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.

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.

Predictive analysis and mapping of indoor radon concentrations in a complex environment using kernel estimation: an application to Switzerland.

PURPOSE: The aim of this study was to develop models based on kernel regression and probability estimation in order to predict and map IRC in Switzerland by taking into account all of the following: architectural factors, spatial relationships between the measurements, as well as geological information. METHODS: We looked at about 240,000 IRC measurements carried out in about 150,000 houses. As predictor variables we included: building type, foundation type, year of construction, detector type, geographical coordinates, altitude, temperature and lithology into the kernel estimation models. We developed predictive maps as well as a map of the local probability to exceed 300 Bq/m(3). Additionally, we developed a map of a confidence index in order to estimate the reliability of the probability map. RESULTS: Our models were able to explain 28% of the variations of IRC data. All variables added information to the model. The model estimation revealed a bandwidth for each variable, making it possible to characterize the influence of each variable on the IRC estimation. Furthermore, we assessed the mapping characteristics of kernel estimation overall as well as by municipality. Overall, our model reproduces spatial IRC patterns which were already obtained earlier. On the municipal level, we could show that our model accounts well for IRC trends within municipal boundaries. Finally, we found that different building characteristics result in different IRC maps. Maps corresponding to detached houses with concrete foundations indicate systematically smaller IRC than maps corresponding to farms with earth foundation. CONCLUSIONS: IRC mapping based on kernel estimation is a powerful tool to predict and analyze IRC on a large-scale as well as on a local level. This approach enables to develop tailor-made maps for different architectural elements and measurement conditions and to account at the same time for geological information and spatial relations between IRC measurements.

Mapping the human connectome at multiple scales with diffusion spectrum MRI.

The global structural connectivity of the brain, the human connectome, is now accessible at millimeter scale with the use of MRI. In this paper, we describe an approach to map the connectome by constructing normalized whole-brain structural connection matrices derived from diffusion MRI tractography at 5 different scales. Using a template-based approach to match cortical landmarks of different subjects, we propose a robust method that allows (a) the selection of identical cortical regions of interest of desired size and location in different subjects with identification of the associated fiber tracts (b) straightforward construction and interpretation of anatomically organized whole-brain connection matrices and (c) statistical inter-subject comparison of brain connectivity at various scales. The fully automated post-processing steps necessary to build such matrices are detailed in this paper. Extensive validation tests are performed to assess the reproducibility of the method in a group of 5 healthy subjects and its reliability is as well considerably discussed in a group of 20 healthy subjects.

Automated Lesion Detection-Based Symptome Mapping In Patients With Sensory Deficits

The human brain is the most complex structure known. With its high number of cells, number of connections and number of pathways it is the source of every thought in the world. It consumes 25% of our oxygen and suffers very fast from a disruption of its supply. An acute event, like a stroke, results in rapid dysfunction referable to the affected area. A few minutes without oxygen and neuronal cells die and subsequently degenerate. Changes in the brains incoming blood flow alternate the anatomy and physiology of the brain. All stroke events leave behind a brain tissue lesion. To rapidly react and improve the prediction of outcome in stroke patients, accurate lesion detection and reliable lesion-based function correlation would be very helpful. With a number of neuroimaging and clinical data of cerebral injured patients this study aims to investigate correlations of structural lesion locations with sensory functions.

Feasibility of direct mapping of cerebral fluorodeoxy-D-glucose metabolism in situ at subcellular resolution using soft X-ray fluorescence.

Glucose metabolism is difficult to image with cellular resolution in mammalian brain tissue, particularly with (18) fluorodeoxy-D-glucose (FDG) positron emission tomography (PET). To this end, we explored the potential of synchrotron-based low-energy X-ray fluorescence (LEXRF) to image the stable isotope of fluorine (F) in phosphorylated FDG (DG-6P) at 1 μm(2) spatial resolution in 3-μm-thick brain slices. The excitation-dependent fluorescence F signal at 676 eV varied linearly with FDG concentration between 0.5 and 10 mM, whereas the endogenous background F signal was undetectable in brain. To validate LEXRF mapping of fluorine, FDG was administered in vitro and in vivo, and the fluorine LEXRF signal from intracellular trapped FDG-6P over selected brain areas rich in radial glia was spectrally quantitated at 1 μm(2) resolution. The subsequent generation of spatial LEXRF maps of F reproduced the expected localization and gradients of glucose metabolism in retinal Müller glia. In addition, FDG uptake was localized to periventricular hypothalamic tanycytes, whose morphological features were imaged simultaneously by X-ray absorption. We conclude that the high specificity of photon emission from F and its spatial mapping at ≤1 μm resolution demonstrates the ability to identify glucose uptake at subcellular resolution and holds remarkable potential for imaging glucose metabolism in biological tissue. © 2012 Wiley Periodicals, Inc.

