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.

Impact of the Advisa MRI pacing system on the diagnostic quality of cardiac MR images and contraction patterns of cardiac muscle during scans: Advisa MRI randomized clinical multicenter study results.

BACKGROUND: The Advisa MRI system is designed to safely undergo magnetic resonance imaging (MRI). Its influence on image quality is not well known. OBJECTIVE: To evaluate cardiac magnetic resonance (CMR) image quality and to characterize myocardial contraction patterns by using the Advisa MRI system. METHODS: In this international trial with 35 participating centers, an Advisa MRI system was implanted in 263 patients. Of those, 177 were randomized to the MRI group and 150 underwent MRI scans at the 9-12-week visit. Left ventricular (LV) and right ventricular (RV) cine long-axis steady-state free precession MR images were graded for quality. Signal loss along the implantable pulse generator and leads was measured. The tagging CMR data quality was assessed as the percentage of trackable tagging points on complementary spatial modulation of magnetization acquisitions (n=16) and segmental circumferential fiber shortening was quantified. RESULTS: Of all cine long-axis steady-state free precession acquisitions, 95% of LV and 98% of RV acquisitions were of diagnostic quality, with 84% and 93%, respectively, being of good or excellent quality. Tagging points were trackable from systole into early diastole (360-648 ms after the R-wave) in all segments. During RV pacing, tagging demonstrated a dyssynchronous contraction pattern, which was not observed in nonpaced (n = 4) and right atrial-paced (n = 8) patients. CONCLUSIONS: In the Advisa MRI study, high-quality CMR images for the assessment of cardiac anatomy and function were obtained in most patients with an implantable pacing system. In addition, this study demonstrated the feasibility of acquiring tagging data to study the LV function during pacing.

Total hip replacement with a collarless polished cemented anatomic stem: clinical and gait analysis results at ten years follow-up.

PURPOSE: The aim of this study was to determine outcomes of total hip replacement (THR) with the Lemania cemented femoral stem. METHODS: A total of 78 THR patients were followed and compared to 17 "fit", healthy, elderly and 72 "frail" elderly subjects without THR, using clinical outcome measures and a portable, in-field gait analysis device at five and ten years follow-up. RESULTS: Forty-one patients (53%), mean age 83.4 years, available at ten years follow-up, reported very good to excellent satisfaction. Mean Harris Hip and Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) scores were 81.2 and 10.5 points, respectively, with excellent radiological preservation of proximal femur bone stock. Spatial and temporal gait parameters were close to the fit group and better than the frail group. CONCLUSIONS: Lemania THR demonstrated very good, stable clinical and radiological results at ten years in an older patient group, comparable to other cemented systems for primary THR. Gait analysis confirmed good walking performance in a real-life environment.

Spatial Scale of Clustering of Motor Vehicle Crash Types and Appropriate Countermeasures, 2009

This project analyzes the characteristics and spatial distributions of motor vehicle crash types in order to evaluate the degree and scale of their spatial clustering. Crashes occur as the result of a variety of vehicle, roadway, and human factors and thus vary in their clustering behavior. Clustering can occur at a variety of scales, from the intersection level, to the corridor level, to the area level. Conversely, other crash types are less linked to geographic factors and are more spatially “random.” The degree and scale of clustering have implications for the use of strategies to promote transportation safety. In this project, Iowa's crash database, geographic information systems, and recent advances in spatial statistics methodologies and software tools were used to analyze the degree and spatial scale of clustering for several crash types within the counties of the Iowa Northland Regional Council of Governments. A statistical measure called the K function was used to analyze the clustering behavior of crashes. Several methodological issues, related to the application of this spatial statistical technique in the context of motor vehicle crashes on a road network, were identified and addressed. These methods facilitated the identification of crash clusters at appropriate scales of analysis for each crash type. This clustering information is useful for improving transportation safety through focused countermeasures directly linked to crash causes and the spatial extent of identified problem locations, as well as through the identification of less location-based crash types better suited to non-spatial countermeasures. The results of the K function analysis point to the usefulness of the procedure in identifying the degree and scale at which crashes cluster, or do not cluster, relative to each other. Moreover, for many individual crash types, different patterns and processes and potentially different countermeasures appeared at different scales of analysis. This finding highlights the importance of scale considerations in problem identification and countermeasure formulation.

