Decision fusion for the classification of hyperspectral data: Outcome of the 2008 GRS-S data fusion contest

The 2008 Data Fusion Contest organized by the IEEE Geoscience and Remote Sensing Data Fusion Technical Committee deals with the classification of high-resolution hyperspectral data from an urban area. Unlike in the previous issues of the contest, the goal was not only to identify the best algorithm but also to provide a collaborative effort: The decision fusion of the best individual algorithms was aiming at further improving the classification performances, and the best algorithms were ranked according to their relative contribution to the decision fusion. This paper presents the five awarded algorithms and the conclusions of the contest, stressing the importance of decision fusion, dimension reduction, and supervised classification methods, such as neural networks and support vector machines.

A novel method for automated classification of epileptiform activity in the human electroencephalogram-based on independent component analysis.

Diagnosis of several neurological disorders is based on the detection of typical pathological patterns in the electroencephalogram (EEG). This is a time-consuming task requiring significant training and experience. Automatic detection of these EEG patterns would greatly assist in quantitative analysis and interpretation. We present a method, which allows automatic detection of epileptiform events and discrimination of them from eye blinks, and is based on features derived using a novel application of independent component analysis. The algorithm was trained and cross validated using seven EEGs with epileptiform activity. For epileptiform events with compensation for eyeblinks, the sensitivity was 65 +/- 22% at a specificity of 86 +/- 7% (mean +/- SD). With feature extraction by PCA or classification of raw data, specificity reduced to 76 and 74%, respectively, for the same sensitivity. On exactly the same data, the commercially available software Reveal had a maximum sensitivity of 30% and concurrent specificity of 77%. Our algorithm performed well at detecting epileptiform events in this preliminary test and offers a flexible tool that is intended to be generalized to the simultaneous classification of many waveforms in the EEG.

