A sequential algorithm for sparse support vector classifiers

Support vector machines (SVMs), though accurate, are not preferred in applications requiring high classification speed or when deployed in systems of limited computational resources, due to the large number of support vectors involved in the model. To overcome this problem we have devised a primal SVM method with the following properties: (1) it solves for the SVM representation without the need to invoke the representer theorem, (2) forward and backward selections are combined to approach the final globally optimal solution, and (3) a criterion is introduced for identification of support vectors leading to a much reduced support vector set. In addition to introducing this method the paper analyzes the complexity of the algorithm and presents test results on three public benchmark problems and a human activity recognition application. These applications demonstrate the effectiveness and efficiency of the proposed algorithm.


--------------------------------------------------------------------------------

Practical automated activity recognition using standard smartphones

This paper describes a simple application for mobile devices which automatically recognizes different physical activity. This application could be used to log exercise sessions for the purpose of aiding weight management or improving sporting performance. © 2012 IEEE.

Gender Classification via Lips: Static and Dynamic Features

Automatic gender classification has many security and commercial applications. Various modalities have been investigated for gender classification with face-based classification being the most popular. In some real-world scenarios the face may be partially occluded. In these circumstances a classification based on individual parts of the face known as local features must be adopted. We investigate gender classification using lip movements. We show for the first time that important gender specific information can be obtained from the way in which a person moves their lips during speech. Furthermore our study indicates that the lip dynamics during speech provide greater gender discriminative information than simply lip appearance. We also show that the lip dynamics and appearance contain complementary gender information such that a model which captures both traits gives the highest overall classification result. We use Discrete Cosine Transform based features and Gaussian Mixture Modelling to model lip appearance and dynamics and employ the XM2VTS database for our experiments. Our experiments show that a model which captures lip dynamics along with appearance can improve gender classification rates by between 16-21% compared to models of only lip appearance.

The neurodevelopmental progress of infants less than 33 weeks into adolescence

Background: Several studies have shown an increased incidence of neurodevelopmental impairment in very preterm survivors at school age compared with controls.

Aim: To compare findings in the same cohort at 8 years and 15 years.

Methods: A total of 151 of the 224 eligible infants born before 33 weeks of gestation from 1979 to 1982, and who were living in the UK, were assessed at 8 and 15 years. Items common to both assessments were compared to evaluate changes in neurodevelopmental function. The assessment included a structured neurological examination, psychometric tests using the WISC-R (in subjects born in 1981-82), a test of visuomotor integration (Beery), and a school questionnaire.

Results: There was a significant increase in the proportion of subjects classified as impaired with disability from 11% at 8 to 22% at 14-15 years of age. The proportion of subjects classified as impaired without disability increased from 16% at 8 to 26% at 14-15 years of age. Full scale IQ decreased from 104 to 95 from childhood to adolescence, and more adolescents (24%) were requiring extra educational provision than they had at the age of 8 years (15%).

Conclusion: Results indicate that between the ages of 8 and 15 years in this cohort of very preterm survivors there is an apparent deterioration in neurodevelopmental outcome category, cognitive function, and extra educational support. It is not clear whether this represents a genuine deterioration in neurocognitive function or whether it represents the expression of pre-existing cerebral pathology in an increasingly complex environment.

Effects of very low birthweight on brain structure in adulthood

Very-low-birthweight (VLBW) individuals are at high risk of brain injury in the perinatal period. We wished to determine how such early brain lesions affect brain structure in adulthood. Thirty-two VLBW adults (20 female, 12 male) and, 18 term, normal birthweight sibling control individuals (nine female, nine male) underwent structural MRI at a mean age of 23 years 4 months (range 17 to 33 years; SD 3.4). Images were analyzed using an automated tissue segmentation algorithm in order to estimate whole brain tissue class volumes in native space. Images were then warped to a template image in standard space. There was no significant between-group difference in whole brain, greymatter, white matter, or total cerebral spinal fluid (CSF) volumes. However, lateral ventricular volume was significantly increased by 41% in those with VLBW. The ratio of grey to white matter was also significantly increased (by 10%) in those with VLBW. Group comparison maps showed widespread changes in the distribution of grey and white matter, and relative excess of ventricular CSF, in the brains of VLBW individuals. Increased ventricular volume predicted decreased grey matter in subcortical nuclei and limbic cortical structures, and decreased periventricular white matter. We conclude that these diffuse abnormalities of grey and white matter are a consequence,of the interaction of perinatal brain injury and ongoing neurodevelopmental processes.

