Application of higher-order spectra for automated grading of diabetic maculopathy

Diabetic macular edema (DME) is one of the most common causes of visual loss among diabetes mellitus patients. Early detection and successive treatment may improve the visual acuity. DME is mainly graded into non-clinically significant macular edema (NCSME) and clinically significant macular edema according to the location of hard exudates in the macula region. DME can be identified by manual examination of fundus images. It is laborious and resource intensive. Hence, in this work, automated grading of DME is proposed using higher-order spectra (HOS) of Radon transform projections of the fundus images. We have used third-order cumulants and bispectrum magnitude, in this work, as features, and compared their performance. They can capture subtle changes in the fundus image. Spectral regression discriminant analysis (SRDA) reduces feature dimension, and minimum redundancy maximum relevance method is used to rank the significant SRDA components. Ranked features are fed to various supervised classifiers, viz. Naive Bayes, AdaBoost and support vector machine, to discriminate No DME, NCSME and clinically significant macular edema classes. The performance of our system is evaluated using the publicly available MESSIDOR dataset (300 images) and also verified with a local dataset (300 images). Our results show that HOS cumulants and bispectrum magnitude obtained an average accuracy of 95.56 and 94.39 % for MESSIDOR dataset and 95.93 and 93.33 % for local dataset, respectively.

Investigating in-domain data requirements for PLDA training

This paper analyzes the limitations upon the amount of in- domain (NIST SREs) data required for training a probabilistic linear discriminant analysis (PLDA) speaker verification system based on out-domain (Switchboard) total variability subspaces. By limiting the number of speakers, the number of sessions per speaker and the length of active speech per session available in the target domain for PLDA training, we investigated the relative effect of these three parameters on PLDA speaker verification performance in the NIST 2008 and NIST 2010 speaker recognition evaluation datasets. Experimental results indicate that while these parameters depend highly on each other, to beat out-domain PLDA training, more than 10 seconds of active speech should be available for at least 4 sessions/speaker for a minimum of 800 speakers. If further data is available, considerable improvement can be made over solely out-domain PLDA training.

Improving PLDA speaker verification using WMFD and linear-weighted approaches in limited microphone data conditions

This paper proposes the addition of a weighted median Fisher discriminator (WMFD) projection prior to length-normalised Gaussian probabilistic linear discriminant analysis (GPLDA) modelling in order to compensate the additional session variation. In limited microphone data conditions, a linear-weighted approach is introduced to increase the influence of microphone speech dataset. The linear-weighted WMFD-projected GPLDA system shows improvements in EER and DCF values over the pooled LDA- and WMFD-projected GPLDA systems in inter-view-interview condition as WMFD projection extracts more speaker discriminant information with limited number of sessions/ speaker data, and linear-weighted GPLDA approach estimates reliable model parameters with limited microphone data.

Whole-genome screening in ankylosing spondylitis: Evidence of non-MHC genetic-susceptibility loci

Ankylosing spondylitis (AS) is a common inflammatory arthritis predominantly affecting the axial skeleton. Susceptibility to the disease is thought to be oligogenic. To identify the genes involved, we have performed a genomewide scan in 185 families containing 255 affected sibling pairs. Two-point and multipoint nonparametric linkage analysis was performed. Regions were identified showing "suggestive" or stronger linkage with the disease on chromosomes 1p, 2q, 6p, 9q, 10q, 16q, and 19q. The MHC locus was identified as encoding the greatest component of susceptibility, with an overall LOD score of 15.6. The strongest non-MHC linkage lies on chromosome 16q (overall LOD score 4.7). These results strongly support the presence of non-MHC genetic-susceptibility factors in AS and point to their likely locations.

Marker selection by Akaike information criterion and Bayesian information criterion

We carried out a discriminant analysis with identity by descent (IBD) at each marker as inputs, and the sib pair type (affected-affected versus affected-unaffected) as the output. Using simple logistic regression for this discriminant analysis, we illustrate the importance of comparing models with different number of parameters. Such model comparisons are best carried out using either the Akaike information criterion (AIC) or the Bayesian information criterion (BIC). When AIC (or BIC) stepwise variable selection was applied to the German Asthma data set, a group of markers were selected which provide the best fit to the data (assuming an additive effect). Interestingly, these 25-26 markers were not identical to those with the highest (in magnitude) single-locus lod scores.

