Clustering in non-parametric multivariate analyses.

Non-parametric multivariate analyses of complex ecological datasets are widely used. Following appropriate pre-treatment of the data inter-sample resemblances are calculated using appropriate measures. Ordination and clustering derived from these resemblances are used to visualise relationships among samples (or variables). Hierarchical agglomerative clustering with group-average (UPGMA) linkage is often the clustering method chosen. Using an example dataset of zooplankton densities from the Bristol Channel and Severn Estuary, UK, a range of existing and new clustering methods are applied and the results compared. Although the examples focus on analysis of samples, the methods may also be applied to species analysis. Dendrograms derived by hierarchical clustering are compared using cophenetic correlations, which are also used to determine optimum  in flexible beta clustering. A plot of cophenetic correlation against original dissimilarities reveals that a tree may be a poor representation of the full multivariate information. UNCTREE is an unconstrained binary divisive clustering algorithm in which values of the ANOSIM R statistic are used to determine (binary) splits in the data, to form a dendrogram. A form of flat clustering, k-R clustering, uses a combination of ANOSIM R and Similarity Profiles (SIMPROF) analyses to determine the optimum value of k, the number of groups into which samples should be clustered, and the sample membership of the groups. Robust outcomes from the application of such a range of differing techniques to the same resemblance matrix, as here, result in greater confidence in the validity of a clustering approach.

Clustering in non-parametric multivariate analyses.

Non-parametric multivariate analyses of complex ecological datasets are widely used. Following appropriate pre-treatment of the data inter-sample resemblances are calculated using appropriate measures. Ordination and clustering derived from these resemblances are used to visualise relationships among samples (or variables). Hierarchical agglomerative clustering with group-average (UPGMA) linkage is often the clustering method chosen. Using an example dataset of zooplankton densities from the Bristol Channel and Severn Estuary, UK, a range of existing and new clustering methods are applied and the results compared. Although the examples focus on analysis of samples, the methods may also be applied to species analysis. Dendrograms derived by hierarchical clustering are compared using cophenetic correlations, which are also used to determine optimum  in flexible beta clustering. A plot of cophenetic correlation against original dissimilarities reveals that a tree may be a poor representation of the full multivariate information. UNCTREE is an unconstrained binary divisive clustering algorithm in which values of the ANOSIM R statistic are used to determine (binary) splits in the data, to form a dendrogram. A form of flat clustering, k-R clustering, uses a combination of ANOSIM R and Similarity Profiles (SIMPROF) analyses to determine the optimum value of k, the number of groups into which samples should be clustered, and the sample membership of the groups. Robust outcomes from the application of such a range of differing techniques to the same resemblance matrix, as here, result in greater confidence in the validity of a clustering approach.

The Open Door Mission: Measuring and predicting outcomes of one community-based substance abuse treatment program

The objectives of this study were to identify and measure the average outcomes of the Open Door Mission's nine-month community-based substance abuse treatment program, identify predictors of successful outcomes, and make recommendations to the Open Door Mission for improving its treatment program.^ The Mission's program is exclusive to adult men who have limited financial resources: most of which were homeless or dependent on parents or other family members for basic living needs. Many, but not all, of these men are either chemically dependent or have a history of substance abuse.^ This study tracked a cohort of the Mission's graduates throughout this one-year study and identified various indicators of success at short-term intervals, which may be predictive of longer-term outcomes. We tracked various levels of 12-step program involvement, as well as other social and spiritual activities, such as church affiliation and recovery support.^ Twenty-four of the 66 subjects, or 36% met the Mission's requirements for success. Specific to this success criteria; Fifty-four, or 82% reported affiliation with a home church; Twenty-six, or 39% reported full-time employment; Sixty-one, or 92% did not report or were not identified as having any post-treatment arrests or incarceration, and; Forty, or 61% reported continuous abstinence from both drugs and alcohol.^ Five research-based hypotheses were developed and tested. The primary analysis tool was the web-based non-parametric dependency modeling tool, B-Course, which revealed some strong associations with certain variables, and helped the researchers generate and test several data-driven hypotheses. Full-time employment is the greatest predictor of abstinence: 95% of those who reported full time employment also reported continuous post-treatment abstinence, while 50% of those working part-time were abstinent and 29% of those with no employment were abstinent. Working with a 12-step sponsor, attending aftercare, and service with others were identified as predictors of abstinence.^ This study demonstrates that associations with abstinence and the ODM success criteria are not simply based on one social or behavioral factor. Rather, these relationships are interdependent, and show that abstinence is achieved and maintained through a combination of several 12-step recovery activities. This study used a simple assessment methodology, which demonstrated strong associations across variables and outcomes, which have practical applicability to the Open Door Mission for improving its treatment program. By leveraging the predictive capability of the various success determination methodologies discussed and developed throughout this study, we can identify accurate outcomes with both validity and reliability. This assessment instrument can also be used as an intervention that, if operationalized to the Mission’s clients during the primary treatment program, may measurably improve the effectiveness and outcomes of the Open Door Mission.^