Phylogeny of shrews (Soricidae) : taxonomic and biogeographic implications

Résumé Les Soricidae sont l'une des plus grandes familles de mammifères avec plus de 300 espèces décrites. Elle a été récemment divisée en trois sous-familles, les Soricidae, qui sont distribuées dans la région Holarctique, les Crocidurinae en Afrique et en Eurasie, et les Myosoricinae en Afrique. La diversité spécifique de cette famille a conduit à des interprétations taxonomiques multiples, qui sont à l'origine de polémiques entre spécialistes, et même les premiers résultats moléculaires ont été fortement contradictoires. Le but de cette thèse est donc d'appliquer des meilleures techniques sur des échantillons mieux ciblés, afin de résoudre les contradictions taxonomiques et comprendre l'histoire de cette famille. Par le biais de marqueurs génétiques mitochondriaux et nucléaires, j'ai étudié: (i) Les relations taxonomiques à différent niveaux hiérarchiques au sein des Soricidae, c'est-à dire, entre les sous-familles, tribus, et genres, ainsi qu'au sein de deux complexes d'espèces largement distribués, et d'une espèce européenne, le but étant d'établir la congruence entre les données génétiques et les interprétations morphologiques classiques. (ii) Les relations biogéographiques, soit l'origine potentielle des différentes sous-familles, tribus, et genres, le nombre d'échanges intercontinentaux, ainsi que la structure phylogéographique à un niveau (péri)-spécifique, afin d'établir l'histoire de la diversification de cette famille. Les analyses combinées d'ADN mitochondrial et nucléaire ont montré un rapport clair entre les taxa à un niveau taxonomique élevé, mettant en évidence les rapports entre les sous-familles, les tribus, et les genres. Bien que Myosorex constitue un groupe monophylétique distinct, sa définition en tant que sous-famille séparée ne peut pas être reconnue. Ainsi, nous proposons d'attribuer un niveau de tribu pour ce clade (inclus dans les Crocidurinae). Nous avons également montré l'inclusion du genre Anourosorex dans les Soricinae et non en position basale dans les Soricidae. Au sein des Crocidurinae, Suncus s'est révélé être paraphylétique, et le genre Diplomesodon devrait être considéré d'un point de vue génétique comme invalide, puisque il se trouve au sein du clade du genre Crocidura. À un niveau taxonomique plus bas, nous avons montré la monophylie de deux complexes d'espèces largement distribués, le groupe de C. suaveolens et de C. olivieri. Néanmoins à l'intérieur de ceux-ci, des différences majeures avec la classification morphologique se sont révélées. Par exemples, C. sibirica n'est pas une espèce valide, les analyses de phylogénie moléculaire ne montrant pas de variations génétiques entre celle-ci et un échantillon de la localité type de C. suaveolens. D'un point de vue biogéographique, les fluctuations climatiques et les activités tectoniques des 20 derniers millions d'années ont fortement influencé la diversité actuelle des Soricidae. À un niveau taxonomique élevé, l'apparition de connexions de terre temporaires entre le Vieux et le Nouveau Monde au Miocène moyen ont mené à plusieurs colonisations indépendantes de l'Amérique par les Soricinae. Celles-ci ónt conduit à une diversification d'une tribu (Notiosoricini), ainsi que de genres (par ex: Cryptotis, Blarina) et d'un sous-genre (Otisorex) endémique au Néarctique. Dans le Vieux Monde, les barrières entre l'Afrique et Eurasie étaient plus perméables, menant à plusieurs échanges bidirectionnels de Crocidurinae. La diversification des clades principaux s'est produite au Miocène, certains clades étant endémiques d'Afrique ou d'Eurasie, tandis que d'autres se sont diversifiés à travers le Vieux Monde. À un niveau spécifique ou péri-spécifique, la fluctuation climatique du Pliocène et les glaciations du Pléistocène ont fortement divisé les populations dans tout le Paléarctique, menant à des entités génétiques distinctes. En Europe, les populations du groupe de C. suaveolens ont été divisées en une lignée Sud-Ouest et une Sud-Est, alors qu'au Proche-Orient et au Moyen-Orient, la diversité de clades est plus importante. En conclusion, mes études ont révélé que du Miocène à nos jours, la diversification des Soricidae a été provoquée par la colonisation de nouveaux habitats (dispersion), ainsi que par l'isolement des populations par diverses barrières (vicariance). Abstract The Soricidae is one of the largest mammalian families with more than 300 species described. It has been recently divided into three subfamilies, the Soricinae, which are distributed in the Holartic region, the Crocidurinae in Africa and Eurasia, and the Myosoricinae in Africa. The specific diversity of this family have led to multiple systematic interpretations and controversies between authors. Fortunately, today, cytotaxonomic, allozymic and molecular studies have permitted to clarify some uncertainties. Nevertheless, the Soricidae remains still poorly known. In this thesis, we aim at understanding with the use of mitochondrial and nuclear markers: (i) the taxonomic relationships at different hierarchical levels within Soricidae, i.e., between the subfamilies, tribes, and genera, as well as within two largely distributed species complexes, and within a European species, the goal being to establish congruence between the genetic data and traditional morphological interpretations; (ii) the biogeographic relationships, especially the potential origin of the different subfamilies, tribes, and genera, the number of transcontinental exchanges, as well as the phylogeographic structure at a (peri)-specific level, in order to establish the history of the genetic diversification of this family. The combined analyses of mitochondrial and nuclear DNA highlight for the first time a clear relationship between taxa at a high taxonomical level, permitting to distinguish the relationships between subfamilies, tribes, and genera. Although Myosorex formed a distinct monophyletic group, its definition as a distinct sub-family cannot be advocated. Thus, we propose to attribute a tribe level for this Glade (included within the Crocidurinae). Additionally, this combination of genes pleads in favour of the inclusion of the genus Anourosorex within the Soricinae and not in a basal position within the Soricidae. Within the Crocidurinae, Suncus appeared to be paraphyletic, and Diplomesodon should be considered from a genetic point of view as invalid, and is presently considered as Crocidura. At a lower taxonomic level, we showed the monophyly of two widely distributed species complexes, the C. suaveolens group and the C. olivieri group. Nevertheless within those, we showed major differences compared to morphological classification. For examples, C. sibirica revealed to not be a valid species, the molecular phylogenetic analyses failed to evidence genetical variations between it and samples of the type locality of C. suaveolens. In a biogeographic point of view, the climatic fluctuations and the tectonic plate activities of the last 20 Myr have strongly influenced the actual diversity of the family. At a high taxonomic level, the successive land bridge connections between the Old and the New World, which occurred during the Middle Miocene, have led to several independent colonisations of America by Soricinae, and a subsequent diversification of endemic Nearctic's tribe (Notiosoricini), genera (e.g. Cryptotis, Blaring) and sub-genus (Otisorex) within the Soricinae. Within the Old World, the barriers between Africa and Eurasia were more permeable, leading to several bidirectional exchanges within the Crocidurinae. The diversification of major clades occurred through the Miocene, some clades being endemic to Africa or Eurasia, whereas others diversified through the Old World. At a species level or a peri-specific level, the Pliocene climatic fluctuation and the Pleistocene glaciations have strongly divided the populations throughout the Palaearctic, leading to well defined genetic entities. In Europe, populations of the C. suaveolens group were split in a classical south-western and south-eastern lineage. In contrast, the Near East and the Middle East reveal many differentiated clades. In conclusion, our studies revealed that, from the Miocene to present, the diversification and speciation events within the Soricidae were caused by natural colonisation of new habitats (dispersion) and isolation of populations by various barriers (vicariance).