Acoustic identification of 12 species of echolocating bat by discriminant function analysis and artificial neural networks

We recorded echolocation calls from 14 sympatric species of bat in Britain. Once digitised, one temporal and four spectral features were measured from each call. The frequency-time course of each call was approximated by fitting eight mathematical functions, and the goodness of fit, represented by the mean-squared error, was calculated. Measurements were taken using an automated process that extracted a single call from background noise and measured all variables without intervention. Two species of Rhinolophus were easily identified from call duration and spectral measurements. For the remaining 12 species, discriminant function analysis and multilayer back-propagation perceptrons were used to classify calls to species level. Analyses were carried out with and without the inclusion of curve-fitting data to evaluate its usefulness in distinguishing among species. Discriminant function analysis achieved an overall correct classification rate of 79% with curve-fitting data included, while an artificial neural network achieved 87%. The removal of curve-fitting data improved the performance of the discriminant function analysis by 2 %, while the performance of a perceptron decreased by 2 %. However, an increase in correct identification rates when curve-fitting information was included was not found for all species. The use of a hierarchical classification system, whereby calls were first classified to genus level and then to species level, had little effect on correct classification rates by discriminant function analysis but did improve rates achieved by perceptrons. This is the first published study to use artificial neural networks to classify the echolocation calls of bats to species level. Our findings are discussed in terms of recent advances in recording and analysis technologies, and are related to factors causing convergence and divergence of echolocation call design in bats.

Acoustic identification of twelve species of echolocating bat by discriminant function analysis and artificial neural networks

We recorded echolocation calls from 14 sympatric species of bat in Britain. Once digitised, one temporal and four spectral features were measured from each call. The frequency-time course of each call was approximated by fitting eight mathematical functions, and the goodness of fit, represented by the mean-squared error, was calculated. Measurements were taken using an automated process that extracted a single call from background noise and measured all variables without intervention. Two species of Rhinolophus were easily identified from call duration and spectral measurements. For the remaining 12 species, discriminant function analysis and multilayer back-propagation perceptrons were used to classify calls to species level. Analyses were carried out with and without the inclusion of curve-fitting data to evaluate its usefulness in distinguishing among species. Discriminant function analysis achieved an overall correct classification rate of 79% with curve-fitting data included, while an artificial neural network achieved 87%. The removal of curve-fitting data improved the performance of the discriminant function analysis by 2 %, while the performance of a perceptron decreased by 2 %. However, an increase in correct identification rates when curve-fitting information was included was not found for all species. The use of a hierarchical classification system, whereby calls were first classified to genus level and then to species level, had little effect on correct classification rates by discriminant function analysis but did improve rates achieved by perceptrons. This is the first published study to use artificial neural networks to classify the echolocation calls of bats to species level. Our findings are discussed in terms of recent advances in recording and analysis technologies, and are related to factors causing convergence and divergence of echolocation call design in bats.

The Relative Efficiency of International, Domestic, and Budget Airlines : Nonparametric Evidence

This study determines whether the inclusion of low-cost airlines in a dataset of international and domestic airlines has an impact on the efficiency scores of so-called ‘prestigious’ purportedly ‘efficient’ airlines. This is because while many airline studies concern efficiency, none has truly included a combination of international, domestic and budget airlines. The present study employs the nonparametric technique of data envelopment analysis (DEA) to investigate the technical efficiency of 53 airlines in 2006. The findings reveal that the majority of budget airlines are efficient relative to their more prestigious counterparts. Moreover, most airlines identified as inefficient are so largely because of the overutilization of non-flight assets.

Efficiency and productivity of Singapore’s manufacturing sector 2001-2010: An analysis using Simar and Wilson’s (2007) bootstrapped truncated approach

This paper seeks to explain the lagging productivity in Singapore’s manufacturing noted in the statements of the Economic Strategies Committee Report 2010. Two methods are employed: the Malmquist productivity to measure total factor productivity change and Simar and Wilson’s (J Econ, 136:31–64, 2007) bootstrapped truncated regression approach. In the first stage, the nonparametric data envelopment analysis is used to measure technical efficiency. To quantify the economic drivers underlying inefficiencies, the second stage employs a bootstrapped truncated regression whereby bias-corrected efficiency estimates are regressed against explanatory variables. The findings reveal that growth in total factor productivity was attributed to efficiency change with no technical progress. Most industries were technically inefficient throughout the period except for ‘Pharmaceutical Products’. Sources of efficiency were attributed to quality of worker and flexible work arrangements while incessant use of foreign workers lowered efficiency.

Statistical and scientometric analysis of international research in geographical and environmental education

Certain statistic and scientometric features of articles published in the journal “International Research in Geographical and Environmental Education” are examined in this paper, for the period 1992-2009, by applying nonparametric statistics and Shannon’s entropy (diversity) formula. The main findings of this analysis are: a) after 2004 the research priorities of researchers in geographical and environmental education seem to have changed, b) “teacher education” has been the most recurrent theme throughout these 18 years, followed by “values & attitudes” and “inquiry & problem solving” c) the themes “GIS” and “Sustainability” were the most “stable” throughout the 18 years, meaning that they maintained their ranks as publication priorities more than other themes, d) citations of IRGEE increase annually, e) the average thematic diversity of articles published during the period 1992-2009 is 82.7% of the maximum thematic diversity (very high), meaning that the Journal has the capacity to attract a wide readership for the 10 themes it has successfully covered throughout the 18 years of its publication.