Nutrient-based dietary patterns and the risk of head and neck cancer: a pooled analysis in the International Head and Neck Cancer Epidemiology consortium.

Background The association between dietary patterns and head and neck cancer has rarely been addressed. Patients and methods We used individual-level pooled data from five case-control studies (2452 cases and 5013 controls) participating in the International Head and Neck Cancer Epidemiology consortium. A posteriori dietary patterns were identified through a principal component factor analysis carried out on 24 nutrients derived from study-specific food-frequency questionnaires. Odds ratios (ORs) and corresponding 95% confidence intervals (CIs) were estimated using unconditional logistic regression models on quintiles of factor scores. Results We identified three major dietary patterns named 'animal products and cereals', 'antioxidant vitamins and fiber', and 'fats'. The 'antioxidant vitamins and fiber' pattern was inversely related to oral and pharyngeal cancer (OR = 0.57, 95% CI 0.43-0.76 for the highest versus the lowest score quintile). The 'animal products and cereals' pattern was positively associated with laryngeal cancer (OR = 1.54, 95% CI 1.12-2.11), whereas the 'fats' pattern was inversely associated with oral and pharyngeal cancer (OR = 0.78, 95% CI 0.63-0.97) and positively associated with laryngeal cancer (OR = 1.69, 95% CI 1.22-2.34). Conclusions These findings suggest that diets rich in animal products, cereals, and fats are positively related to laryngeal cancer, and those rich in fruit and vegetables inversely related to oral and pharyngeal cancer.

The odd couple: contrasting phylogeographic patterns in two sympatric sibling species of woodlouse-­‐hunter spiders in the Canary Islands

Comparative phylogeography seeks for commonalities in the spatial demographic history of sympatric organisms to characterize the mechanisms that shaped such patterns. The unveiling of incongruent phylogeographic patterns in co-occurring species, on the other hand, may hint to overlooked differences in their life histories or microhabitat preferences. The woodlouse-hunter spiders of the genus Dysdera have undergone a major diversi cation on the Canary Islands. The species pair Dysdera alegranzaensis and Dysdera nesiotes are endemic to the island of Lanzarote and nearby islets, where they co-occur at most of their known localities. The two species stand in sharp contrast to other sympatric endemic Dysdera in showing no evidence of somatic (non-genitalic) differentiation. Phylogenetic and population genetic analyses of mitochondrial cox1 sequences from an exhaustive sample of D. alegranzaensis and D. nesiotes specimens, and additional mitochondrial (16S, L1, nad1) and nuclear genes (28S, H3) were analysed to reveal their phylogeographic patterns and clarify their phylogenetic relationships. Relaxed molecular clock models using ve calibration points were further used to estimate divergence times between species and populations. Striking differences in phylogeography and population structure between the two species were observed. Dysdera nesiotes displayed a metapopulation-like structure, while D. alegranzaensis was characterized by a weaker geographical structure but greater genetic divergences among its main haplotype lineages, suggesting more complex population dynamics. Our study con rms that co-distributed sibling species may exhibit contrasting phylogeographic patterns in the absence of somatic differentiation. Further ecological studies, however, will be necessary to clarify whether the contrasting phylogeographies may hint at an overlooked niche partitioning between the two species. In addition, further comparisons with available phylogeographic data of other eastern Canarian Dysdera endemics con rm the key role of lava ows in structuring local populations in oceanic islands and identify localities that acted as refugia during volcanic eruptions

Analysis of a finite volume method for a cross-diffusion model in population dynamics

The main goal of this paper is to propose a convergent finite volume method for a reactionâeuro"diffusion system with cross-diffusion. First, we sketch an existence proof for a class of cross-diffusion systems. Then the standard two-point finite volume fluxes are used in combination with a nonlinear positivity-preserving approximation of the cross-diffusion coefficients. Existence and uniqueness of the approximate solution are addressed, and it is also shown that the scheme converges to the corresponding weak solution for the studied model. Furthermore, we provide a stability analysis to study pattern-formation phenomena, and we perform two-dimensional numerical examples which exhibit formation of nonuniform spatial patterns. From the simulations it is also found that experimental rates of convergence are slightly below second order. The convergence proof uses two ingredients of interest for various applications, namely the discrete Sobolev embedding inequalities with general boundary conditions and a space-time $L^1$ compactness argument that mimics the compactness lemma due to Kruzhkov. The proofs of these results are given in the Appendix.

