Développement, implémentation et utilisation pratique d'un score diagnostique [Development, implementation and practical use of a diagnostic score].

In recent years many clinical prediction rules (CPR) have been developed. Before a CPR can be used in clinical practice, different methodical steps are necessary, from the development of the score, the internal and external validation to the impact study. Before using a CPR in daily practice family doctors have to verify how the rules have been developed and whether this has been done in a population similar to the population in which they would use them. The aim of this paper is to describe the development of a CPR, and to discuss advantages and risks related to the use of CPR in order to help family doctors in their choice of scores for use in their daily practice.

Analysis, modelling and classification of geospatial data using machine learning

The research considers the problem of spatial data classification using machine learning algorithms: probabilistic neural networks (PNN) and support vector machines (SVM). As a benchmark model simple k-nearest neighbor algorithm is considered. PNN is a neural network reformulation of well known nonparametric principles of probability density modeling using kernel density estimator and Bayesian optimal or maximum a posteriori decision rules. PNN is well suited to problems where not only predictions but also quantification of accuracy and integration of prior information are necessary. An important property of PNN is that they can be easily used in decision support systems dealing with problems of automatic classification. Support vector machine is an implementation of the principles of statistical learning theory for the classification tasks. Recently they were successfully applied for different environmental topics: classification of soil types and hydro-geological units, optimization of monitoring networks, susceptibility mapping of natural hazards. In the present paper both simulated and real data case studies (low and high dimensional) are considered. The main attention is paid to the detection and learning of spatial patterns by the algorithms applied.

Comparative assessment of intimal hyperplasia development after 14 days in two different experimental settings: tissue culture versus ex vivo continuous perfusion of human saphenous vein.

BACKGROUND: Intimal hyperplasia (IH) is a vascular remodeling process which often leads to failure of arterial bypass or hemodialysis access. Experimental and clinical work have provided insight in IH development; however, further studies under precise controlled conditions are required to improve therapeutic strategies to inhibit IH development. Ex vivo perfusion of human vessel segments under standardized hemodynamic conditions may provide an adequate experimental approach for this purpose. Therefore, chronically perfused venous segments were studied and compared to traditional static culture procedures with regard to functional and histomorphologic characteristics as well as gene expression. MATERIALS AND METHODS: Static vein culture allowing high tissue viability was performed as previously described. Ex vivo vein support system (EVVSS) was performed using a vein support system consisting of an incubator with a perfusion chamber and a pump. EVVSS allows vessel perfusion under continuous flow while maintaining controlled hemodynamic conditions. Each human saphenous vein was divided in two parts, one cultured in a Pyrex dish and the other part perfused in EVVSS for 14days. Testing of vasomotion, histomorphometry, expression of CD 31, Factor VIII, MIB 1, alpha-actin, and PAI-l were determined before and after 14days of either experimental conditions. RESULTS: Human venous segments cultured under traditional or perfused conditions exhibited similar IH after 14 days as shown by histomorphometry. Smooth-muscle cell (SMC) was preserved after chronic perfusion. Although integrity of both endothelial and smooth-muscle cells appears to be maintained in both culture conditions as confirmed by CD31, factor VIII, and alpha-actin expression, a few smooth-muscle cells in the media stained positive for factor VIII. Cell-proliferation marker MIB-1 was also detected in the two settings and PAI-1 mRNA expression and activity increased significantly after 14 days of culture and perfusion. CONCLUSION: This study demonstrates the feasibility to chronically perfuse human vessels under sterile conditions with preservation of cellular integrity and vascular contractility. To gain insights into the mechanisms leading to IH, it will now be possible to study vascular remodeling not only under static conditions but also in hemodynamic environment mimicking as closely as possible the flow conditions encountered in reconstructive vascular surgery.

Role of hemodynamic forces in the ex vivo arterialization of human saphenous veins.

