Learning Methods and Algorithms for Semantic Text Classification across Multiple Domains

Information is nowadays a key resource: machine learning and data mining techniques have been developed to extract high-level information from great amounts of data. As most data comes in form of unstructured text in natural languages, research on text mining is currently very active and dealing with practical problems. Among these, text categorization deals with the automatic organization of large quantities of documents in priorly defined taxonomies of topic categories, possibly arranged in large hierarchies. In commonly proposed machine learning approaches, classifiers are automatically trained from pre-labeled documents: they can perform very accurate classification, but often require a consistent training set and notable computational effort. Methods for cross-domain text categorization have been proposed, allowing to leverage a set of labeled documents of one domain to classify those of another one. Most methods use advanced statistical techniques, usually involving tuning of parameters. A first contribution presented here is a method based on nearest centroid classification, where profiles of categories are generated from the known domain and then iteratively adapted to the unknown one. Despite being conceptually simple and having easily tuned parameters, this method achieves state-of-the-art accuracy in most benchmark datasets with fast running times. A second, deeper contribution involves the design of a domain-independent model to distinguish the degree and type of relatedness between arbitrary documents and topics, inferred from the different types of semantic relationships between respective representative words, identified by specific search algorithms. The application of this model is tested on both flat and hierarchical text categorization, where it potentially allows the efficient addition of new categories during classification. Results show that classification accuracy still requires improvements, but models generated from one domain are shown to be effectively able to be reused in a different one.

Antarctic cloud spectral emission from ground-based measurements, a focus on far infrared signatures

The present work belongs to the PRANA project, the first extensive field campaign of observation of atmospheric emission spectra covering the Far InfraRed spectral region, for more than two years. The principal deployed instrument is REFIR-PAD, a Fourier transform spectrometer used by us to study Antarctic cloud properties. A dataset covering the whole 2013 has been analyzed and, firstly, a selection of good quality spectra is performed, using, as thresholds, radiance values in few chosen spectral regions. These spectra are described in a synthetic way averaging radiances in selected intervals, converting them into BTs and finally considering the differences between each pair of them. A supervised feature selection algorithm is implemented with the purpose to select the features really informative about the presence, the phase and the type of cloud. Hence, training and test sets are collected, by means of Lidar quick-looks. The supervised classification step of the overall monthly datasets is performed using a SVM. On the base of this classification and with the help of Lidar observations, 29 non-precipitating ice cloud case studies are selected. A single spectrum, or at most an average over two or three spectra, is processed by means of the retrieval algorithm RT-RET, exploiting some main IR window channels, in order to extract cloud properties. Retrieved effective radii and optical depths are analyzed, to compare them with literature studies and to evaluate possible seasonal trends. Finally, retrieval output atmospheric profiles are used as inputs for simulations, assuming two different crystal habits, with the aim to examine our ability to reproduce radiances in the FIR. Substantial mis-estimations are found for FIR micro-windows: a high variability is observed in the spectral pattern of simulation deviations from measured spectra and an effort to link these deviations to cloud parameters has been performed.

Classification of thoracolumbar fractures and dislocations

A classification of injuries is necessary in order to develop a common language for treatment indications and outcomes. Several classification systems have been proposed, the most frequently used is the Denis classification. The problem of this classification system is that it is based on an assumption, which is anatomically unidentifiable: the so-called middle column. For this reason, few years ago, a group of spine surgeons has developed a new classification system, which is based on the severity of the injury. The severity is defined by the pathomorphological findings, the prognosis in terms of healing and potential of neurological damage. This classification is based on three major groups: A = isolated anterior column injuries by axial compression, B = disruption of the posterior ligament complex by distraction posteriorly, and group C = corresponding to group B but with rotation. There is an increasing severity from A to C, and within each group, the severity usually increases within the subgroups from .1, .2, .3. All these pathomorphologies are supported by a mechanism of injury, which is responsible for the extent of the injury. The type of injury with its groups and subgroups is able to suggest the treatment modality.