Neuro developmental status at 1 year predicts neuropsychiatric outcome at 14-15 years of age in very preterm infants

Background: Neurodevelopmental and behavioural problems have been repeatedly reported in very preterm. survivors, often showing themselves later in childhood as poor school performance. Early identification of problems would mean that appropriate remedial therapy can be implemented. We have previously shown that neurodevelopmental status at 1 year was predictive of outcome at 8 years in a cohort of preterm. infants. The aim of this paper was to see if neurodevelopmental outcome in adolescence could be predicted by assessment by 1 year in the same cohort of pretem infants. Study design: Prospective cohort study. Subjects: 150 adolescents, born before 33 weeks gestation. Outcome measures: Neurological examination, developmental quotient, vision and hearing by 1 year. At 14-15 years, neurological examination, school performance questionnaire, Schonnell test of reading age, a premorbid adjustment score, Rutter behavioural score and for those born from 1981, cognitive tests (WISC-R). Results: A highly significant relationship existed between neurological status by 1 year and the need for extra educational provision, overall neurodevelopmental status, cognitive function in those that had their IQs measured and premorbid adjustment score of prepsychotic symptoms in adolescence. However, status at 1 year was not predictive of adolescent reading age or behavioural score. Conclusions: Neurodevelopmental assessment at 1 year ispredictive of school performance and outcome in the adolescent period. (C) 2001 Elsevier Science Ireland Ltd. All rights reserved.

Corpus callosum size and very preterm birth: relationship to neuropsychological outcome

Thinning of the corpus callosum (CC) is often observed in individuals who were born very preterm. Damage to the CC during neurodevelopment may be associated with poor neuropsychological performance. This study aimed to explore any evidence of CC pathology in adolescents aged 14-15 years who were born very preterm, and to investigate the relationship between CC areas and verbal skills. Seventy-two individuals born before 33 weeks of gestation and 51 age- and sex-matched full-term controls received structural MRI and neuropsychological assessment. Total CC area in very preterm adolescents was 7.5% smaller than in controls, after adjusting for total white matter volume (P=0.015). The absolute size of callosal subregions differed between preterm and fullterm adolescents: preterm individuals had a 14.7% decrease in posterior (P<0.0001) and an 11.6% decrease in mid-posterior CC quarters (P=0.029). Preterm individuals who had experienced periventricular haemorrhage and ventricular dilatation in the neonatal period showed the greatest decrease in CC area. In very preterm boys only, verbal IQ and verbal fluency scores were positively associated with total mid-sagittal CC size and midposterior surface area. These results suggest that very preterm birth adversely affects the development of the CC, particularly its posterior quarter, and this impairs verbal skills in boys.

Lateralisation of language function in young adults born very preterm

Objective: To explore, using functional magnetic resonance imaging (MRI), the functional organisation of phonological processing in young adults born very preterm.

Subjects: Six right handed male subjects with radiological evidence of thinning of the corpus callosum were selected from a cohort of very preterm subjects. Six normal right handed male volunteers acted as controls.

Method: Blood oxygenation level dependent contrast echoplanar images were acquired over five minutes at 1.5 T while subjects performed the tasks. During the ON condition, subjects were visually presented with pairs of non-words and asked to press a key when a pair of words rhymed (phonological processing). This task alternated with the OFF condition, which required subjects to make letter case judgments of visually presented pairs of consonant letter strings (orthographic processing). Generic brain activation maps were constructed from individual images by sinusoidal regression and non-parametric testing. Between group differences in the mean power of experimental response were identified on a voxel wise basis by analysis of variance.