BACKGROUND: Human saphenous vein grafts are one of the salvage bypass conduits when endovascular procedures are not feasible or fail. Understanding the remodeling process that venous grafts undergo during exposure to arterial conditions is crucial to improve their patency, which is often compromised by intimal hyperplasia. The precise role of hemodynamic forces such as shear stress and arterial pressure in this remodeling is not fully characterized. The aim of this study was to determine the involvement of arterial shear stress and pressure on vein wall remodeling and to unravel the underlying molecular mechanisms. METHODS: An ex vivo vein support system was modified for chronic (up to 1 week), pulsatile perfusion of human saphenous veins under controlled conditions that permitted the separate control of arterial shear stress and different arterial pressure (7 mm Hg or 70 mm Hg). RESULTS: Veins perfused for 7 days under high pressure (70 mm Hg) underwent significant development of a neointima compared with veins exposed to low pressure (7 mm Hg). These structural changes were associated with altered expression of several molecular markers. Exposure to an arterial shear stress under low pressure increased the expression of matrix metalloproteinase (MMP)-2 and MMP-9 and tissue inhibitor of metalloproteinase (TIMP)-1 at the transcript, protein, and activity levels. This increase was enhanced by high pressure, which also increased TIMP-2 protein expression despite decreased levels of the cognate transcript. In contrast, the expression of plasminogen activator inhibitor-1 increased with shear stress but was not modified by pressure. Levels of the venous marker Eph-B4 were decreased under arterial shear stress, and levels of the arterial marker Ephrin-B2 were downregulated under high-pressure conditions. CONCLUSIONS: This model is a valuable tool to identify the role of hemodynamic forces and to decipher the molecular mechanisms leading to failure of human saphenous vein grafts. Under ex vivo conditions, arterial perfusion is sufficient to activate the remodeling of human veins, a change that is associated with the loss of specific vein markers. Elevation of pressure generates intimal hyperplasia, even though veins do not acquire arterial markers. CLINICAL RELEVANCE: The pathological remodeling of the venous wall, which leads to stenosis and ultimately graft failure, is the main limiting factor of human saphenous vein graft bypass. This remodeling is due to the hemodynamic adaptation of the vein to the arterial environment and cannot be prevented by conventional therapy. To develop a more targeted therapy, a better understanding of the molecular mechanisms involved in intimal hyperplasia is essential, which requires the development of ex vivo models of chronic perfusion of human veins.

Ex vivo Pulsatile Perfusion of Human Saphenous Veins Induces Intimal Hyperplasia and Increased Levels of the Plasminogen Activator Inhibitor 1.

Vessel wall trauma induces vascular remodeling processes including the development of intimal hyperplasia (IH). To assess the development of IH in human veins, we have used an ex vivo vein support system (EVVSS) allowing the perfusion of freshly isolated segments of saphenous veins in the presence of a pulsatile flow which reproduced arterial conditions regarding shear stress, flow rate and pressure during a period of 7 and 14 days. Compared to the corresponding freshly harvested human veins, histomorphometric analysis showed a significant increase in the intimal thickness which was already maximal after 7 days of perfusion. Expression of the endothelial marker CD31 demonstrated the presence of endothelium up to 14 days of perfusion. In our EVVSS model, the activity as well as the mRNA and protein expression levels of plasminogen activator inhibitor 1, the inhibitor of urokinase-type plasminogen activator (uPA) and tissue-type plasminogen activator (tPA), were increased after 7 days of perfusion, whereas the expression levels of tPA and uPA were not altered. No major change was observed between 7 and 14 days of perfusion. These data show that our newly developed EVVSS is a valuable setting to study ex vivo remodeling of human veins submitted to a pulsatile flow.

Ruling out coronary heart disease in primary care: external validation of a clinical prediction rule.