Validation of statistical shape model based reconstruction of the proximal femur - A morphology study

Seventeen bones (sixteen cadaveric bones and one plastic bone) were used to validate a method for reconstructing a surface model of the proximal femur from 2D X-ray radiographs and a statistical shape model that was constructed from thirty training surface models. Unlike previously introduced validation studies, where surface-based distance errors were used to evaluate the reconstruction accuracy, here we propose to use errors measured based on clinically relevant morphometric parameters. For this purpose, a program was developed to robustly extract those morphometric parameters from the thirty training surface models (training population), from the seventeen surface models reconstructed from X-ray radiographs, and from the seventeen ground truth surface models obtained either by a CT-scan reconstruction method or by a laser-scan reconstruction method. A statistical analysis was then performed to classify the seventeen test bones into two categories: normal cases and outliers. This classification step depends on the measured parameters of the particular test bone. In case all parameters of a test bone were covered by the training population's parameter ranges, this bone is classified as normal bone, otherwise as outlier bone. Our experimental results showed that statistically there was no significant difference between the morphometric parameters extracted from the reconstructed surface models of the normal cases and those extracted from the reconstructed surface models of the outliers. Therefore, our statistical shape model based reconstruction technique can be used to reconstruct not only the surface model of a normal bone but also that of an outlier bone.

Comprehensive analysis of CpG island methylator phenotype (CIMP)-high, -low, and -negative colorectal cancers based on protein marker expression and molecular features

CpG island methylator phenotype (CIMP) is being investigated for its role in the molecular and prognostic classification of colorectal cancer patients but is also emerging as a factor with the potential to influence clinical decision-making. We report a comprehensive analysis of clinico-pathological and molecular features (KRAS, BRAF and microsatellite instability, MSI) as well as of selected tumour- and host-related protein markers characterizing CIMP-high (CIMP-H), -low, and -negative colorectal cancers. Immunohistochemical analysis for 48 protein markers and molecular analysis of CIMP (CIMP-H: ? 4/5 methylated genes), MSI (MSI-H: ? 2 instable genes), KRAS, and BRAF were performed on 337 colorectal cancers. Simple and multiple regression analysis and receiver operating characteristic (ROC) curve analysis were performed. CIMP-H was found in 24 cases (7.1%) and linked (p < 0.0001) to more proximal tumour location, BRAF mutation, MSI-H, MGMT methylation (p = 0.022), advanced pT classification (p = 0.03), mucinous histology (p = 0.069), and less frequent KRAS mutation (p = 0.067) compared to CIMP-low or -negative cases. Of the 48 protein markers, decreased levels of RKIP (p = 0.0056), EphB2 (p = 0.0045), CK20 (p = 0.002), and Cdx2 (p < 0.0001) and increased numbers of CD8+ intra-epithelial lymphocytes (p < 0.0001) were related to CIMP-H, independently of MSI status. In addition to the expected clinico-pathological and molecular associations, CIMP-H colorectal cancers are characterized by a loss of protein markers associated with differentiation, and metastasis suppression, and have increased CD8+ T-lymphocytes regardless of MSI status. In particular, Cdx2 loss seems to strongly predict CIMP-H in both microsatellite-stable (MSS) and MSI-H colorectal cancers. Cdx2 is proposed as a surrogate marker for CIMP-H.

Mining for Class-Specific Motifs in Protein Sequence Classification

Background: In protein sequence classification, identification of the sequence motifs or n-grams that can precisely discriminate between classes is a more interesting scientific question than the classification itself. A number of classification methods aim at accurate classification but fail to explain which sequence features indeed contribute to the accuracy. We hypothesize that sequences in lower denominations (n-grams) can be used to explore the sequence landscape and to identify class-specific motifs that discriminate between classes during classification. Discriminative n-grams are short peptide sequences that are highly frequent in one class but are either minimally present or absent in other classes. In this study, we present a new substitution-based scoring function for identifying discriminative n-grams that are highly specific to a class. Results: We present a scoring function based on discriminative n-grams that can effectively discriminate between classes. The scoring function, initially, harvests the entire set of 4- to 8-grams from the protein sequences of different classes in the dataset. Similar n-grams of the same size are combined to form new n-grams, where the similarity is defined by positive amino acid substitution scores in the BLOSUM62 matrix. Substitution has resulted in a large increase in the number of discriminatory n-grams harvested. Due to the unbalanced nature of the dataset, the frequencies of the n-grams are normalized using a dampening factor, which gives more weightage to the n-grams that appear in fewer classes and vice-versa. After the n-grams are normalized, the scoring function identifies discriminative 4- to 8-grams for each class that are frequent enough to be above a selection threshold. By mapping these discriminative n-grams back to the protein sequences, we obtained contiguous n-grams that represent short class-specific motifs in protein sequences. Our method fared well compared to an existing motif finding method known as Wordspy. We have validated our enriched set of class-specific motifs against the functionally important motifs obtained from the NLSdb, Prosite and ELM databases. We demonstrate that this method is very generic; thus can be widely applied to detect class-specific motifs in many protein sequence classification tasks. Conclusion: The proposed scoring function and methodology is able to identify class-specific motifs using discriminative n-grams derived from the protein sequences. The implementation of amino acid substitution scores for similarity detection, and the dampening factor to normalize the unbalanced datasets have significant effect on the performance of the scoring function. Our multipronged validation tests demonstrate that this method can detect class-specific motifs from a wide variety of protein sequence classes with a potential application to detecting proteome-specific motifs of different organisms.

