Machine learning algorithms for spatial data. Case studies: Environmental pollution, natural hazards, renewable resources

This paper presents general problems and approaches for the spatial data analysis using machine learning algorithms. Machine learning is a very powerful approach to adaptive data analysis, modelling and visualisation. The key feature of the machine learning algorithms is that they learn from empirical data and can be used in cases when the modelled environmental phenomena are hidden, nonlinear, noisy and highly variable in space and in time. Most of the machines learning algorithms are universal and adaptive modelling tools developed to solve basic problems of learning from data: classification/pattern recognition, regression/mapping and probability density modelling. In the present report some of the widely used machine learning algorithms, namely artificial neural networks (ANN) of different architectures and Support Vector Machines (SVM), are adapted to the problems of the analysis and modelling of geo-spatial data. Machine learning algorithms have an important advantage over traditional models of spatial statistics when problems are considered in a high dimensional geo-feature spaces, when the dimension of space exceeds 5. Such features are usually generated, for example, from digital elevation models, remote sensing images, etc. An important extension of models concerns considering of real space constrains like geomorphology, networks, and other natural structures. Recent developments in semi-supervised learning can improve modelling of environmental phenomena taking into account on geo-manifolds. An important part of the study deals with the analysis of relevant variables and models' inputs. This problem is approached by using different feature selection/feature extraction nonlinear tools. To demonstrate the application of machine learning algorithms several interesting case studies are considered: digital soil mapping using SVM, automatic mapping of soil and water system pollution using ANN; natural hazards risk analysis (avalanches, landslides), assessments of renewable resources (wind fields) with SVM and ANN models, etc. The dimensionality of spaces considered varies from 2 to more than 30. Figures 1, 2, 3 demonstrate some results of the studies and their outputs. Finally, the results of environmental mapping are discussed and compared with traditional models of geostatistics.

Evaluation of organ-specific peripheral doses after 2-dimensional, 3-dimensional and hybrid intensity modulated radiation therapy for breast cancer based on Monte Carlo and convolution/superposition algorithms: implications for secondary cancer risk assessment.

To make a comprehensive evaluation of organ-specific out-of-field doses using Monte Carlo (MC) simulations for different breast cancer irradiation techniques and to compare results with a commercial treatment planning system (TPS). Three breast radiotherapy techniques using 6MV tangential photon beams were compared: (a) 2DRT (open rectangular fields), (b) 3DCRT (conformal wedged fields), and (c) hybrid IMRT (open conformal+modulated fields). Over 35 organs were contoured in a whole-body CT scan and organ-specific dose distributions were determined with MC and the TPS. Large differences in out-of-field doses were observed between MC and TPS calculations, even for organs close to the target volume such as the heart, the lungs and the contralateral breast (up to 70% difference). MC simulations showed that a large fraction of the out-of-field dose comes from the out-of-field head scatter fluence (>40%) which is not adequately modeled by the TPS. Based on MC simulations, the 3DCRT technique using external wedges yielded significantly higher doses (up to a factor 4-5 in the pelvis) than the 2DRT and the hybrid IMRT techniques which yielded similar out-of-field doses. In sharp contrast to popular belief, the IMRT technique investigated here does not increase the out-of-field dose compared to conventional techniques and may offer the most optimal plan. The 3DCRT technique with external wedges yields the largest out-of-field doses. For accurate out-of-field dose assessment, a commercial TPS should not be used, even for organs near the target volume (contralateral breast, lungs, heart).

Trabecular Bone Score: A Noninvasive Analytical Method Based Upon the DXA Image.

The trabecular bone score (TBS) is a gray-level textural metric that can be extracted from the two-dimensional lumbar spine dual-energy X-ray absorptiometry (DXA) image. TBS is related to bone microarchitecture and provides skeletal information that is not captured from the standard bone mineral density (BMD) measurement. Based on experimental variograms of the projected DXA image, TBS has the potential to discern differences between DXA scans that show similar BMD measurements. An elevated TBS value correlates with better skeletal microstructure; a low TBS value correlates with weaker skeletal microstructure. Lumbar spine TBS has been evaluated in cross-sectional and longitudinal studies. The following conclusions are based upon publications reviewed in this article: 1) TBS gives lower values in postmenopausal women and in men with previous fragility fractures than their nonfractured counterparts; 2) TBS is complementary to data available by lumbar spine DXA measurements; 3) TBS results are lower in women who have sustained a fragility fracture but in whom DXA does not indicate osteoporosis or even osteopenia; 4) TBS predicts fracture risk as well as lumbar spine BMD measurements in postmenopausal women; 5) efficacious therapies for osteoporosis differ in the extent to which they influence the TBS; 6) TBS is associated with fracture risk in individuals with conditions related to reduced bone mass or bone quality. Based on these data, lumbar spine TBS holds promise as an emerging technology that could well become a valuable clinical tool in the diagnosis of osteoporosis and in fracture risk assessment. © 2014 American Society for Bone and Mineral Research.