The science of citizen science

Biodiversity citizen science projects are growing in number, size, and scope, and are gaining recognition as valuable data sources that build public engagement. Yet publication rates indicate that citizen science is still infrequently used as a primary tool for conservation research and the causes of this apparent disconnect have not been quantitatively evaluated. To uncover the barriers to the use of citizen science as a research tool, we surveyed professional biodiversity scientists (n = 423) and citizen science project managers (n = 125). We conducted three analyses using non-parametric recursive modeling (random forest), using questions that addressed: scientists' perceptions and preferences regarding citizen science, scientists' requirements for their own data, and the actual practices of citizen science projects. For all three analyses we identified the most important factors that influence the probability of publication using citizen science data. Four general barriers emerged: a narrow awareness among scientists of citizen science projects that match their needs; the fact that not all biodiversity science is well-suited for citizen science; inconsistency in data quality across citizen science projects; and bias among scientists for certain data sources (institutions and ages/education levels of data collectors). Notably, we find limited evidence to suggest a relationship between citizen science projects that satisfy scientists' biases and data quality or probability of publication. These results illuminate the need for greater visibility of citizen science practices with respect to the requirements of biodiversity science and show that addressing bias among scientists could improve application of citizen science in conservation.

Applying quantile regression for modeling equivalent property damage only crashes to identify accident blackspots

Hot spot identification (HSID) aims to identify potential sites—roadway segments, intersections, crosswalks, interchanges, ramps, etc.—with disproportionately high crash risk relative to similar sites. An inefficient HSID methodology might result in either identifying a safe site as high risk (false positive) or a high risk site as safe (false negative), and consequently lead to the misuse the available public funds, to poor investment decisions, and to inefficient risk management practice. Current HSID methods suffer from issues like underreporting of minor injury and property damage only (PDO) crashes, challenges of accounting for crash severity into the methodology, and selection of a proper safety performance function to model crash data that is often heavily skewed by a preponderance of zeros. Addressing these challenges, this paper proposes a combination of a PDO equivalency calculation and quantile regression technique to identify hot spots in a transportation network. In particular, issues related to underreporting and crash severity are tackled by incorporating equivalent PDO crashes, whilst the concerns related to the non-count nature of equivalent PDO crashes and the skewness of crash data are addressed by the non-parametric quantile regression technique. The proposed method identifies covariate effects on various quantiles of a population, rather than the population mean like most methods in practice, which more closely corresponds with how black spots are identified in practice. The proposed methodology is illustrated using rural road segment data from Korea and compared against the traditional EB method with negative binomial regression. Application of a quantile regression model on equivalent PDO crashes enables identification of a set of high-risk sites that reflect the true safety costs to the society, simultaneously reduces the influence of under-reported PDO and minor injury crashes, and overcomes the limitation of traditional NB model in dealing with preponderance of zeros problem or right skewed dataset.

The reasons behind the progression in PISA scores: An education production function approach using semi-parametric techniques

Research Masters

Large-scale statistical modeling of motion patterns : a Bayesian nonparametric approach

We propose a novel framework for large-scale scene understanding in static camera surveillance. Our techniques combine fast rank-1 constrained robust PCA to compute the foreground, with non-parametric Bayesian models for inference. Clusters are extracted in foreground patterns using a joint multinomial+Gaussian Dirichlet process model (DPM). Since the multinomial distribution is normalized, the Gaussian mixture distinguishes between similar spatial patterns but different activity levels (eg. car vs bike). We propose a modification of the decayed MCMC technique for incremental inference, providing the ability to discover theoretically unlimited patterns in unbounded video streams. A promising by-product of our framework is online, abnormal activity detection. A benchmark video and two surveillance videos, with the longest being 140 hours long are used in our experiments. The patterns discovered are as informative as existing scene understanding algorithms. However, unlike existing work, we achieve near real-time execution and encouraging performance in abnormal activity detection.