Contrasting spatio-temporal climatic niche dynamics during the eastern and western invasions of spotted knapweed in North America

Aim The spotted knapweed (Centaurea stoebe), a plant native to south-east and central Europe, is highly invasive in North America. We investigated the spatio-temporal climatic niche dynamics of the spotted knapweed in North America along two putative eastern and western invasion routes. We then considered the patterns observed in the light of historical, ecological and evolutionary factors. Location Europe and North America. Methods The niche characteristics of the east and west invasive populations of spotted knapweed in North America were determined from documented occurrences over 120 consecutive years (1890-2010). The 2.5 and 97.5 percentiles of values along temperature and precipitation gradients, as given by the two first axes of a principal component axis (PCA), were then calculated. We additionally measured the climatic dissimilarity between invaded and native niches using a multivariate environmental similarity surface (MESS) analysis. Results Along both invasion routes, the species established in regions with climatic conditions that were similar to those in the native range in Europe. An initial spread in ruderal habitats always preceded spread in (semi-)natural habitats. In the east, the niche gradually increased over time until it reached limits similar to the native niche. Conversely, in the west the niche abruptly expanded after an extended time lag into climates not occupied in the native range; only the native cold niche limit was conserved. Main conclusions Our study reveals that different niche dynamics have taken place during the eastern and western invasions. This pattern indicates different combinations of historical, ecological and evolutionary factors in the two ranges. We hypothesize that the lack of a well-developed transportation network in the west at the time of the introduction of spotted knapweed confined the species to a geographically and climatically isolated region. The invasion of dry rangelands may have been favoured during the agricultural transition in the 1930s by release from natural enemies, local adaptation and less competitive vegetation, but further experimental and molecular studies are needed to explain these contrasting niche patterns fully. Our study illustrates the need and benefit of applying large-scale, temporally explicit approaches to understanding biological invasions.

HAMAP in 2015: updates to the protein family classification and annotation system.

HAMAP (High-quality Automated and Manual Annotation of Proteins-available at http://hamap.expasy.org/) is a system for the automatic classification and annotation of protein sequences. HAMAP provides annotation of the same quality and detail as UniProtKB/Swiss-Prot, using manually curated profiles for protein sequence family classification and expert curated rules for functional annotation of family members. HAMAP data and tools are made available through our website and as part of the UniRule pipeline of UniProt, providing annotation for millions of unreviewed sequences of UniProtKB/TrEMBL. Here we report on the growth of HAMAP and updates to the HAMAP system since our last report in the NAR Database Issue of 2013. We continue to augment HAMAP with new family profiles and annotation rules as new protein families are characterized and annotated in UniProtKB/Swiss-Prot; the latest version of HAMAP (as of 3 September 2014) contains 1983 family classification profiles and 1998 annotation rules (up from 1780 and 1720). We demonstrate how the complex logic of HAMAP rules allows for precise annotation of individual functional variants within large homologous protein families. We also describe improvements to our web-based tool HAMAP-Scan which simplify the classification and annotation of sequences, and the incorporation of an improved sequence-profile search algorithm.

Climatic extremes improve predictions of spatial patterns of tree species.