Urban Expansion and its impact on urban agriculture - remote sensing based change analysis of Kizinga and Mzinga Valley - Dar Es Salaam, Tanzania

Urban agriculture is a phenomenon that can be observed world-wide, particularly in cities of devel- oping countries. It is contributing significantly to food security and food safety and has sustained livelihood of the urban and peri-urban low income dwe llers in developing countries for many years. Population increase due to rural-urban migration and natural - formal as well as informal - urbani- sation are competing with urban farming for available space and scarce water resources. A mul- titemporal and multisensoral urban change analysis over the period of 25 years (1982-2007) was performed in order to measure and visualise the urban expansion along the Kizinga and Mzinga valley in the south of Dar Es Salaam. Airphotos and VHR satellite data were analysed by using a combination of a composition of anisotropic textural measures and spectral information. The study revealed that unplanned built-up area is expanding continuously, and vegetation covers and agricultural lands decline at a fast rate. The validation showed that the overall classification accuracy varied depending on the database. The extracted built-up areas were used for visual in- terpretation mapping purposes and served as information source for another research project. The maps visualise an urban congestion and expansion of nearly 18% of the total analysed area that had taken place in the Kizinga valley between 1982 and 2007. The same development can be ob- served in the less developed and more remote Mzinga valley between 1981 and 2002. Both areas underwent fast changes where land prices still tend to go up and an influx of people both from rural and urban areas continuously increase the density with the consequence of increasing multiple land use interests.

Interactions between selected photosensitizers and model membranes: an NMR classification

Membrane interactions of porphyrinic photosensitizers (PSs) are known to play a crucial role for PS efficiency in photodynamic therapy (PDT). In the current paper, the interactions between 15 different porphyrinic PSs with various hydrophilic/lipophilic properties and phospholipid bilayers were probed by NMR spectroscopy. Unilamellar vesicles consisting of dioleoyl-phosphatidyl-choline (DOPC) were used as membrane models. PS-membrane interactions were deduced from analysis of the main DOPC (1)H-NMR resonances (choline and lipid chain signals). Initial membrane adsorption of the PSs was indicated by induced changes to the DOPC choline signal, i.e. a split into inner and outer choline peaks. Based on this parameter, the PSs could be classified into two groups, Type-A PSs causing a split and the Type-B PSs causing no split. A further classification into two subgroups each, A1, A2 and B1, B2 was based on the observed time-dependent changes of the main DOPC NMR signals following initial PS adsorption. Four different time-correlated patterns were found indicating different levels and rates of PS penetration into the hydrophobic membrane interior. The type of interaction was mainly affected by the amphiphilicity and the overall lipophilicity of the applied PS structures. In conclusion, the NMR data provided valuable structural and dynamic insights into the PS-membrane interactions which allow deriving the structural constraints for high membrane affinity and high membrane penetration of a given PS. (C) 2011 Elsevier B.V. All rights reserved.