Le football comme religion populaire et comme culture mondialisée : brèves notations en vue d'une interprétation critique d'une quasi-religion contemporaine

(Résumé de l'ouvrage) Le présent recueil d'hommages veut témoigner de l'importance de l'oeuvre théologique de Jean Richard, professeur à la Faculté de théologie et de sciences religieuses de l'Université Laval. Désireux de signifier leur reconnaissance au père Richard, les auteurs de ce livre ont proposé des textes s'inscrivant dans l'un ou l'autre de ses champs de recherche privilégiés. Le tableau des contributeurs à ces mélanges regroupe des théologiens de tous les âges, hommes et femmes, provenant du Canada, mais aussi des États-Unis et de l'Europe, de langue française, anglaise ou allemande, de confession catholique ou protestante, oeuvrant dans des champs disciplinaires aussi divers que la théologie systématique, la théologie fondamentale, l'éthique théologique, l'exégèse ou encore la philosophie. Cette diversité est révélatrice de la personnalité théologique du père Richard, du caractère international de son réseau d'amitiés et de recherche et de l'ampleur de son influence. Elle témoigne aussi de l'audace d'une oeuvre qui, entreprise dans les années 1960, a dès le départ relevé le défi de l'aggiornamento que venait de poser en toute urgence son Église et qui a très vite conduit son auteur « aux confins », pour reprendre - et pour cause - une expression de son mentor Paul Tillich qui lui sied à merveille. Les contributions réunies dans ce volume ne résultent pas d'un plan concerté. D'autant plus significatifs sont les recoupements thématiques qu'on y observe, qui, par un hasard seulement apparent, restituent les principaux axes de la production théologique du père Jean Richard. L'ouverture au présent et a ses problématiques propres, ainsi que les défis conceptuels par là posés à l'effort théologique constituent les deux constantes de ces contributions, comme de l'oeuvre qui les a inspirées.

Analytical validation of a lymphocytic choriomeningitis virus real-time RT-PCR assay.

Lymphocytic choriomeningitis virus (LCMV) is a rare cause of central nervous system disease in humans. Screening by real-time RT-PCR assay is of interest in the case of aseptic meningitis of unknown etiology. A specific LCMV real-time RT-PCR assay, based on the detection of genomic sequences of the viral nucleoprotein (NP), was developed to assess the presence of LCMV in cerebrospinal fluids (CSF) sent for viral screening to a Swiss university hospital laboratory. A 10-fold dilution series assay using a plasmid containing the cDNA of the viral NP of the LCMV isolate Armstrong (Arm) 53b demonstrated the high sensitivity of the assay with a lowest detection limit of ≤50 copies per reaction. High sensitivity was confirmed by dilution series assays in a pool of human CSF using four different LCMV isolates (Arm53b, WE54, Traub and E350) with observed detection limits of ≤10PFU/ml (Arm53b and WE54) and 1PFU/ml (Traub and E350). Analysis of 130 CSF showed no cases of acute infection. The absence of positive cases was confirmed by a published PCR assay detecting all Old World arenaviruses. This study validates a specific and sensitive real-time RT-PCR assay for the diagnosis of LCMV infections. Results showed that LCMV infections are extremely rare in hospitalized patients western in Switzerland.

Artificial neural network study of whole-cell bacterial bioreporter response determined using fluorescence flow cytometry.

Genetically engineered bioreporters are an excellent complement to traditional methods of chemical analysis. The application of fluorescence flow cytometry to detection of bioreporter response enables rapid and efficient characterization of bacterial bioreporter population response on a single-cell basis. In the present study, intrapopulation response variability was used to obtain higher analytical sensitivity and precision. We have analyzed flow cytometric data for an arsenic-sensitive bacterial bioreporter using an artificial neural network-based adaptive clustering approach (a single-layer perceptron model). Results for this approach are far superior to other methods that we have applied to this fluorescent bioreporter (e.g., the arsenic detection limit is 0.01 microM, substantially lower than for other detection methods/algorithms). The approach is highly efficient computationally and can be implemented on a real-time basis, thus having potential for future development of high-throughput screening applications.

