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.

Paper 5: Surveillance of multiple congenital anomalies: implementation of a computer algorithm in European registers for classification of cases.

BACKGROUND: Surveillance of multiple congenital anomalies is considered to be more sensitive for the detection of new teratogens than surveillance of all or isolated congenital anomalies. Current literature proposes the manual review of all cases for classification into isolated or multiple congenital anomalies. METHODS: Multiple anomalies were defined as two or more major congenital anomalies, excluding sequences and syndromes. A computer algorithm for classification of major congenital anomaly cases in the EUROCAT database according to International Classification of Diseases (ICD)v10 codes was programmed, further developed, and implemented for 1 year's data (2004) from 25 registries. The group of cases classified with potential multiple congenital anomalies were manually reviewed by three geneticists to reach a final agreement of classification as "multiple congenital anomaly" cases. RESULTS: A total of 17,733 cases with major congenital anomalies were reported giving an overall prevalence of major congenital anomalies at 2.17%. The computer algorithm classified 10.5% of all cases as "potentially multiple congenital anomalies". After manual review of these cases, 7% were agreed to have true multiple congenital anomalies. Furthermore, the algorithm classified 15% of all cases as having chromosomal anomalies, 2% as monogenic syndromes, and 76% as isolated congenital anomalies. The proportion of multiple anomalies varies by congenital anomaly subgroup with up to 35% of cases with bilateral renal agenesis. CONCLUSIONS: The implementation of the EUROCAT computer algorithm is a feasible, efficient, and transparent way to improve classification of congenital anomalies for surveillance and research.

Classification of current anticancer immunotherapies.

During the past decades, anticancer immunotherapy has evolved from a promising therapeutic option to a robust clinical reality. Many immunotherapeutic regimens are now approved by the US Food and Drug Administration and the European Medicines Agency for use in cancer patients, and many others are being investigated as standalone therapeutic interventions or combined with conventional treatments in clinical studies. Immunotherapies may be subdivided into "passive" and "active" based on their ability to engage the host immune system against cancer. Since the anticancer activity of most passive immunotherapeutics (including tumor-targeting monoclonal antibodies) also relies on the host immune system, this classification does not properly reflect the complexity of the drug-host-tumor interaction. Alternatively, anticancer immunotherapeutics can be classified according to their antigen specificity. While some immunotherapies specifically target one (or a few) defined tumor-associated antigen(s), others operate in a relatively non-specific manner and boost natural or therapy-elicited anticancer immune responses of unknown and often broad specificity. Here, we propose a critical, integrated classification of anticancer immunotherapies and discuss the clinical relevance of these approaches.

Enchondromatosis revisited: New classification with molecular basis.

The so-called "enchondromatoses" are skeletal disorders defined by the presence of ectopic cartilaginous tissue within bone tissue. The clinical and radiographic features of the different enchondromatoses are distinct, and grouping them does not reflect a common pathogenesis but simply a similar radiographic appearance and thus the need for a differential diagnosis. Recent advances in the understanding of their molecular and cellular bases confirm the heterogeneous nature of the different enchondromatoses. Some, like Ollier disease, Maffucci disease, metaphyseal chondromatosis with hydroxyglutaric aciduria, and metachondromatosis are produced by a dysregulation of chondrocyte proliferation, while others (such as spondyloenchondrodysplasia or dysspondyloenchondromatosis) are caused by defects in structure or metabolism of cartilage or bone matrix. In other forms (e.g., the dominantly inherited genochondromatoses), the basic defect remains to be determined. The classification, proposed by Spranger and associates in 1978 and tentatively revised twice, was based on the radiographic appearance, the anatomic sites involved, and the mode of inheritance. The new classification proposed here integrates the molecular genetic advances and delineates phenotypic families based on the molecular defects. Reference radiographs are provided to help in the diagnosis of the well-defined forms. In spite of advances, many cases remain difficult to diagnose and classify, implying that more variants remain to be defined at both the clinical and molecular levels. © 2012 Wiley Periodicals, Inc.