Understanding niche evolution, dynamics, and the response of species to climate change requires knowledge of the determinants of the environmental niche and species range limits. Mean values of climatic variables are often used in such analyses. In contrast, the increasing frequency of climate extremes suggests the importance of understanding their additional influence on range limits. Here, we assess how measures representing climate extremes (i.e., interannual variability in climate parameters) explain and predict spatial patterns of 11 tree species in Switzerland. We find clear, although comparably small, improvement (+20% in adjusted D(2), +8% and +3% in cross-validated True Skill Statistic and area under the receiver operating characteristics curve values) in models that use measures of extremes in addition to means. The primary effect of including information on climate extremes is a correction of local overprediction and underprediction. Our results demonstrate that measures of climate extremes are important for understanding the climatic limits of tree species and assessing species niche characteristics. The inclusion of climate variability likely will improve models of species range limits under future conditions, where changes in mean climate and increased variability are expected.

Introducing a new MRI classification of lumbar spinal stenosis based on cross-sectional morphology of the dural sac : 37

Introduction: Measures of the degree of lumbar spinal stenosis (LSS) such as antero-posterior diameter of the canal, and dural sac cross sectional area vary, and do not correlate with symptoms or results of surgery. We created a grading system, comprised of seven categories, based on the morphology of the dural sac and its contents as seen on T2 axial images. The categories take into account the ratio of rootlet/ CSF content. Grade A indicates no significant compression, grade D is equivalent to a total myelograhic block. We compared this classification with commonly used criteria of severity of stenosis. Methods: Fifty T2 axial MRI images taken at disc level from 27 symptomatic LSS patients undergoing decompressive surgery were classified twice by two radiologists and three spinal surgeons working at different institutions and countries. Dural sac cross-sectional surface area and AP diameter of the canal were measured both at disc and pedicle level from DICOM images using OsiriX software. Intraand inter-observer reliability were assessed using Cohen's, Fleiss' kappa statistics, and t test. Results: For the morphological grading the average intra-and inter observer kappas were 0.76 and 0.69+, respectively, for physicians working in the study originating country. Combining all observers the kappa values were 0.57 ± 0.19. and 0.44 ± 0.19, respectively. AP diameter and dural sac cross-sectional area measurements showed no statistically significant differences between observers. No correlation between morphological grading and AP diameter or dural sac crosssectional areawas observed in 13 (26%) and 8 cases (16%), respectively. Discussion: The proposed morphological grading relies on the identification of the dural sac and CSF better seen on full MRI series. This was not available to the external observers, which might explain the lower overall kappa values. Since no specific measurement tools are needed the grading suits everyday clinical practice and favours communication of degree of stenosis between practising physicians. The absence of a strict correlation with the dural sac surface suggests that measuring the surface alone might be insufficient in defining LSS as it is essentially a mismatch between the spinal canal and its contents. This grading is now adopted in our unit and further studies concentrating on relation between morphology, clinical symptoms and surgical results are underway.

Event-related potential analyses via single-subject topographic classification.

