MDN). Cette gamme d'algorithmes permet de couvrir des tâches variées telle que la classification, la régression ou l'estimation de densité de probabilité. L'analyse exploratoire des données (Exploratory Data Analysis, EDA) est le premier pas de toute analyse de données. Dans cette thèse les concepts d'analyse exploratoire de données spatiales (Exploratory Spatial Data Analysis, ESDA) sont traités selon l'approche traditionnelle de la géostatistique avec la variographie expérimentale et selon les principes de l'apprentissage automatique. La variographie expérimentale, qui étudie les relations entre pairs de points, est un outil de base pour l'analyse géostatistique de corrélations spatiales anisotropiques qui permet de détecter la présence de patterns spatiaux descriptible par une statistique. L'approche de l'apprentissage automatique pour l'ESDA est présentée à travers l'application de la méthode des k plus proches voisins qui est très simple et possède d'excellentes qualités d'interprétation et de visualisation. Une part importante de la thèse traite de sujets d'actualité comme la cartographie automatique de données spatiales. Le réseau de neurones de régression généralisée est proposé pour résoudre cette tâche efficacement. Les performances du GRNN sont démontrées par des données de Comparaison d'Interpolation Spatiale (SIC) de 2004 pour lesquelles le GRNN bat significativement toutes les autres méthodes, particulièrement lors de situations d'urgence. La thèse est composée de quatre chapitres : théorie, applications, outils logiciels et des exemples guidés. Une partie importante du travail consiste en une collection de logiciels : Machine Learning Office. Cette collection de logiciels a été développée durant les 15 dernières années et a été utilisée pour l'enseignement de nombreux cours, dont des workshops internationaux en Chine, France, Italie, Irlande et Suisse ainsi que dans des projets de recherche fondamentaux et appliqués. Les cas d'études considérés couvrent un vaste spectre de problèmes géoenvironnementaux réels à basse et haute dimensionnalité, tels que la pollution de l'air, du sol et de l'eau par des produits radioactifs et des métaux lourds, la classification de types de sols et d'unités hydrogéologiques, la cartographie des incertitudes pour l'aide à la décision et l'estimation de risques naturels (glissements de terrain, avalanches). Des outils complémentaires pour l'analyse exploratoire des données et la visualisation ont également été développés en prenant soin de créer une interface conviviale et facile à l'utilisation. Machine Learning for geospatial data: algorithms, software tools and case studies Abstract The thesis is devoted to the analysis, modeling and visualisation of spatial environmental data using machine learning algorithms. In a broad sense machine learning can be considered as a subfield of artificial intelligence. It mainly concerns with the development of techniques and algorithms that allow computers to learn from data. In this thesis machine learning algorithms are adapted to learn from spatial environmental data and to make spatial predictions. Why machine learning? In few words most of machine learning algorithms are universal, adaptive, nonlinear, robust and efficient modeling tools. They can find solutions for the classification, regression, and probability density modeling problems in high-dimensional geo-feature spaces, composed of geographical space and additional relevant spatially referenced features. They are well-suited to be implemented as predictive engines in decision support systems, for the purposes of environmental data mining including pattern recognition, modeling and predictions as well as automatic data mapping. They have competitive efficiency to the geostatistical models in low dimensional geographical spaces but are indispensable in high-dimensional geo-feature spaces. The most important and popular machine learning algorithms and models interesting for geo- and environmental sciences are presented in details: from theoretical description of the concepts to the software implementation. The main algorithms and models considered are the following: multi-layer perceptron (a workhorse of machine learning), general regression neural networks, probabilistic neural networks, self-organising (Kohonen) maps, Gaussian mixture models, radial basis functions networks, mixture density networks. This set of models covers machine learning tasks such as classification, regression, and density estimation. Exploratory data analysis (EDA) is initial and very important part of data analysis. In this thesis the concepts of exploratory spatial data analysis (ESDA) is considered using both traditional geostatistical approach such as_experimental variography and machine learning. Experimental variography is a basic tool for geostatistical analysis of anisotropic spatial correlations which helps to understand the presence of spatial patterns, at least described by two-point statistics. A machine learning approach for ESDA is presented by applying the k-nearest neighbors (k-NN) method which is simple and has very good interpretation and visualization properties. Important part of the thesis deals with a hot topic of nowadays, namely, an automatic mapping of geospatial data. General regression neural networks (GRNN) is proposed as efficient model to solve this task. Performance of the GRNN model is demonstrated on Spatial Interpolation Comparison (SIC) 2004 data where GRNN model significantly outperformed all other approaches, especially in case of emergency conditions. The thesis consists of four chapters and has the following structure: theory, applications, software tools, and how-to-do-it examples. An important part of the work is a collection of software tools - Machine Learning Office. Machine Learning Office tools were developed during last 15 years and was used both for many teaching courses, including international workshops in China, France, Italy, Ireland, Switzerland and for realizing fundamental and applied research projects. Case studies considered cover wide spectrum of the real-life low and high-dimensional geo- and environmental problems, such as air, soil and water pollution by radionuclides and heavy metals, soil types and hydro-geological units classification, decision-oriented mapping with uncertainties, natural hazards (landslides, avalanches) assessments and susceptibility mapping. Complementary tools useful for the exploratory data analysis and visualisation were developed as well. The software is user friendly and easy to use.