Mechanical non-surgical treatment of peri-implantitis: a single-blinded randomized longitudinal clinical study. II. Microbiological results

BACKGROUND: Peri-implantitis is common in patients with dental implants. We performed a single-blinded longitudinal randomized study to assess the effects of mechanical debridement on the peri-implant microbiota in peri-implantitis lesions. MATERIALS AND METHODS: An expanded checkerboard DNA-DNA hybridization assay encompassing 79 different microorganisms was used to study bacterial counts before and during 6 months following mechanical treatment of peri-implantitis in 17 cases treated with curettes and 14 cases treated with an ultrasonic device. Statistics included non-parametric tests and GLM multivariate analysis with p<0001 indicating significance and 80% power. RESULTS: At selected implant test sites, the most prevalent bacteria were: Fusobacterium nucleatum sp., Staphylococci sp., Aggregatibacter actinomycetemcomitans, Helicobacter pylori, and Tannerella forsythia. 30 min. after treatment with curettes, A. actinomycetemcomitans (serotype a), Lactobacillus acidophilus, Streptococcus anginosus, and Veillonella parvula were found at lower counts (p<0.001). No such differences were found for implants treated with the ultrasonic device. Inconsistent changes occurred following the first week. No microbiological differences between baseline and 6-month samples were found for any species or between treatment study methods in peri-implantitis. CONCLUSIONS: Both methods failed to eliminate or reduce bacterial counts in peri-implantitis. No group differences were found in the ability to reduce the microbiota in peri-implantitis.

First trimester markers for pre-eclampsia: placental vs. non-placental protein serum levels

BACKGROUND/AIM: Parallel investigation, in a matched case-control study, of the association of different first-trimester markers with the risk of subsequent pre-eclampsia (PE). METHOD: The levels of different first trimester serum markers and fetal nuchal translucency thickness were compared between 52 cases of PE and 104 control women by non-parametric two-group comparisons and by calculating matched odds ratios. RESULTS: In univariable analysis increased concentrations of inhibin A and activin A were associated with subsequent PE (p < 0.02). Multivariable conditional logistic regression models revealed an association between increased risk of PE and increased inhibin A and translucency thickness and respectively reduced pregnancy-associated plasma protein A (PAPP-A) and placental lactogen . However, these associations varied with the gestational age at sample collection. For blood samples taken in pregnancy weeks 12 and 13 only, increased levels of activin A, inhibin A and nuchal translucency thickness, and lower levels of placenta growth factor and PAPP-A were associated with an increased risk of PE. CONCLUSIONS: Members of the inhibin family and to some extent PAPP-A and placental growth factor are superior to other serum markers, and the predictive value of these depends on the gestational age at blood sampling. The availability of a single, early pregnancy 'miracle' serum marker for PE risk assessment seems unlikely in the near future.

Modeling multilevel sleep transitional data via Poisson log-linear multilevel models

This paper proposes Poisson log-linear multilevel models to investigate population variability in sleep state transition rates. We specifically propose a Bayesian Poisson regression model that is more flexible, scalable to larger studies, and easily fit than other attempts in the literature. We further use hierarchical random effects to account for pairings of individuals and repeated measures within those individuals, as comparing diseased to non-diseased subjects while minimizing bias is of epidemiologic importance. We estimate essentially non-parametric piecewise constant hazards and smooth them, and allow for time varying covariates and segment of the night comparisons. The Bayesian Poisson regression is justified through a re-derivation of a classical algebraic likelihood equivalence of Poisson regression with a log(time) offset and survival regression assuming piecewise constant hazards. This relationship allows us to synthesize two methods currently used to analyze sleep transition phenomena: stratified multi-state proportional hazards models and log-linear models with GEE for transition counts. An example data set from the Sleep Heart Health Study is analyzed.

Mechanical non-surgical treatment of peri-implantitis: a double-blind randomized longitudinal clinical study. I: clinical results

BACKGROUND: Peri-implantitis is a frequent finding in patients with dental implants. The present study compared two non-surgical mechanical debridement methods of peri-implantitis. MATERIAL AND METHODS: Thirty-seven subjects (mean age 61.5; S.D+/-12.4), with one implant each, demonstrating peri-implantitis were randomized, and those treated either with titanium hand-instruments or with an ultrasonic device were enrolled. Data were obtained before treatment, and at 1, 3, and 6 months. Parametric and non-parametric statistics were used. RESULTS: Thirty-one subjects completed the study. The mean bone loss at implants in both groups was 1.5 mm (SD +/-1.2 mm). No group differences for plaque or gingival indices were found at any time point. Baseline and 6-month mean probing pocket depths (PPD) at implants were 5.1 and 4.9 mm (p=0.30) in both groups. Plaque scores at treated implants decreased from 73% to 53% (p<0.01). Bleeding scores also decreased (p<0.01), with no group differences. No differences in the total bacterial counts were found over time. Higher total bacterial counts were found immediately after treatment (p<0.01) and at 1 week for ultrasonic-treated implants (p<0.05). CONCLUSIONS: No group differences were found in the treatment outcomes. While plaque and bleeding scores improved, no effects on PPD were identified.