University Student Depression Inventory (USDI) : confirmatory factor analysis and review of psychometric properties

Background: The 30-item USDI is a self-report measure that assesses depressive symptoms among university students. It consists of three correlated three factors: Lethargy, Cognitive-Emotional and Academic motivation. The current research used confirmatory factor analysis to asses construct validity and determine whether the original factor structure would be replicated in a different sample. Psychometric properties were also examined. Method: Participants were 1148 students (mean age 22.84 years, SD = 6.85) across all faculties from a large Australian metropolitan university. Students completed a questionnaire comprising of the USDI, the Depression Anxiety Stress Scale (DASS) and Life Satisfaction Scale (LSS). Results: The three correlated factor model was shown to be an acceptable fit to the data, indicating sound construct validity. Internal consistency of the scale was also demonstrated to be sound, with high Cronbach Alpha values. Temporal stability of the scale was also shown to be strong through test-retest analysis. Finally, concurrent and discriminant validity was examined with correlations between the USDI and DASS subscales as well as the LSS, with sound results contributing to further support the construct validity of the scale. Cut-off points were also developed to aid total score interpretation. Limitations: Response rates are unclear. In addition, the representativeness of the sample could be improved potentially through targeted recruitment (i.e. reviewing the online sample statistics during data collection, examining the representativeness trends and addressing particular faculties within the university that were underrepresented). Conclusions: The USDI provides a valid and reliable method of assessing depressive symptoms found among university students.

Productivity and convergence in India : a state-level analysis

Total factor productivity plays an important role in the growth of the Indian economy. Using state-level data from 1993 to 2005 that were recently made available, we find widespread regional variation in productivity changes. In the years immediately following economic liberalization, productivity growth improved technical efficiency; however, in subsequent years, productivity growth was propelled by technological progress. We find a tendency toward convergence with regard to productivity growth among states; however, the states that were technically efficient when the economic reforms were instituted remained innovative in later years.

Total factor productivity growth and convergence in the petroleum industry : empirical analysis testing for convexity

While economic theory acknowledges that some features of technology (e.g., indivisibilities, economies of scale and specialization) can fundamentally violate the traditional convexity assumption, almost all empirical studies accept the convexity property on faith. In this contribution, we apply two alternative flexible production technologies to measure total factor productivity growth and test the significance of the convexity axiom using a nonparametric test of closeness between unknown distributions. Based on unique field level data on the petroleum industry, the empirical results reveal significant differences, indicating that this production technology is most likely non-convex. Furthermore, we also show the impact of convexity on answers to traditional convergence questions in the productivity growth literature.

Modal choice between air and rail : a social efficiency benchmarking analysis that considers CO2 emissions

The problem of modal choice between rail and air arises as public awareness of carbon dioxide (CO2) emissions by the transportation sector rises. In this paper, we answer this question quantitatively by performing an efficiency benchmarking analysis that takes into account life-cycle CO2 emission due to transport service provision. The paper employs nonparametric efficiency estimation methods, namely a slacks-based inefficiency measure, as well as a more conventional directional distance function approach. We apply them to a panel data set for three major railway companies and the aviation sector in Japan for the period from 1999 to 2007. Results shows that, contrary to the common argument, air transport can still be more socially efficient than rail transport, even when the environmental load due to CO2 emission is incorporated. This is due to the aviation sector's extremely low user cost, measured in terms of in-vehicle time. In other words, aviation is a necessary transportation mode for those with a very high willingness to pay for their time.

Nonparametric rank regression for analyzing water quality concentration data with multiple detection limits

Environmental data usually include measurements, such as water quality data, which fall below detection limits, because of limitations of the instruments or of certain analytical methods used. The fact that some responses are not detected needs to be properly taken into account in statistical analysis of such data. However, it is well-known that it is challenging to analyze a data set with detection limits, and we often have to rely on the traditional parametric methods or simple imputation methods. Distributional assumptions can lead to biased inference and justification of distributions is often not possible when the data are correlated and there is a large proportion of data below detection limits. The extent of bias is usually unknown. To draw valid conclusions and hence provide useful advice for environmental management authorities, it is essential to develop and apply an appropriate statistical methodology. This paper proposes rank-based procedures for analyzing non-normally distributed data collected at different sites over a period of time in the presence of multiple detection limits. To take account of temporal correlations within each site, we propose an optimal linear combination of estimating functions and apply the induced smoothing method to reduce the computational burden. Finally, we apply the proposed method to the water quality data collected at Susquehanna River Basin in United States of America, which dearly demonstrates the advantages of the rank regression models.