BACKGROUND: The Marburg Heart Score (MHS) aims to assist GPs in safely ruling out coronary heart disease (CHD) in patients presenting with chest pain, and to guide management decisions. AIM: To investigate the diagnostic accuracy of the MHS in an independent sample and to evaluate the generalisability to new patients. DESIGN AND SETTING: Cross-sectional diagnostic study with delayed-type reference standard in general practice in Hesse, Germany. METHOD: Fifty-six German GPs recruited 844 males and females aged ≥ 35 years, presenting between July 2009 and February 2010 with chest pain. Baseline data included the items of the MHS. Data on the subsequent course of chest pain, investigations, hospitalisations, and medication were collected over 6 months and were reviewed by an independent expert panel. CHD was the reference condition. Measures of diagnostic accuracy included the area under the receiver operating characteristic curve (AUC), sensitivity, specificity, likelihood ratios, and predictive values. RESULTS: The AUC was 0.84 (95% confidence interval [CI] = 0.80 to 0.88). For a cut-off value of 3, the MHS showed a sensitivity of 89.1% (95% CI = 81.1% to 94.0%), a specificity of 63.5% (95% CI = 60.0% to 66.9%), a positive predictive value of 23.3% (95% CI = 19.2% to 28.0%), and a negative predictive value of 97.9% (95% CI = 96.2% to 98.9%). CONCLUSION: Considering the diagnostic accuracy of the MHS, its generalisability, and ease of application, its use in clinical practice is recommended.

Outpatient prescription writing quality in a paediatric general hospital

Introduction The writing of prescriptions is an important aspect of medical practice. Since 2006, the Swiss authorities have decided to impose incentives to prescribe generic drugs. The objectives of this study were 1) to determine the evolution of the outpatient prescription practice in our paediatric university hospital during 2 periods separated by 5 years; 2) to assess the writing quality of outpatient prescriptions during the same period.Materials & Methods Design: Copies of prescriptions written by physicians were collected twice from community pharmacies in the region of our hospital for a 2-month period in 2005 and 2010. They were analysed according to standard criteria regarding both formal and pharmaceutical aspects. Drug prescriptions were classified as a) complete when all criteria for safety were fulfilled, b) ambiguous when there was a danger of a dispensing error because of one or more missing criteria, or c) containing an error.Setting: Paediatric university hospital.Main outcome measures: Proportion of generic drugs; outpatient prescription writing quality.Results: A total of 651 handwritten prescriptions were reviewed in 2005 and 693 in 2010. They contained 1570 drug prescriptions in 2005 (2.4 ± 1.2 drugs per patient) and 1462 in 2010 (2.1 ± 1.1). The most common drugs were paracetamol, ibuprofen, and sodium chloride. A higher proportion of drugs were prescribed as generic names or generics in 2010. Formal data regarding the physicians and the patients were almost complete, except for the patients' weight. Of the drug prescriptions, 48.5% were incomplete, 11.3% were ambiguous, and 3.0% contained an error in 2005. These proportions rose to 64.2%, 15.5% and 7.4% in 2010, respectively.Discussions, Conclusion This study showed that physicians' prescriptions comprised numerous omissions and errors with minimal potential for harm. Computerized prescription coupled with advanced decision support is eagerly awaited.Disclosure of Interest None Declared

Low-cost microprocessed instrument for evaluating soil temperature profile

This paper describes a low-cost microprocessed instrument for in situ evaluating soil temperature profile ranging from -20.0°C to 99.9°C, and recording soil temperature data at eight depths from 2 to 128 cm. Of great importance in agriculture, soil temperature affects plant growth directly, and nutrient uptake as well as indirectly in soil water and gas flow, soil structure and nutrient availability. The developed instrument has potential applications in the soil science, when temperature monitoring is required. Results show that the instrument with its individual sensors guarantees ±0.25°C accuracy and 0.1°C resolution, making possible localized management changes within decision support systems. The instrument, based on complementary metal oxide semiconductor devices as well as thermocouples, operates in either automatic or non-automatic mode.