Segmentación de Secuencias de Vídeo: análisis y desarrollo de estrategias para la detección de cambios de toma y para la detección de objetos móviles

The important technological advances experienced along the last years have resulted in an important demand for new and efficient computer vision applications. On the one hand, the increasing use of video editing software has given rise to a necessity for faster and more efficient editing tools that, in a first step, perform a temporal segmentation in shots. On the other hand, the number of electronic devices with integrated cameras has grown enormously. These devices require new, fast, and efficient computer vision applications that include moving object detection strategies. In this dissertation, we propose a temporal segmentation strategy and several moving object detection strategies, which are suitable for the last generation of computer vision applications requiring both low computational cost and high quality results. First, a novel real-time high-quality shot detection strategy is proposed. While abrupt transitions are detected through a very fast pixel-based analysis, gradual transitions are obtained from an efficient edge-based analysis. Both analyses are reinforced with a motion analysis that allows to detect and discard false detections. This analysis is carried out exclusively over a reduced amount of candidate transitions, thus maintaining the computational requirements. On the other hand, a moving object detection strategy, which is based on the popular Mixture of Gaussians method, is proposed. This strategy, taking into account the recent history of each image pixel, adapts dynamically the amount of Gaussians that are required to model its variations. As a result, we improve significantly the computational efficiency with respect to other similar methods and, additionally, we reduce the influence of the used parameters in the results. Alternatively, in order to improve the quality of the results in complex scenarios containing dynamic backgrounds, we propose different non-parametric based moving object detection strategies that model both background and foreground. To obtain high quality results regardless of the characteristics of the analyzed sequence we dynamically estimate the most adequate bandwidth matrices for the kernels that are used in the background and foreground modeling. Moreover, the application of a particle filter allows to update the spatial information and provides a priori knowledge about the areas to analyze in the following images, enabling an important reduction in the computational requirements and improving the segmentation results. Additionally, we propose the use of an innovative combination of chromaticity and gradients that allows to reduce the influence of shadows and reflects in the detections.

Moving Object Detection for Real-Time High-Quality Lightweight Applications on Smart Cameras

Here, a novel and efficient strategy for moving object detection by non-parametric modeling on smart cameras is presented. Whereas the background is modeled using only color information, the foreground model combines color and spatial information. The application of a particle filter allows the update of the spatial information and provides a priori information about the areas to analyze in the following images, enabling an important reduction in the computational requirements and improving the segmentation results

Carotid artery intimal medial thickness, brachial artery flow-mediated vasodilation and cardiovascular risk factors in diabetic and non-diabetic indigenous Australians

Background: Indigenous Australians are at high risk for cardiovascular disease and type 2 diabetes. Carotid artery intimal medial thickness (CIMT) and brachial artery flow-mediated vasodilation (FMD) are ultrasound imaging based surrogate markers of cardiovascular risk. This study examines the relative contributions of traditional cardiovascular risk factors on CIMT and FMD in adult Indigenous Australians with and without type 2 diabetes mellitus. Method: One hundred and nineteen Indigenous Australians were recruited. Physical and biochemical markers of cardiovascular risk, together with CIMT and FMD were meausred for all subjects. Results: Fifty-three Indigenous Australians subjects (45%) had type 2 diabetes mellitus. There was a significantly greater mean CIMT in diabetic versus non-diabetic subjects (p = 0.049). In the non-diabetic group with non-parametric analyses, there were significant correlations between CIMT and: age (r = 0.64, p < 0.001), systolic blood pressure (r = 0.47, p < 0.001) and non-smokers (r = -0.30, p = 0.018). In the diabetic group, non-parametric analysis showed correlations between CIMT, age (r = 0.36, p = 0.009) and duration of diabetes (r = 0.30, p = 0.035) only. Adjusting forage, sex, smoking and history of cardiovascular disease, Hb(A1c) became the sole significant correlate of CIMT (r = 0.35,p = 0.01) in the diabetic group. In non-parametric analysis, age was the sole significant correlate of FMD (r = -0.31,p = 0.013), and only in non-diabetic subjects. Linear regression analysis showed significant associations between CIMT and age (t = 4.6,p < 0.001), systolic blood pressure (t = 2.6, p = 0.010) and Hb(A1c) (t = 2.6, p = 0.012), smoking (t = 2.1, p = 0.04) and fasting LDL-cholesterol (t = 2.1, p = 0.04). There were no significant associations between FMD and examined cardiovascular risk factors with linear regression analysis Conclusions: CIMT appears to be a useful surrogate marker of cardiovascular risk in this sample of Indigenous Australian subjects, correlating better than FMD with established cardiovascular risk factors. A lifestyle intervention programme may alleviate the burden of cardiovascular disease in Indigenous Australians by reducing central obesity, lowering blood pressure, correcting dyslipidaemia and improving glycaemic control. CIMT may prove to be a useful tool to assess efficacy of such an intervention programme. (c) 2004 Elsevier Ireland Ltd. All rights reserved.