Lake dwellers occupation gap in Lake Geneva (France-Switzerland) possibly explained by an earthquake-mass movement-tsunami event during Early Bronze Age

High-resolution seismic and sediment core data from the 'Grand Lac' basin of Lake Geneva reveal traces of repeated slope instabilities with one main slide-evolved mass-flow (minimum volume 0.13 km3) that originated from the northern lateral slope of the lake near the city of Lausanne. Radiocarbon dating of organic remains sampled from the top of the main deposit gives an age interval of 1865-1608 BC. This date coincides with the age interval for a mass movement event described in the 'Petit Lac' basin of Lake Geneva (1872-1622 BC). Because multiple mass movements took place at the same time in different parts of the lake, we consider the most likely trigger mechanism to be a strong earthquake (Mw 6) that occurred in the period between 1872 and 1608 BC. Based on numerical simulations, we show the major deposit near Lausanne would have generated a tsunami with local wave heights of up to 6 m. The combined effects of the earthquake and the following tsunami provide a possible explanation for a gap in lake dwellers occupation along the shores of Lake Geneva revealed by dendrochronological dating of two palafitte archaeological sites.

Low latitude Paleocene to early Eocene radiolarian systematics, biochronology and paleoenvironmental analysis

Highly diverse radiolarian faunas of latest Maastrichtian to early Eocene age have been recovered from the low latitude realm in order to contribute to the clarification of radiolarian taxonomy, construct a zonation based on a discrete sequence of co-existence intervals of species ranging from the late Paleocene to early Eocene and to describe a rich low latitude latest Cretaceous to late Paleocene fauna. 225 samples of late Paleocene to early Eocene age have been collected from ODP Leg 171 B-Hole 1051 A (Blake Nose), DSDP Leg 43-Site 384 (Northwest Atlantic) and DSDP Leg 10-Sites 86, 94, 95, 96. Sequences consist of mainly pelagic oozes and chalks, with some clay and ash layers. A new imaging technique is devised to perform (in particular on topotypic material) both transmitted light microscopy and SEM imaging on individual radiolarian specimens. SEM precedes transmitted light imaging. Radiolarians are adhered to a cover slip (using nail varnish) which is secured to a stub using conductive levers. Specimens are then photographed in low vacuum (40-50Pa; 0.5mbar), which enables charge neutralization by ionized molecules of the chamber atmosphere. Thus gold coating is avoided and subsequently this allows transmitted light imaging to follow. The conductive levers are unscrewed and the cover slip is simply overturned and mounted with Canada balsam. In an attempt towards a post-Haeckelian classification, the initial spicule (Entactinaria), micro- or macrosphere (Spumellaria) and initial spicule and cephalis (Nassellaria) have been studied by slicing Entactinaria and Spumellaria, and by tilting Nassellaria in the SEM chamber. A new genus of the family Coccodiscidae is erected and Spongatractus HAECKEL is re-located to the subfamily Axopruinae. The biochronology has been carried out using the Unitary Association Method (Guex 1977, 1991). A database recording the occurrences of 112 species has been used to establish a succession of 22 Unitary Associations. Each association is correlated to chronostratigraphy via calcareous microfossils that were previously studied by other authors. The 22 UAs have been united into seven Unitary Associations Zones (UAZones) (JP10- JE4). The established zones permit to distinguish supplementary subdivisions within the existing zonation. The low-latitude Paleocene radiolarian zonation established by Sanfilippo and Nigrini (1998a) is incomplete due to the lack of radiolarian-bearing early Paleocene sediments. In order to contribute to the study of sparsely known low latitude early Paleocene faunas, 80 samples were taken from the highly siliceous Guayaquil Formation (Ecuador). The sequence consists of black cherts, shales, siliceous limestones and volcanic ash layers. The carbonate content increases up section. Age control is supplied by sporadic occurrences of silicified planktonic foraminifera casts. One Cretaceous zone and seven Paleocene zones have been identified. The existing zonation for the South Pacific can be applied to the early-early late Paleocene sequence, although certain marker species have significantly shorter ranges (notably Buryella foremanae and B. granulata). Despite missing marker species in the late Paleocene, faunal distribution correlates reasonably to the Low-Latitude zonation. An assemblage highly abundant in Lithomelissa, Lophophaena and Cycladophora in the upper RP6 zone (correlated by the presence of Pterocodon poculum, Circodiscus circularis, Pterocodon? sp. aff. P. tenellus and Stylotrochus nitidus) shows a close affinity to contemporaneous