Predictive analysis and mapping of indoor radon concentrations in Switzerland

It is estimated that around 230 people die each year due to radon (222Rn) exposure in Switzerland. 222Rn occurs mainly in closed environments like buildings and originates primarily from the subjacent ground. Therefore it depends strongly on geology and shows substantial regional variations. Correct identification of these regional variations would lead to substantial reduction of 222Rn exposure of the population based on appropriate construction of new and mitigation of already existing buildings. Prediction of indoor 222Rn concentrations (IRC) and identification of 222Rn prone areas is however difficult since IRC depend on a variety of different variables like building characteristics, meteorology, geology and anthropogenic factors. The present work aims at the development of predictive models and the understanding of IRC in Switzerland, taking into account a maximum of information in order to minimize the prediction uncertainty. The predictive maps will be used as a decision-support tool for 222Rn risk management. The construction of these models is based on different data-driven statistical methods, in combination with geographical information systems (GIS). In a first phase we performed univariate analysis of IRC for different variables, namely the detector type, building category, foundation, year of construction, the average outdoor temperature during measurement, altitude and lithology. All variables showed significant associations to IRC. Buildings constructed after 1900 showed significantly lower IRC compared to earlier constructions. We observed a further drop of IRC after 1970. In addition to that, we found an association of IRC with altitude. With regard to lithology, we observed the lowest IRC in sedimentary rocks (excluding carbonates) and sediments and the highest IRC in the Jura carbonates and igneous rock. The IRC data was systematically analyzed for potential bias due to spatially unbalanced sampling of measurements. In order to facilitate the modeling and the interpretation of the influence of geology on IRC, we developed an algorithm based on k-medoids clustering which permits to define coherent geological classes in terms of IRC. We performed a soil gas 222Rn concentration (SRC) measurement campaign in order to determine the predictive power of SRC with respect to IRC. We found that the use of SRC is limited for IRC prediction. The second part of the project was dedicated to predictive mapping of IRC using models which take into account the multidimensionality of the process of 222Rn entry into buildings. We used kernel regression and ensemble regression tree for this purpose. We could explain up to 33% of the variance of the log transformed IRC all over Switzerland. This is a good performance compared to former attempts of IRC modeling in Switzerland. As predictor variables we considered geographical coordinates, altitude, outdoor temperature, building type, foundation, year of construction and detector type. Ensemble regression trees like random forests allow to determine the role of each IRC predictor in a multidimensional setting. We found spatial information like geology, altitude and coordinates to have stronger influences on IRC than building related variables like foundation type, building type and year of construction. Based on kernel estimation we developed an approach to determine the local probability of IRC to exceed 300 Bq/m3. In addition to that we developed a confidence index in order to provide an estimate of uncertainty of the map. All methods allow an easy creation of tailor-made maps for different building characteristics. Our work is an essential step towards a 222Rn risk assessment which accounts at the same time for different architectural situations as well as geological and geographical conditions. For the communication of 222Rn hazard to the population we recommend to make use of the probability map based on kernel estimation. The communication of 222Rn hazard could for example be implemented via a web interface where the users specify the characteristics and coordinates of their home in order to obtain the probability to be above a given IRC with a corresponding index of confidence. Taking into account the health effects of 222Rn, our results have the potential to substantially improve the estimation of the effective dose from 222Rn delivered to the Swiss population.