Robust Nonlinear Mapping of Soil Contamination Using Support Regression

Spatial data analysis mapping and visualization is of great importance in various fields: environment, pollution, natural hazards and risks, epidemiology, spatial econometrics, etc. A basic task of spatial mapping is to make predictions based on some empirical data (measurements). A number of state-of-the-art methods can be used for the task: deterministic interpolations, methods of geostatistics: the family of kriging estimators (Deutsch and Journel, 1997), machine learning algorithms such as artificial neural networks (ANN) of different architectures, hybrid ANN-geostatistics models (Kanevski and Maignan, 2004; Kanevski et al., 1996), etc. All the methods mentioned above can be used for solving the problem of spatial data mapping. Environmental empirical data are always contaminated/corrupted by noise, and often with noise of unknown nature. That's one of the reasons why deterministic models can be inconsistent, since they treat the measurements as values of some unknown function that should be interpolated. Kriging estimators treat the measurements as the realization of some spatial randomn process. To obtain the estimation with kriging one has to model the spatial structure of the data: spatial correlation function or (semi-)variogram. This task can be complicated if there is not sufficient number of measurements and variogram is sensitive to outliers and extremes. ANN is a powerful tool, but it also suffers from the number of reasons. of a special type ? multiplayer perceptrons ? are often used as a detrending tool in hybrid (ANN+geostatistics) models (Kanevski and Maignank, 2004). Therefore, development and adaptation of the method that would be nonlinear and robust to noise in measurements, would deal with the small empirical datasets and which has solid mathematical background is of great importance. The present paper deals with such model, based on Statistical Learning Theory (SLT) - Support Vector Regression. SLT is a general mathematical framework devoted to the problem of estimation of the dependencies from empirical data (Hastie et al, 2004; Vapnik, 1998). SLT models for classification - Support Vector Machines - have shown good results on different machine learning tasks. The results of SVM classification of spatial data are also promising (Kanevski et al, 2002). The properties of SVM for regression - Support Vector Regression (SVR) are less studied. First results of the application of SVR for spatial mapping of physical quantities were obtained by the authorsin for mapping of medium porosity (Kanevski et al, 1999), and for mapping of radioactively contaminated territories (Kanevski and Canu, 2000). The present paper is devoted to further understanding of the properties of SVR model for spatial data analysis and mapping. Detailed description of the SVR theory can be found in (Cristianini and Shawe-Taylor, 2000; Smola, 1996) and basic equations for the nonlinear modeling are given in section 2. Section 3 discusses the application of SVR for spatial data mapping on the real case study - soil pollution by Cs137 radionuclide. Section 4 discusses the properties of the modelapplied to noised data or data with outliers.

MRS glucose mapping and PET joining forces: re-evaluation of the lumped constant in the rat brain under isoflurane anaesthesia.

Although numerous positron emission tomography (PET) studies with (18) F-fluoro-deoxyglucose (FDG) have reported quantitative results on cerebral glucose kinetics and consumption, there is a large variation between the absolute values found in the literature. One of the underlying causes is the inconsistent use of the lumped constants (LCs), the derivation of which is often based on multiple assumptions that render absolute numbers imprecise and errors hard to quantify. We combined a kinetic FDG-PET study with magnetic resonance spectroscopic imaging (MRSI) of glucose dynamics in Sprague-Dawley rats to obtain a more comprehensive view of brain glucose kinetics and determine a reliable value for the LC under isoflurane anaesthesia. Maps of Tmax /CMRglc derived from MRSI data and Tmax determined from PET kinetic modelling allowed to obtain an LC-independent CMRglc . The LC was estimated to range from 0.33 ± 0.07 in retrosplenial cortex to 0.44 ± 0.05 in hippocampus, yielding CMRglc between 62 ± 14 and 54 ± 11 μmol/min/100 g, respectively. These newly determined LCs for four distinct areas in the rat brain under isoflurane anaesthesia provide means of comparing the growing amount of FDG-PET data available from translational studies.