faunas reported from Site 1121, Campbell Plateau. Coupled with a high diatom abundance (notably Aulacodiscus spp. and Arachnoidiscus spp.), these faunas are interpreted as reflecting a period of enhanced biosiliceous productivity during the late Paleocene. The youngest sample is void of radiolarians, diatoms and sponge spicules yet contains many pyritized infaunal benthic foraminifera which are akin to the midway-type fauna. The presence of this fauna suggests deposition in a neritic environment. This is in contrast to the inferred bathyal slope depositional environment of the older Paleocene sediments and suggests a shoaling of the depositional environment which may be related to a coeval major accretionary event. RESUME DE LA THESE Des faunes de radiolaires de basses latitudes très diversifiées d'âge Maastrichtien terminal à Eocène inférieur, ont été étudiées afin de contribuer à la clarification de leur taxonomie, de construire une biozonation basée sur une séquence discrète d'intervalles de coexistence des espèces d'age Paléocène supérieur à Eocène inférieur et de décrire une riche faune de basse latitude allant du Crétacé terminal au Paléocène supérieur. L'étude de cette faune contribue particulièrement à la connaissance des insaisissables radiolaires de basses latitudes du Paléocène inférieur. 225 échantillons d'âge Paléocène supérieur à Eocène inférieur provenant des ODP Leg 171B-Site 1051A (Blake Nose), Leg DSDP 43-Site 384 (Atlantique Nord -Ouest) et des DSDP Leg 10 -Sites 86, 94, 95, 96, ont été étudiés. Ces séquences sont constituées principalement de « ooze » et de « chalks »pélagiques ainsi que de quelques niveaux de cendres et d'argiles. Une nouvelle technique d'imagerie a été conçue afin de pouvoir prendre conjointement des images en lumière transmise et au Microscope Electronique à Balayage (MEB) de spécimens individuels. Ceci à été particulièrement appliqué à l'étude des topotypes. L'imagerie MEB précède l'imagerie en lumière transmise. Les radiolaires sont collés sur une lame pour micropaléontologie (au moyen de vernis à ongles) qui est ensuite fixée à un porte-objet à l'aide de bras métalliques conducteurs. Les spécimens sont ensuite photographiés en vide partiel (40-50Pa; 0.5mbar), ce qui permet la neutralisation des charges électrostatiques dues à la présence de molécules ionisées dans l'atmosphère de la chambre d'observation. Ainsi la métallisation de l'échantillon avec de l'or n'est plus nécessaire et ceci permet l'observation ultérieure en lumière transmise. Les bras conducteurs sont ensuite dévissés et la lame est simplement retournée et immergée dans du baume du Canada. Dans une approche de classification post Haeckelienne, le spicule initial (Entactinaires), la micro- ou macro -sphère (Spumellaires) et le spicule initial et cephalis (Nassellaires) ont été étudiés. Ceci a nécessité le sectionnement d'Entactinaires et de Spumellaires, et de pivoter les Nassellaires dans la chambre d'observation du MEB. Un nouveau genre de la Famille des Coccodiscidae a été érigé et Spongatractus HAECKEL à été réassigné à la sous-famille des Axopruninae. L'analyse biostratigraphique à été effectuée à l'aide de la méthode des Associations Unitaires {Guex 1977, 1991). Une base de données enregistrant les présences de 112 espèces à été utilisée poux établir une succession de 22 Associations Unitaires. Chaque association est corrélée à la chronostratigraphie au moyen de microfossiles calcaires précédemment étudiés par d'autres auteurs. Les 22 UAs ont été combinées en sept Zones d'Associations Unitaires (UAZones) (JP10- JE4). Ces Zones permettent d'insérer des subdivisions supplémentaires dans la zonation actuelle. La zonation de basses latitudes du Paléocène établie par Sanfilippo et Nigrini (1998a) est incomplète due au manque de sédiments du Paléocène inférieur contenant des radiolaires. Afin de contribuer à l'étude des faunes peu connues des basses latitudes du Paléocène inférieur, 80 échantillons ont été prélevés d'une section siliceuse de la Formation de Guayaquil (Equateur). La séquence est composée de cherts noirs, de shales, de calcaires siliceux et de couches de cendres volcaniques. La fraction carbonatée augmente vers le haut de la section. Des contraintes chronologiques sont fournies par la présence sporadique de moules de foraminifères planctoniques. Une zone d'intervalles du Crétacé et sept du Paléocène ont été mises en évidence. Bien que certaines espèces marqueur ont des distributions remarquablement plus courtes (notamment Buryella foremanae et B. granulata), la zonation existante pour le Pacifique Sud est applicable à la séquence d'age Paléocène inférieure à Paléocène supérieur basal étudiée. Malgré l'absence d'espèces marqueur du Paléocène supérieur, la succession faunistique se corrèle raisonnablement avec la zonation pour les basses latitudes. Un assemblage contenant d'abondants représentant du genre Lithomelissa, Lophophaena et Cycladophora