Comparative assessement of intimal hyperplasia development after 14 days in two different experimental settings : tissue culture versus ex vivo continuous perfusion of human saphenous vein

Résumé de l'article : L'hyperplasie intimale est un processus de remodelage vasculaire ubiquitaire après une lésion, pouvant menacer la perméabilité de tout type de reconstruction vasculaire. Les mécanismes physiopathologiques impliqués dans le développement de l'hyperplasie intimale ne sont que partiellement élucidés. Il est par conséquent nécessaire d'effectuer des recherches complémentaires afin d'en améliorer la compréhension et ainsi permettre l'élaboration de nouvelles stratégies thérapeutiques médicamenteuses. La culture de veines en milieu statique permet le développement de l'hyperplasie intimale. Ce modèle maintient la viabilité tissulaire, comme décrit précédemment dans d'autres études, mais empêche l'analyse des paramètres hémodynamiques. La mise au point d'un modèle de perfusion in vitro permettant la perfusion de segments vasculaires représente une approche expérimentale intégrant les différents facteurs hémodynamiques. Le système de perfusion (Ex Vivo Vein Support System) que nous avons élaboré conserve l'intégrité pariétale ainsi que les propriétés vasomotrices des veines pour une durée de 14 jours. Cette étude démontre que les deux modèles permettent le développement de l'hyperplasie intimale. Toutefois, les propriétés vasomotrices ainsi que l'influence des paramètres hémodynamiques ne peuvent être analysées que par l'utilisation du système de perfusion. Ce dernier a permis de perfuser des vaisseaux humains sans contamination bactérienne tout en maintenant l'intégrité cellulaire. Ce modèle de perfusion se rapproche plus des conditions hémodynamiques rencontrées in vivo que le modèle statique. Abstract : Background. Intimal hyperplasia (IH) is a vascular remodeling process which often leads to failure of arterial bypass or hemodialysis access. Experimental and clinical work have provided insight in IH development; however, further studies under precise con-trolled conditions are required to improve therapeutic strategies to inhibit IH development. Ex vivo perfusion of human vessel segments under standardized hemodynamic conditions may provide an adequate experimental approach for this purpose. Therefore, chronically perfused venous segments were studied and compared to traditional static culture procedures with regard to functional and histomorphologic characteristics as well as gene expression. Materials and methods. Static vein culture allowing high tissue viability was performed as previously described. Ex vivo vein support system (EVVSS) was performed using a vein support system consisting of an incubator with a perfusion chamber and a pump. EVVSS allows vessel perfusion under continuous flow while maintaining controlled hemodynamic conditions. Each human saphenous vein was divided in two parts, one cultured in a Pyrex dish and the other part perfused in EVVSS for 14 days. Testing of vasomotion, histomorphometry, expression of CD 31, Factor VIII, MIB 1, α-actin, and PAI-1 were determined before and after 14 days of either experimental conditions. Results, Human venous segments cultured under traditional or perfused conditions exhibited similar IH after 14 days as shown by histomorphometry. Smooth-muscle cell ( SMC) was preserved after chronic perfusion. Although integrity of both endothelial and smooth-muscle cells appears to be maintained in both culture conditions as confirmed by CD31, factor VIII and α-actin expression, a few smooth-muscle cells in the media stained positive for factor VIII. Cell-proliferation marker MIB-1 was also detected in the two settings and PAI-1 mRNA expression and activity increased significantly after 14 days of culture and perfusion. Conclusion. This study demonstrates the feasibility to chronically perfuse human vessels under sterile conditions with preservation of cellular integrity and vascular contractility. To gain insights into the mechanisms leading to IH, it will now be possible to study vascular remodeling not only under static conditions but also in hemodynamic environment mimicking as closely as possible the flow conditions encountered in reconstructive vascular surgery.