Classification of organisms previously reported as the SP and Stewart-Letscher groups, with descriptions of Necropsobacter gen. nov. and of Necropsobacter rosorum sp. nov. for organisms of the SP group

To allow classification of bacteria previously reported as the SP group and the Stewart-Letscher group, 35 isolates from rodents (21), rabbits (eight), a dog and humans (five) were phenotypically and genotypically characterized. Comparison of partial rpoB sequences showed that 34 of the isolates were closely related, demonstrating at least 97.4 % similarity. 16S rRNA gene sequence comparison of 20 selected isolates confirmed the monophyly of the SP group and revealed 98.5 %-100 % similarity between isolates. A blast search using the 16S rRNA gene sequences showed that the highest similarity outside the SP group was 95.5 % to an unclassified rat isolate. The single strain, P625, representing the Stewart-Letscher group showed the highest 16S rRNA gene similarity (94.9-95.5 %) to members of the SP group. recN gene sequence analysis of 11 representative strains resulted in similarities of 97-100 % among the SP group strains, which showed 80 % sequence similarity to the Stewart-Letscher group strain. Sequence similarity values based on the recN gene, indicative for whole genome similarity, showed the SP group being clearly separated from established genera, whereas the Stewart-Letscher group strain was associated with the SP group. A new genus, Necropsobacter gen. nov., with only one species, Necropsobacter rosorum sp. nov., is proposed to include all members of the SP group. The new genus can be separated from existing genera of the family Pasteurellaceae by at least three phenotypic characters. The most characteristic properties of the new genus are that haemolysis is not observed on bovine blood agar, positive reactions are observed in the porphyrin test, acid is produced from (+)-L-arabinose, (+)-D-xylose, dulcitol, (+)-D-galactose, (+)-D-mannose, maltose and melibiose, and negative reactions are observed for symbiotic growth, urease, ornithine decarboxylase and indole. Previous publications have documented that both ubiquinones and demethylmenaquinone were produced by the proposed type strain of the new genus, Michel A/76(T), and that the major polyamine of representative strains (type strain not included) of the genus is 1,3-diaminopropane, spermidine is present in moderate amounts and putrescine and spermine are detectable only in minor amounts. The major fatty acids of strain Michel A/76(T) are C(14 : 0), C(16 : 0), C(16:1)omega7c and summed feature C(14 : 0) 3-OH/iso-C(16 : 1) I. This fatty acid profile is typical for members of the family Pasteurellaceae. The G+C content of DNA of strain Michel A/76(T) was estimated to be 52.5 mol% in a previous investigation. The type strain is P709(T) ( = Michel A/76(T) = CCUG 28028(T) = CIP 110147(T) = CCM 7802(T)).

Stratification and Prognostic Relevance of Jass's Molecular Classification of Colorectal Cancer

Background: The current proposed model of colorectal tumorigenesis is based primarily on CpG island methylator phenotype (CIMP), microsatellite instability (MSI), KRAS, BRAF, and methylation status of 0-6-Methylguanine DNA Methyltransferase (MGMT) and classifies tumors into five subgroups. The aim of this study is to validate this molecular classification and test its prognostic relevance. Methods: Three hundred two patients were included in this study. Molecular analysis was performed for five CIMP-related promoters (CRABP1, MLH1, p16INK4a, CACNA1G, NEUROG1), MGMT, MSI, KRAS, and BRAF. Methylation in at least 4 promoters or in one to three promoters was considered CIMP-high and CIMP-low (CIMP-H/L), respectively. Results: CIMP-H, CIMP-L, and CIMP-negative were found in 7.1, 43, and 49.9% cases, respectively. One hundred twenty-three tumors (41%) could not be classified into any one of the proposed molecular subgroups, including 107 CIMP-L, 14 CIMP-H, and two CIMP-negative cases. The 10 year survival rate for CIMP-high patients [22.6% (95%CI: 7-43)] was significantly lower than for CIMP-L or CIMP-negative (p = 0.0295). Only the combined analysis of BRAF and CIMP (negative versus L/H) led to distinct prognostic subgroups. Conclusion: Although CIMP status has an effect on outcome, our results underline the need for standardized definitions of low- and high-level CIMP, which clearly hinders an effective prognostic and molecular classification of colorectal cancer.