dans la zone RP6 (correlée par la présence de Pterocodon poculum, Circodiscus circularis, Pterocodon? sp. aff. P. tenellus et Stylotrochus nitidus) montre une grande similitude avec certaines faunes issues des hauts latitudes et d'age semblable décrites par Hollis (2002, Site 1121, Campbell Plateau). Ceci, en plus d'une abondance importante en diatomés (notamment Aulacodiscus spp. et Arachnoidiscus spp.) nous mènent à interpréter cette faune comme témoin d'un épisode de productivité biosiliceuse accrue dans le Paléocène supérieur. L'échantillon le plus jeune, dépourvu de radiolaires, de diatomés et de spicules d'éponge contient de nombreux foraminifères benthiques infaunaux pyritisés. Les espèces identifiées sont caractéristiques d'une faune de type midway. La présence de ces foraminifères suggère un environnement de type néritique. Ceci est en contraste avec l'environnement de pente bathyale caractérisent les sédiments sous-jacent. Cette séquence de diminution de la tranche d'eau peut être associée à un événement d'accrétion majeure. RESUME DE LA THESE (POUR LE GRAND PUBLIC) Les radiolaires constituent le groupe de plancton marin le plus divers et le plus largement répandu de l'enregistrement fossile. Un taux d'évolution rapide et une variation géographique considérable des populations font des radiolaires un outil de recherche sans égal pour la biostratigraphie et la paléocéanographie. Néanmoins, avant de pouvoir les utiliser comme outils de travail, il est essentiel d'établir une solide base taxonomique. L'étude des Radiolaires peut impliquer plusieurs techniques d'extraction, d'observation et d'imagerie qui sont dépendantes du degré d'altération diagénétique des spécimens. Le squelette initial, qu'il s'agisse d'un spicule initial (Entactinaria), d'une micro- ou macro -sphère (Spumellaria) ou d'un spicule initial et d'un cephalis (Nassellaria), est l'élément le plus constant au cours de l'évolution et devrait représenter le fondement de la systématique. Des échantillons provenant de carottes de basses latitudes du Deep Sea Drilling Project et de l' Ocean Drilling ont été étudiés. De nouvelles techniques d'imagerie et de sectionnement ont été développées sur des topotypes de radiolaires préservés en opale, dans le but d'étudier les caractéristiques de leur squelette initial qui n'étaient pas visibles dans leur illustration originale. Ceci aide entre autre à comparer des spécimens recristallisés en quartz, provenant de terrains accrétés, avec les holotypes en opale de la littérature. La distribution des espèces étudiés a fourni des données biostratigraphiques qui ont été compilées à l'aide de la méthode des Associations Unitaires (Guez 1977, 1991). Il s'agit d'un modèle mathématique déterministe conçu pour exploiter la totalité de l'assemblage plutôt que de se confiner à l'utilisation de taxons marqueurs individuels. Une séquence de 22 Associations Unitaires a été établie pour la période allant du Paléocène supérieur à l'Éocène inférieur. Chaque Association Unitaire a été corrélée à l'échelle de temps absolue à l'aide de microfossiles calcaires. Les 22 UAs ont été combinées en sept Zones d'Associations Unitaires (JP10- JE4). Ces Zones permettent d'insérer des subdivisions supplémentaires dans la zonation actuelle. Les radiolaires du Paléocène inférieur à moyen des basses latitudes sont rares. Les meilleures sections connues se trouvent dans les hautes latitudes (Nouvelle Zélande). Quelques assemblages épars ont été mentionnés par le passé en Californie, en Équateur et en Russie. Une séquence siliceuse de 190 mètres dans la Formation de Guayaquil (Équateur), s'étendant du Maastrichtien supérieur au Paléocène supérieur, a fourni des faunes relativement bien préservées. L'étude de ces faunes a permis de mettre en évidence la première séquence complète de radiolaires de basses latitudes dans le Paléocène inférieure. Huit zones allant du Crétacé terminal au Paléocène supérieur ont pu être appliqués et la présence de foraminifères planctoniques a fournie plusieurs points d'attache chronologiques. Dans le Paléocène supérieur, un riche assemblage contenant d'abondants diatomés et radiolaires ayant des similitudes faunistiques marquantes avec des assemblages de hautes latitudes de Nouvelle Zélande, témoigne d'un épisode de productivité biosiliceuse accrue pendant cette période. Étant donné que la pointe du continent sud-américain et l'Antarctique étaient plus proches au cours du Paléocène, ce phénomène peut être expliqué par le transport, le long de la côte ouest de l'Amérique du Sud, d'eaux riches en nutriments en provenance de l'Océan Antarctique. Suite à cet épisode, l'enregistrement en radiolaires est interrompu. Ceci peut être associé à des événements tectoniques régionaux qui ont eu pour effet de diminuer la tranche d'eau relative, rendant l'environnement plus favorable aux foraminifères benthiques qui sont abondamment présents dans l'échantillon le plus jeune de la séquence.