An epidemiological evaluation of pediatric long bone fractures - a retrospective cohort study of 2716 patients from two Swiss tertiary pediatric hospitals.

BACKGROUND: Children and adolescents are at high risk of sustaining fractures during growth. Therefore, epidemiological assessment is crucial for fracture prevention. The AO Comprehensive Injury Automatic Classifier (AO COIAC) was used to evaluate epidemiological data of pediatric long bone fractures in a large cohort. METHODS: Data from children and adolescents with long bone fractures sustained between 2009 and 2011, treated at either of two tertiary pediatric surgery hospitals in Switzerland, were retrospectively collected. Fractures were classified according to the AO Pediatric Comprehensive Classification of Long Bone Fractures (PCCF). RESULTS: For a total of 2716 patients (60% boys), 2807 accidents with 2840 long bone fractures (59% radius/ulna; 21% humerus; 15% tibia/fibula; 5% femur) were documented. Children's mean age (SD) was 8.2 (4.0) years (6% infants; 26% preschool children; 40% school children; 28% adolescents). Adolescent boys sustained more fractures than girls (p < 0.001). The leading cause of fractures was falls (27%), followed by accidents occurring during leisure activities (25%), at home (14%), on playgrounds (11%), and traffic (11%) and school accidents (8%). There was boy predominance for all accident types except for playground and at home accidents. The distribution of accident types differed according to age classes (p < 0.001). Twenty-six percent of patients were classed as overweight or obese - higher than data published by the WHO for the corresponding ages - with a higher proportion of overweight and obese boys than in the Swiss population (p < 0.0001). CONCLUSION: Overall, differences in the fracture distribution were sex and age related. Overweight and obese patients seemed to be at increased risk of sustaining fractures. Our data give valuable input into future development of prevention strategies. The AO PCCF proved to be useful in epidemiological reporting and analysis of pediatric long bone fractures.

Statistical significance of quantitative PCR.

BACKGROUND: PCR has the potential to detect and precisely quantify specific DNA sequences, but it is not yet often used as a fully quantitative method. A number of data collection and processing strategies have been described for the implementation of quantitative PCR. However, they can be experimentally cumbersome, their relative performances have not been evaluated systematically, and they often remain poorly validated statistically and/or experimentally. In this study, we evaluated the performance of known methods, and compared them with newly developed data processing strategies in terms of resolution, precision and robustness. RESULTS: Our results indicate that simple methods that do not rely on the estimation of the efficiency of the PCR amplification may provide reproducible and sensitive data, but that they do not quantify DNA with precision. Other evaluated methods based on sigmoidal or exponential curve fitting were generally of both poor resolution and precision. A statistical analysis of the parameters that influence efficiency indicated that it depends mostly on the selected amplicon and to a lesser extent on the particular biological sample analyzed. Thus, we devised various strategies based on individual or averaged efficiency values, which were used to assess the regulated expression of several genes in response to a growth factor. CONCLUSION: Overall, qPCR data analysis methods differ significantly in their performance, and this analysis identifies methods that provide DNA quantification estimates of high precision, robustness and reliability. These methods allow reliable estimations of relative expression ratio of two-fold or higher, and our analysis provides an estimation of the number of biological samples that have to be analyzed to achieve a given precision.