Contemporary consensus proposal on criteria and classification of eosinophilic disorders and related syndromes

Eosinophilia is an important indicator of various neoplastic and nonneoplastic conditions. Depending on the underlying disease and mechanisms, eosinophil infiltration can lead to organ dysfunction, clinical symptoms, or both. During the past 2 decades, several different classifications of eosinophilic disorders and related syndromes have been proposed in various fields of medicine. Although criteria and definitions are, in part, overlapping, no global consensus has been presented to date. The Year 2011 Working Conference on Eosinophil Disorders and Syndromes was organized to update and refine the criteria and definitions for eosinophilic disorders and to merge prior classifications in a contemporary multidisciplinary schema. A panel of experts from the fields of immunology, allergy, hematology, and pathology contributed to this project. The expert group agreed on unifying terminologies and criteria and a classification that delineates various forms of hypereosinophilia, including primary and secondary variants based on specific hematologic and immunologic conditions, and various forms of the hypereosinophilic syndrome. For patients in whom no underlying disease or hypereosinophilic syndrome is found, the term hypereosinophilia of undetermined significance is introduced. The proposed novel criteria, definitions, and terminologies should assist in daily practice, as well as in the preparation and conduct of clinical trials.

Time course based artifact identification for independent components of resting-state FMRI

In functional magnetic resonance imaging (fMRI) coherent oscillations of the blood oxygen level-dependent (BOLD) signal can be detected. These arise when brain regions respond to external stimuli or are activated by tasks. The same networks have been characterized during wakeful rest when functional connectivity of the human brain is organized in generic resting-state networks (RSN). Alterations of RSN emerge as neurobiological markers of pathological conditions such as altered mental state. In single-subject fMRI data the coherent components can be identified by blind source separation of the pre-processed BOLD data using spatial independent component analysis (ICA) and related approaches. The resulting maps may represent physiological RSNs or may be due to various artifacts. In this methodological study, we propose a conceptually simple and fully automatic time course based filtering procedure to detect obvious artifacts in the ICA output for resting-state fMRI. The filter is trained on six and tested on 29 healthy subjects, yielding mean filter accuracy, sensitivity and specificity of 0.80, 0.82, and 0.75 in out-of-sample tests. To estimate the impact of clearly artifactual single-subject components on group resting-state studies we analyze unfiltered and filtered output with a second level ICA procedure. Although the automated filter does not reach performance values of visual analysis by human raters, we propose that resting-state compatible analysis of ICA time courses could be very useful to complement the existing map or task/event oriented artifact classification algorithms.

A mass spectrometric multicenter study supports classification of preeclampsia as heterogeneous disorder

The diagnostic value of affinity-based matrix-assisted laser desorption/ionization time-of-flight mass spectrometry analysis to distinguish preeclampsia (PE) from matched controls was tested in a multicenter setting.

Classification and selection of biomarkers in genomic data using LASSO

High-throughput gene expression technologies such as microarrays have been utilized in a variety of scientific applications. Most of the work has been on assessing univariate associations between gene expression with clinical outcome (variable selection) or on developing classification procedures with gene expression data (supervised learning). We consider a hybrid variable selection/classification approach that is based on linear combinations of the gene expression profiles that maximize an accuracy measure summarized using the receiver operating characteristic curve. Under a specific probability model, this leads to consideration of linear discriminant functions. We incorporate an automated variable selection approach using LASSO. An equivalence between LASSO estimation with support vector machines allows for model fitting using standard software. We apply the proposed method to simulated data as well as data from a recently published prostate cancer study.

The influence of intention on masked priming: A study with semantic classification of words

Unconscious perception is commonly described as a phenomenon that is not under intentional control and relies on automatic processes. We challenge this view by arguing that some automatic processes may indeed be under intentional control, which is implemented in task-sets that define how the task is to be performed. In consequence, those prime attributes that are relevant to the task will be most effective. To investigate this hypothesis, we used a paradigm which has been shown to yield reliable short-lived priming in tasks based on semantic classification of words. This type of study uses fast, well practised classification responses, whereby responses to targets are much less accurate if prime and target belong to a different category than if they belong to the same category. In three experiments, we investigated whether the intention to classify the same words with respect to different semantic categories had a differential effect on priming. The results suggest that this was indeed the case: Priming varied with the task in all experiments. However, although participants reported not seeing the primes, they were able to classify the primes better than chance using the classification task they had used before with the targets. When a lexical task was used for discrimination in experiment 4, masked primes could however not be discriminated. Also, priming was as pronounced when the primes were visible as when they were invisible. The pattern of results suggests that participants had intentional control on prime processing, even if they reported not seeing the primes.