Introduction: Responses to external stimuli are typically investigated by averaging peri-stimulus electroencephalography (EEG) epochs in order to derive event-related potentials (ERPs) across the electrode montage, under the assumption that signals that are related to the external stimulus are fixed in time across trials. We demonstrate the applicability of a single-trial model based on patterns of scalp topographies (De Lucia et al, 2007) that can be used for ERP analysis at the single-subject level. The model is able to classify new trials (or groups of trials) with minimal a priori hypotheses, using information derived from a training dataset. The features used for the classification (the topography of responses and their latency) can be neurophysiologically interpreted, because a difference in scalp topography indicates a different configuration of brain generators. An above chance classification accuracy on test datasets implicitly demonstrates the suitability of this model for EEG data. Methods: The data analyzed in this study were acquired from two separate visual evoked potential (VEP) experiments. The first entailed passive presentation of checkerboard stimuli to each of the four visual quadrants (hereafter, "Checkerboard Experiment") (Plomp et al, submitted). The second entailed active discrimination of novel versus repeated line drawings of common objects (hereafter, "Priming Experiment") (Murray et al, 2004). Four subjects per experiment were analyzed, using approx. 200 trials per experimental condition. These trials were randomly separated in training (90%) and testing (10%) datasets in 10 independent shuffles. In order to perform the ERP analysis we estimated the statistical distribution of voltage topographies by a Mixture of Gaussians (MofGs), which reduces our original dataset to a small number of representative voltage topographies. We then evaluated statistically the degree of presence of these template maps across trials and whether and when this was different across experimental conditions. Based on these differences, single-trials or sets of a few single-trials were classified as belonging to one or the other experimental condition. Classification performance was assessed using the Receiver Operating Characteristic (ROC) curve. Results: For the Checkerboard Experiment contrasts entailed left vs. right visual field presentations for upper and lower quadrants, separately. The average posterior probabilities, indicating the presence of the computed template maps in time and across trials revealed significant differences starting at ~60-70 ms post-stimulus. The average ROC curve area across all four subjects was 0.80 and 0.85 for upper and lower quadrants, respectively and was in all cases significantly higher than chance (unpaired t-test, p<0.0001). In the Priming Experiment, we contrasted initial versus repeated presentations of visual object stimuli. Their posterior probabilities revealed significant differences, which started at 250ms post-stimulus onset. The classification accuracy rates with single-trial test data were at chance level. We therefore considered sub-averages based on five single trials. We found that for three out of four subjects' classification rates were significantly above chance level (unpaired t-test, p<0.0001). Conclusions: The main advantage of the present approach is that it is based on topographic features that are readily interpretable along neurophysiologic lines. As these maps were previously normalized by the overall strength of the field potential on the scalp, a change in their presence across trials and between conditions forcibly reflects a change in the underlying generator configurations. The temporal periods of statistical difference between conditions were estimated for each training dataset for ten shuffles of the data. Across the ten shuffles and in both experiments, we observed a high level of consistency in the temporal periods over which the two conditions differed. With this method we are able to analyze ERPs at the single-subject level providing a novel tool to compare normal electrophysiological responses versus single cases that cannot be considered part of any cohort of subjects. This aspect promises to have a strong impact on both basic and clinical research.

Archaeospicularia, new radiolarian order; a new step for the classification of the Lower Paleozoic Radiolaria.

A new radiolarian order - Archaeospicularia - is proposed for some Lower Paleozoic radiolarians previously considered to belong to Spumellaria and to Collodaria. It is characterized by a globular shell made of several spicules which can be free, interlocked, or fused to formed a latticed wall. The present paper gives the definition of this order and proposes a first classification. It is supposed that the Archaeospicularia represents the oldest radiolarian group and that in the Lower Paleozoic it gave rise to the orders Entactinaria, Albaillellaria, and probably Spumellaria by the reduction of the number of initial spicules. The origin of this order and its relationships with other groups of organisms with siliceous skeletons are also briefly discussed. (C) 2000 Academie des sciences / Editions scientifiques et medicales Elsevier SAS.

The AO comprehensive classification of pediatric long-bone fractures: a web-based multicenter agreement study.

The first AO comprehensive pediatric long-bone fracture classification system has been proposed following a structured path of development and validation with experienced pediatric surgeons. A Web-based multicenter agreement study involving 70 surgeons in 15 clinics and 5 countries was conducted to assess the reliability and accuracy of this classification when used by a wide range of surgeons with various levels of experience. Training was provided at each clinic before the session. Using the Internet, participants could log in at any time and classify 275 supracondylar, radius, and tibia fractures at their own pace. The fracture diagnosis was made following the hierarchy of the classification system using both clinical terminology and codes. kappa coefficients for the single-surgeon diagnosis of epiphyseal, metaphyseal, or diaphyseal fracture type were 0.66, 0.80, and 0.91, respectively. Median accuracy estimates for each bone and type were all greater than 80%. Depending on their experience and specialization, surgeons greatly varied in their ability to classify fractures. Pediatric training and at least 2 years of experience were associated with significant improvement in reliability and accuracy. Kappa coefficients for diagnosis of specific child patterns were 0.51, 0.63, and 0.48 for epiphyseal, metaphyseal, and diaphyseal fractures, respectively. Identified reasons for coding discrepancies were related to different understandings of terminology and definitions, as well as poor quality radiographic images. Results supported some minor adjustments in the coding of fracture type and child patterns. This classification system received wide acceptance and support among the surgeons involved. As long as appropriate training could be performed, the system classification was reliable, especially among surgeons with a minimum of 2 years of clinical experience. We encourage broad-based consultation between surgeons' international societies and the use of this classification system in the context of clinical practice as well as prospectively for clinical studies.