Results: Compared with controls, the subjects with thinning of the corpus callosum showed significantly reduced power of response in the left hemisphere, including the peristriate cortex and the cerebellum, as well as in the right parietal association area. Significantly increased power of response was observed in the right precentral gyrus and the right supplementary motor area.

Conclusions: The data show evidence of increased frontal and decreased occipital activation in male subjects with neurodevelopmental thinning of the corpus callosum, which may be due to the operation of developmental compensatory mechanisms.

Neuropsychological outcome at adolescence of very preterm birth and its relation to brain structure

Neuropsychological outcome at 14 to 15 years of age of a cohort of 75 participants (39 male, 36 female) born at <33 weeks' gestation was investigated. Research was conducted parallel to a recent MRI study by Stewart and colleagues which reported that 55% of this cohort had evidence of brain abnormality. One aim of the studs was to compare neuropsychological function in those very preterm children with and without MRI abnormality. Compared to a control sample of term adolescents, very preterm participants had impairment only on a measure of word production. On measures of attention, memory, perceptual skill, and visuomotor and executive function, the adolescents born very preterm performed in the normal range, whether or not they had evidence of MRI abnormality. Our findings are encouraging as the neuropsychological consequences of damage to the very preterm brain, still evident on MRI at 14 to 15 years of age, appear to be minor.

Changes in lymphocyte subsets, interleukin 2, and soluble interleukin 2 receptor in old and very old age

In this study, the changes in some of the cellular components of the immune system and the activity of the cytokine interleukin 2, important for immune activation and lymphocyte proliferation, were measured in a large cross-sectional study of all age groups including octogenarian and nonagenarian subjects. In 206 apparently well community-living subjects, the absolute lymphocyte count and T and B cell numbers fell a little in old and very old subjects. Within the T cell compartment, helper/inducer CD4+ T cells, together with their subsets identified as 'naive' (CD4+/CD45RA+) and 'memory' (CD4+/CD45RO+) cells, also showed a decline with increased age. The suppressor/cytotoxic CD8+ subset showed no age-related change. The levels of the cytokine interleukin 2 were very low in octogenarian and nonagenarian subjects, while the soluble interleukin 2 receptor levels increased with increasing age. The interleukin 2 levels were associated with number and percentage of the 'memory' (CD4+/CD45RO+) subset of T cells which mediates the host response to previously met antigens. Since the interleukin 2 values were very low in the oldest groups and were associated with a reduced 'memory' (CD4+/CD45RO+) compartment, this suggests a possible mechanism of why the very elderly subject is more susceptible to morbidity and mortality from infectious or other agents.

Improved Detection of Mycobacterium bovis Infection in Bovine Lymph Node Tissue Using Immunomagnetic Separation (IMS)-Based Methods

Immunomagnetic separation (IMS) can selectively isolate and concentrate Mycobacterium bovis cells from lymph node tissue to facilitate subsequent detection by PCR (IMS-PCR) or culture (IMS-MGIT). This study describes application of these novel IMS-based methods to test for M. bovis in a survey of 280 bovine lymph nodes (206 visibly lesioned (VL), 74 non-visibly lesioned (NVL)) collected at slaughter as part of the Northern Ireland bovine TB eradication programme. Their performance was evaluated relative to culture. Overall, 174 (62.1%) lymph node samples tested positive by culture, 162 (57.8%) by IMS-PCR (targeting IS6110), and 196 (70.0%) by IMS-MGIT culture. Twelve (6.9%) of the 174 culture positive lymph node samples were not detected by either of the IMS-based methods. However, an additional 78 M. bovis positive lymph node samples (26 (12.6%) VL and 54 (73.0%) NVL) were detected by the IMS-based methods and not by culture. When low numbers of viable M. bovis are present in lymph nodes (e.g. in NVLs of skin test reactor cattle) decontamination prior to culture may adversely affect viability, leading to false negative culture results. In contrast, IMS specifically captures whole M. bovis cells (live, dead or potentially dormant) which are not subject to any deleterious treatment before detection by PCR or MGIT culture. During this study only 2.7% of NVL lymph nodes tested culture positive, whereas 73% of the same samples tested M. bovis positive by the IMS-based tests. Results clearly demonstrate that not only are the IMS-based methods more rapid but they have greater detection sensitivity than the culture approach currently used for the detection of M. bovis infection in cattle.. Adoption of the IMS-based methods for lymph node testing would have the potential to improve M. bovis detection in clinical samples.