Self-Organising Maps and Correlation Analysis as a Tool to Explore Patterns in Excitation-Emission Matrix Data Sets and to Discriminate Dissolved Organic Matter Fluorescence Components

Dissolved organic matter (DOM) is a complex mixture of organic compounds, ubiquitous in marine and freshwater systems. Fluorescence spectroscopy, by means of Excitation-Emission Matrices (EEM), has become an indispensable tool to study DOM sources, transport and fate in aquatic ecosystems. However the statistical treatment of large and heterogeneous EEM data sets still represents an important challenge for biogeochemists. Recently, Self-Organising Maps (SOM) has been proposed as a tool to explore patterns in large EEM data sets. SOM is a pattern recognition method which clusterizes and reduces the dimensionality of input EEMs without relying on any assumption about the data structure. In this paper, we show how SOM, coupled with a correlation analysis of the component planes, can be used both to explore patterns among samples, as well as to identify individual fluorescence components. We analysed a large and heterogeneous EEM data set, including samples from a river catchment collected under a range of hydrological conditions, along a 60-km downstream gradient, and under the influence of different degrees of anthropogenic impact. According to our results, chemical industry effluents appeared to have unique and distinctive spectral characteristics. On the other hand, river samples collected under flash flood conditions showed homogeneous EEM shapes. The correlation analysis of the component planes suggested the presence of four fluorescence components, consistent with DOM components previously described in the literature. A remarkable strength of this methodology was that outlier samples appeared naturally integrated in the analysis. We conclude that SOM coupled with a correlation analysis procedure is a promising tool for studying large and heterogeneous EEM data sets.

Force Spectroscopy with Dual-Trap Optical Tweezers: Molecular Stiffness Measurements and Coupled Fluctuations Analysis

ABSTRACT Dual-trap optical tweezers are often used in high-resolution measurements in single-molecule biophysics. Such measurements can be hindered by the presence of extraneous noise sources, the most prominent of which is the coupling of fluctuations along different spatial directions, which may affect any optical tweezers setup. In this article, we analyze, both from the theoretical and the experimental points of view, the most common source for these couplings in dual-trap optical-tweezers setups: the misalignment of traps and tether. We give criteria to distinguish different kinds of misalignment, to estimate their quantitative relevance and to include them in the data analysis. The experimental data is obtained in a, to our knowledge, novel dual-trap optical-tweezers setup that directly measures forces. In the case in which misalignment is negligible, we provide a method to measure the stiffness of traps and tether based on variance analysis. This method can be seen as a calibration technique valid beyond the linear trap region. Our analysis is then employed to measure the persistence length of dsDNA tethers of three different lengths spanning two orders of magnitude. The effective persistence length of such tethers is shown to decrease with the contour length, in accordance with previous studies.

Self-Organising Maps and Correlation Analysis as a Tool to Explore Patterns in Excitation-Emission Matrix Data Sets and to Discriminate Dissolved Organic Matter Fluorescence Components

Dissolved organic matter (DOM) is a complex mixture of organic compounds, ubiquitous in marine and freshwater systems. Fluorescence spectroscopy, by means of Excitation-Emission Matrices (EEM), has become an indispensable tool to study DOM sources, transport and fate in aquatic ecosystems. However the statistical treatment of large and heterogeneous EEM data sets still represents an important challenge for biogeochemists. Recently, Self-Organising Maps (SOM) has been proposed as a tool to explore patterns in large EEM data sets. SOM is a pattern recognition method which clusterizes and reduces the dimensionality of input EEMs without relying on any assumption about the data structure. In this paper, we show how SOM, coupled with a correlation analysis of the component planes, can be used both to explore patterns among samples, as well as to identify individual fluorescence components. We analysed a large and heterogeneous EEM data set, including samples from a river catchment collected under a range of hydrological conditions, along a 60-km downstream gradient, and under the influence of different degrees of anthropogenic impact. According to our results, chemical industry effluents appeared to have unique and distinctive spectral characteristics. On the other hand, river samples collected under flash flood conditions showed homogeneous EEM shapes. The correlation analysis of the component planes suggested the presence of four fluorescence components, consistent with DOM components previously described in the literature. A remarkable strength of this methodology was that outlier samples appeared naturally integrated in the analysis. We conclude that SOM coupled with a correlation analysis procedure is a promising tool for studying large and heterogeneous EEM data sets.