How Wages and Employment Adjust to Trade Liberalization: Quasi-Experimental Evidence from Austria

We study the response of regional employment and nominal wages to trade liberalization, exploiting the natural experiment provided by the opening of Central and Eastern European markets after the fall of the Iron Curtain in 1990. Using data for Austrian municipalities, we examine differential pre- and post-1990 wage and employment growth rates between regions bordering the formerly communist economies and interior regions. If the "border regions" are defined narrowly, within a band of less than 50 kilometers, we can identify statistically significant liberalization effects on both employment and wages. While wages responded earlier than employment, the employment effect over the entire adjustment period is estimated to be around three times as large as the wage effect. The implied slope of the regional labor supply curve can be replicated in an economic geography model that features obstacles to labor migration due to immobile housing and to heterogeneous locational preferences.

Urinary analysis of four testosterone metabolites and pregnanediol by gas chromatography-combustion-isotope ratio mass spectrometry after oral administrations of testosterone.

The most frequently used method to demonstrate testosterone abuse is the determination of the testosterone and epitestosterone concentration ratio (T/E ratio) in urine. Nevertheless, it is known that factors other than testosterone administration may increase the T/E ratio. In the last years, the determination of the carbon isotope ratio has proven to be the most promising method to help discriminate between naturally elevated T/E ratios and those reflecting T use. In this paper, an excretion study following oral administration of 40 mg testosterone undecanoate initially and 13 h later is presented. Four testosterone metabolites (androsterone, etiocholanolone, 5 alpha-androstanediol, and 5 beta-androstanediol) together with an endogenous reference (5 beta-pregnanediol) were extracted from the urines and the delta(13)C/(12)C ratio of each compound was analyzed by gas chromatography-combustion-isotope ratio mass spectrometry. The results show similar maximum delta(13)C-value variations (parts per thousand difference of delta(13)C/(12)C ratio from the isotope ratio standard) for the T metabolites and concomitant changes of the T/E ratios after administration of the first and the second dose of T. Whereas the T/E ratios as well as the androsterone, etiocholanolone and 5 alpha-androstanediol delta(13)C-values returned to the baseline 15 h after the second T administration, a decrease of the 5 beta-androstanediol delta-values could be detected for over 40 h. This suggests that measurements of 5 beta-androstanediol delta-values allow the detection of a testosterone ingestion over a longer post-administration period than other T metabolites delta(13)C-values or than the usual T/E ratio approach.

Study of common database feeding with results coming from different analytical methods in the framework of illicit drugs chemical profiling

Analytical results harmonisation is investigated in this study to provide an alternative to the restrictive approach of analytical methods harmonisation which is recommended nowadays for making possible the exchange of information and then for supporting the fight against illicit drugs trafficking. Indeed, the main goal of this study is to demonstrate that a common database can be fed by a range of different analytical methods, whatever the differences in levels of analytical parameters between these latter ones. For this purpose, a methodology making possible the estimation and even the optimisation of results similarity coming from different analytical methods was then developed. In particular, the possibility to introduce chemical profiles obtained with Fast GC-FID in a GC-MS database is studied in this paper. By the use of the methodology, the similarity of results coming from different analytical methods can be objectively assessed and the utility in practice of database sharing by these methods can be evaluated, depending on profiling purposes (evidential vs. operational perspective tool). This methodology can be regarded as a relevant approach for database feeding by different analytical methods and puts in doubt the necessity to analyse all illicit drugs seizures in one single laboratory or to implement analytical methods harmonisation in each participating laboratory.

Echo-time independent signal modulations for strongly coupled systems in triple echo localization schemes: an extension of S-PRESS editing.

The double spin-echo point resolved spectroscopy sequence (PRESS) is a widely used method and standard in clinical MR spectroscopy. Existence of important J-modulations at constant echo times, depending on the temporal delays between the rf-pulses, have been demonstrated recently for strongly coupled spin systems and were exploited for difference editing, removing singlets from the spectrum (strong-coupling PRESS, S-PRESS). A drawback of this method for in vivo applications is that large signal modulations needed for difference editing occur only at relatively long echo times. In this work we demonstrate that, by simply adding a third refocusing pulse (3S-PRESS), difference editing becomes possible at substantially shorter echo times while, as applied to citrate, more favorable lineshapes can be obtained. For the example of an AB system an analytical description of the MR signal, obtained with this triple refocusing sequence (3S-PRESS), is provided.