Interest rates mapping

The present study deals with the analysis and mapping of Swiss franc interest rates. Interest rates depend on time and maturity, defining term structure of the interest rate curves (IRC). In the present study IRC are considered in a two-dimensional feature space - time and maturity. Exploratory data analysis includes a variety of tools widely used in econophysics and geostatistics. Geostatistical models and machine learning algorithms (multilayer perceptron and Support Vector Machines) were applied to produce interest rate maps. IR maps can be used for the visualisation and pattern perception purposes, to develop and to explore economical hypotheses, to produce dynamic asset-liability simulations and for financial risk assessments. The feasibility of an application of interest rates mapping approach for the IRC forecasting is considered as well. (C) 2008 Elsevier B.V. All rights reserved.

The role of interval nodes in sentinel lymph node mapping and dissection for melanoma patients.

In sentinel node (SN) biopsy, an interval SN is defined as a lymph node or group of lymph nodes located between the primary melanoma and an anatomically well-defined lymph node group directly draining the skin. As shown in previous reports, these interval SNs seem to be at the same metastatic risk as are SNs in the usual, classic areas. This study aimed to review the incidence, lymphatic anatomy, and metastatic risk of interval SNs. METHODS: SN biopsy was performed at a tertiary center by a single surgical team on a cohort of 402 consecutive patients with primary melanoma. The triple technique of localization was used-that is, lymphoscintigraphy, blue dye, and gamma-probe. Otolaryngologic melanoma and mucosal melanoma were excluded from this analysis. SNs were examined by serial sectioning and immunohistochemistry. All patients with metastatic SNs were recommended to undergo a radical selective lymph node dissection. RESULTS: The primary locations of the melanomas included the trunk (188), an upper limb (67), or a lower limb (147). Overall, 97 (24.1%) of the 402 SNs were metastatic. Interval SNs were observed in 18 patients, in all but 2 of whom classic SNs were also found. The location of the primary was truncal in 11 (61%) of the 18, upper limb in 5, and lower limb in 2. One patient with a dorsal melanoma had drainage exclusively in a cervicoscapular area that was shown on removal to contain not lymph node tissue but only a blue lymph channel without tumor cells. Apart from the interval SN, 13 patients had 1 classic SN area and 3 patients 2 classic SN areas. Of the 18 patients, 2 had at least 1 metastatic interval SN and 2 had a classic SN that was metastatic; overall, 4 (22.2%) of 18 patients were node-positive. CONCLUSION: We found that 2 of 18 interval SNs were metastatic: This study showed that preoperative lymphoscintigraphy must review all known lymphatic areas in order to exclude an interval SN.

Quantitative multi-parameter mapping of R1, PD(*), MT, and R2(*) at 3T: a multi-center validation.

Multi-center studies using magnetic resonance imaging facilitate studying small effect sizes, global population variance and rare diseases. The reliability and sensitivity of these multi-center studies crucially depend on the comparability of the data generated at different sites and time points. The level of inter-site comparability is still controversial for conventional anatomical T1-weighted MRI data. Quantitative multi-parameter mapping (MPM) was designed to provide MR parameter measures that are comparable across sites and time points, i.e., 1 mm high-resolution maps of the longitudinal relaxation rate (R1 = 1/T1), effective proton density (PD(*)), magnetization transfer saturation (MT) and effective transverse relaxation rate (R2(*) = 1/T2(*)). MPM was validated at 3T for use in multi-center studies by scanning five volunteers at three different sites. We determined the inter-site bias, inter-site and intra-site coefficient of variation (CoV) for typical morphometric measures [i.e., gray matter (GM) probability maps used in voxel-based morphometry] and the four quantitative parameters. The inter-site bias and CoV were smaller than 3.1 and 8%, respectively, except for the inter-site CoV of R2(*) (<20%). The GM probability maps based on the MT parameter maps had a 14% higher inter-site reproducibility than maps based on conventional T1-weighted images. The low inter-site bias and variance in the parameters and derived GM probability maps confirm the high comparability of the quantitative maps across sites and time points. The reliability, short acquisition time, high resolution and the detailed insights into the brain microstructure provided by MPM makes it an efficient tool for multi-center imaging studies.