The odd couple: contrasting phylogeographic patterns in two sympatric sibling species of woodlouse-­‐hunter spiders in the Canary Islands

Comparative phylogeography seeks for commonalities in the spatial demographic history of sympatric organisms to characterize the mechanisms that shaped such patterns. The unveiling of incongruent phylogeographic patterns in co-occurring species, on the other hand, may hint to overlooked differences in their life histories or microhabitat preferences. The woodlouse-hunter spiders of the genus Dysdera have undergone a major diversi cation on the Canary Islands. The species pair Dysdera alegranzaensis and Dysdera nesiotes are endemic to the island of Lanzarote and nearby islets, where they co-occur at most of their known localities. The two species stand in sharp contrast to other sympatric endemic Dysdera in showing no evidence of somatic (non-genitalic) differentiation. Phylogenetic and population genetic analyses of mitochondrial cox1 sequences from an exhaustive sample of D. alegranzaensis and D. nesiotes specimens, and additional mitochondrial (16S, L1, nad1) and nuclear genes (28S, H3) were analysed to reveal their phylogeographic patterns and clarify their phylogenetic relationships. Relaxed molecular clock models using ve calibration points were further used to estimate divergence times between species and populations. Striking differences in phylogeography and population structure between the two species were observed. Dysdera nesiotes displayed a metapopulation-like structure, while D. alegranzaensis was characterized by a weaker geographical structure but greater genetic divergences among its main haplotype lineages, suggesting more complex population dynamics. Our study con rms that co-distributed sibling species may exhibit contrasting phylogeographic patterns in the absence of somatic differentiation. Further ecological studies, however, will be necessary to clarify whether the contrasting phylogeographies may hint at an overlooked niche partitioning between the two species. In addition, further comparisons with available phylogeographic data of other eastern Canarian Dysdera endemics con rm the key role of lava ows in structuring local populations in oceanic islands and identify localities that acted as refugia during volcanic eruptions

Spatial and Temporal Trends in Sunshine Duration over Western Europe

This work analyzes sunshine duration variability in the western part of Europe (WEU) over the 1938– 2004 period. A principal component analysis is applied to cluster the original series from 79 sites into 6 regions, and then annual and seasonal mean series are constructed on regional and also for the whole WEU scales. Over the entire period studied here, the linear trend of annual sunshine duration is found to be nonsignificant. However, annual sunshine duration shows an overall decrease since the 1950s until the early 1980s, followed by a subsequent recovery during the last two decades. This behavior is in good agreement with the dimming and brightening phenomena described in previous literature. From the seasonal analysis, the most remarkable result is the similarity between spring and annual series, although the spring series has a negative trend; and the clear significant increase found for the whole WEU winter series, being especially large since the 1970s. The behavior of the major synoptic patterns for two seasons is investigated, resulting in some indications that sunshine duration evolution may be partially explained by changes in the frequency of some of them

LSB Matching Steganalysis Based on Patterns of Pixel Differences and Random Embedding

Peer-reviewed

Un análisis comparativo del papel del bucle fonológico versus la agenda visoespacial en el cálculo en niños de 7-8 años = A comparative analysis of the phonological loop versus the visuo-spatial sketchpad in mental arithmetic tasks in 7-8 y.o. children

En esta investigación se ha estudiado la relación entre dos subsistemas de la memoria de trabajo (buclefonológico y agenda viso-espacial) y el rendimiento en cálculo con una muestra de 94 niños españolesde 7-8 años. Hemos administrado dos pruebas de cálculo diseñadas para este estudio y seis medidassimples de memoria de trabajo (de contenido verbal, numérico y espacial) de la «Batería de Testsde Memoria de Treball» de Pickering, Baqués y Gathercole (1999), y dos pruebas visuales complementarias.Los resultados muestran una correlación importante entre las medidas de contenido verbaly numérico y el rendimiento en cálculo. En cambio, no hemos encontrado ninguna relación con las medidasespaciales. Se concluye, por lo tanto, que en escolares españoles existe una relación importanteentre el bucle fonológico y el rendimiento en tareas de cálculo. En cambio, el rol de la agenda viso-espaciales nulo