Production and evaluation of the utility of novel phage display-derived peptide ligands to Salmonella spp. for magnetic separation

Aims: The objectives of this study were to produce Salmonella-specific peptide ligands by phage display biopanning and evaluate their use for magnetic separation (MS).
Methods and Results: Four phage display biopanning rounds were performed and the peptides expressed by the two most Salmonella-specific (on the basis of phage binding ELISA results) phage clones, MSal020401 and MSal020417, were chemically synthesized and coupled to MyOne™ tosylactivated Dynabeads®. Peptide capture capability for whole Salmonella cells from non-enriched broth cultures was quantified by MS + plate counts and MS + Greenlight™ detection, and compared to capture capability of anti-Salmonella (antibody-coated) Dynabeads®. MS + Greenlight™ gave a more comprehensive picture of capture capability than MS + plate counts and showed that Peptide MSal020417-coated beads exhibited at least similar, if not better, capture capability to anti-Salmonella Dynabeads® (mean capture values of 36.0 ± 18.2 % and 31.2 ± 20.1 %, respectively, over Salmonella spp. concentration range 3 x 101 - 3 x 106 cfu ml-1) with minimal cross-reactivity (= 1.9 %) to three other foodborne bacteria.
Conclusions: One of the phage display-derived peptide ligands was demonstrated by MS + Greenlight™ to be a viable antibody-alternative for MS of Salmonella spp.
Significance and Impact of Study: This study demonstrates an antibody-free approach to Salmonella detection and opens substantial possibilities for more rapid tests for this bacterium.

Investigation of the Human Brain Metabolome to Identify Potential Markers for Early Diagnosis and Therapeutic Targets of Alzheimer’s Disease

A study combining high resolution mass spectrometry (liquid chromatography-quadrupole time-of-flight-mass spectrometry, UPLC-QTof-MS) and chemometrics for the analysis of post-mortem brain tissue from subjects with Alzheimer’s disease (AD) (n = 15) and healthy age-matched controls (n = 15) was undertaken. The huge potential of this metabolomics approach for distinguishing AD cases is underlined by the correct prediction of disease status in 94–97% of cases. Predictive power was confirmed in a blind test set of 60 samples, reaching 100% diagnostic accuracy. The approach also indicated compounds significantly altered in concentration following the onset of human AD. Using orthogonal partial least-squares discriminant analysis (OPLS-DA), a multivariate model was created for both modes of acquisition explaining the maximum amount of variation between sample groups (Positive Mode-R2 = 97%; Q2 = 93%; root mean squared error of validation (RMSEV) = 13%; Negative Mode-R2 = 99%; Q2 = 92%; RMSEV = 15%). In brain extracts, 1264 and 1457 ions of interest were detected for the different modes of acquisition (positive and negative, respectively). Incorporation of gender into the model increased predictive accuracy and decreased RMSEV values. High resolution UPLC-QTof-MS has not previously been employed to biochemically profile post-mortem brain tissue, and the novel methods described and validated herein prove its potential for making new discoveries related to the etiology, pathophysiology, and treatment of degenerative brain disorders.

A Data Model for Content Modelling of Temporal Media

This paper describes a data model for content representation of temporal media in an IP based sensor network. The model is formed by introducing the idea of semantic-role from linguistics into the underlying concepts of formal event representation with the aim of developing a common event model. The architecture of a prototype system for a multi camera surveillance system, based on the proposed model is described. The important aspects of the proposed model are its expressiveness, its ability to model content of temporal media, and its suitability for use with a natural language interface. It also provides a platform for temporal information fusion, as well as organizing sensor annotations by help of ontologies.