Data-driven topo-climatic mapping with machine learning methods

Automatic environmental monitoring networks enforced by wireless communication technologies provide large and ever increasing volumes of data nowadays. The use of this information in natural hazard research is an important issue. Particularly useful for risk assessment and decision making are the spatial maps of hazard-related parameters produced from point observations and available auxiliary information. The purpose of this article is to present and explore the appropriate tools to process large amounts of available data and produce predictions at fine spatial scales. These are the algorithms of machine learning, which are aimed at non-parametric robust modelling of non-linear dependencies from empirical data. The computational efficiency of the data-driven methods allows producing the prediction maps in real time which makes them superior to physical models for the operational use in risk assessment and mitigation. Particularly, this situation encounters in spatial prediction of climatic variables (topo-climatic mapping). In complex topographies of the mountainous regions, the meteorological processes are highly influenced by the relief. The article shows how these relations, possibly regionalized and non-linear, can be modelled from data using the information from digital elevation models. The particular illustration of the developed methodology concerns the mapping of temperatures (including the situations of Föhn and temperature inversion) given the measurements taken from the Swiss meteorological monitoring network. The range of the methods used in the study includes data-driven feature selection, support vector algorithms and artificial neural networks.

Randomized trial of a computerized coronary heart disease risk assessment tool in HIV-infected patients receiving combination antiretroviral therapy.

BACKGROUND: Exposure to combination antiretroviral therapy (cART) can lead to important metabolic changes and increased risk of coronary heart disease (CHD). Computerized clinical decision support systems have been advocated to improve the management of patients at risk for CHD but it is unclear whether such systems reduce patients' risk for CHD. METHODS: We conducted a cluster trial within the Swiss HIV Cohort Study (SHCS) of HIV-infected patients, aged 18 years or older, not pregnant and receiving cART for >3 months. We randomized 165 physicians to either guidelines for CHD risk factor management alone or guidelines plus CHD risk profiles. Risk profiles included the Framingham risk score, CHD drug prescriptions and CHD events based on biannual assessments, and were continuously updated by the SHCS data centre and integrated into patient charts by study nurses. Outcome measures were total cholesterol, systolic and diastolic blood pressure and Framingham risk score. RESULTS: A total of 3,266 patients (80% of those eligible) had a final assessment of the primary outcome at least 12 months after the start of the trial. Mean (95% confidence interval) patient differences where physicians received CHD risk profiles and guidelines, rather than guidelines alone, were total cholesterol -0.02 mmol/l (-0.09-0.06), systolic blood pressure -0.4 mmHg (-1.6-0.8), diastolic blood pressure -0.4 mmHg (-1.5-0.7) and Framingham 10-year risk score -0.2% (-0.5-0.1). CONCLUSIONS: Systemic computerized routine provision of CHD risk profiles in addition to guidelines does not significantly improve risk factors for CHD in patients on cART.

Derivation and validation of a prognostic model for pulmonary embolism.

RATIONALE: An objective and simple prognostic model for patients with pulmonary embolism could be helpful in guiding initial intensity of treatment. OBJECTIVES: To develop a clinical prediction rule that accurately classifies patients with pulmonary embolism into categories of increasing risk of mortality and other adverse medical outcomes. METHODS: We randomly allocated 15,531 inpatient discharges with pulmonary embolism from 186 Pennsylvania hospitals to derivation (67%) and internal validation (33%) samples. We derived our prediction rule using logistic regression with 30-day mortality as the primary outcome, and patient demographic and clinical data routinely available at presentation as potential predictor variables. We externally validated the rule in 221 inpatients with pulmonary embolism from Switzerland and France. MEASUREMENTS: We compared mortality and nonfatal adverse medical outcomes across the derivation and two validation samples. MAIN RESULTS: The prediction rule is based on 11 simple patient characteristics that were independently associated with mortality and stratifies patients with pulmonary embolism into five severity classes, with 30-day mortality rates of 0-1.6% in class I, 1.7-3.5% in class II, 3.2-7.1% in class III, 4.0-11.4% in class IV, and 10.0-24.5% in class V across the derivation and validation samples. Inpatient death and nonfatal complications were <or= 1.1% among patients in class I and <or= 1.9% among patients in class II. CONCLUSIONS: Our rule accurately classifies patients with pulmonary embolism into classes of increasing risk of mortality and other adverse medical outcomes. Further validation of the rule is important before its implementation as a decision aid to guide the initial management of patients with pulmonary embolism.

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.