Validation of tissue modelization and classification techniques in T1-weighted MR brain images

We propose a deep study on tissue modelization andclassification Techniques on T1-weighted MR images. Threeapproaches have been taken into account to perform thisvalidation study. Two of them are based on FiniteGaussian Mixture (FGM) model. The first one consists onlyin pure gaussian distributions (FGM-EM). The second oneuses a different model for partial volume (PV) (FGM-GA).The third one is based on a Hidden Markov Random Field(HMRF) model. All methods have been tested on a DigitalBrain Phantom image considered as the ground truth. Noiseand intensity non-uniformities have been added tosimulate real image conditions. Also the effect of ananisotropic filter is considered. Results demonstratethat methods relying in both intensity and spatialinformation are in general more robust to noise andinhomogeneities. However, in some cases there is nosignificant differences between all presented methods.

A tutorial review of electrical neuroimaging from group-average to single-trial event-related potentials.

This tutorial review details some of the recent advances in signal analyses applied to event-related potential (ERP) data. These "electrical neuroimaging" analyses provide reference-independent measurements of response strength and response topography that circumvent statistical and interpretational caveats of canonical ERP analysis methods while also taking advantage of the greater information provided by high-density electrode montages. Electrical neuroimaging can be applied across scales ranging from group-averaged ERPs to single-subject and single-trial datasets. We illustrate these methods with a tutorial dataset and place particular emphasis on their suitability for studies of clinical and/or developmental populations.

Classification of very high spatial resolution imagery using mathematical morphology and support vector machines

We investigate the relevance of morphological operators for the classification of land use in urban scenes using submetric panchromatic imagery. A support vector machine is used for the classification. Six types of filters have been employed: opening and closing, opening and closing by reconstruction, and opening and closing top hat. The type and scale of the filters are discussed, and a feature selection algorithm called recursive feature elimination is applied to decrease the dimensionality of the input data. The analysis performed on two QuickBird panchromatic images showed that simple opening and closing operators are the most relevant for classification at such a high spatial resolution. Moreover, mixed sets combining simple and reconstruction filters provided the best performance. Tests performed on both images, having areas characterized by different architectural styles, yielded similar results for both feature selection and classification accuracy, suggesting the generalization of the feature sets highlighted.