Edward Wegman - Academia.edu (original) (raw)
Papers by Edward Wegman
Brain and Behavior, 2016
We have explored the potential prefrontal hemodynamic biomarkers to characterize subjects with Tr... more We have explored the potential prefrontal hemodynamic biomarkers to characterize subjects with Traumatic Brain Injury (TBI) by employing the multivariate machine learning approach and introducing a novel task-related hemodynamic response detection followed by a heuristic search for optimum set of hemodynamic features. To achieve this goal, the hemodynamic response from a group of 31 healthy controls and 30 chronic TBI subjects were recorded as they performed a complexity task. To determine the optimum hemodynamic features, we considered 11 features and their combinations in characterizing TBI subjects. We investigated the significance of the features by utilizing a machine learning classification algorithm to score all the possible combinations of features according to their predictive power. The identified optimum feature elements resulted in classification accuracy, sensitivity, and specificity of 85%, 85%, and 84%, respectively. Classification improvement was achieved for TBI subject classification through feature combination. It signified the major advantage of the multivariate analysis over the commonly used univariate analysis suggesting that the features that are individually irrelevant in characterizing the data may become relevant when used in combination. We also conducted a spatio-temporal classification to identify regions within the prefrontal cortex (PFC) that contribute in distinguishing between TBI and healthy subjects. As expected, Brodmann areas (BA) 10 within the PFC were isolated as the region that healthy subjects (unlike subjects with TBI), showed major hemodynamic activity in response to the High Complexity task. Overall, our results indicate that identified temporal and spatio-temporal features from PFC's hemodynamic activity are promising biomarkers in classifying subjects with TBI.
Genetic and Evolutionary Computation Conference, 2000
We are developing GA-based tools for use in constructing high quality approximations to continuou... more We are developing GA-based tools for use in constructing high quality approximations to continuous functions. In this paper we report on a GA-based method for adaptively select-ing e ective interpolation points. We evalu-ate our approach on a variety of test func-tions, and we compare are results to more traditional approaches. The results are quite promising and suggest directions for further
Proceedings 3rd IEEE International Workshop on Distributed Interactive Simulation and Real-Time Applications, 2000
Simulation of particle systems is time consuming. However, many particle system applications requ... more Simulation of particle systems is time consuming. However, many particle system applications require fast interactive animations. For example, simulation of physically realistic complex dust behaviors is very useful in training, education, art, advertising, and entertainment. There are no published models for real-time simulation of dust behavior generated by a traveling vehicle. In this paper we use particle systems, computational fluid dynamics, and behavioral simulation techniques to simulate dust behavior in real time. First, we analyze the forces and factors that affect dust generation and the behavior after dust particles are generated. Then, we construct physically-based empirical models to generate dust particles and control the behavior accordingly. We further simplify the numerical calculations by dividing dust behavior into three stages, and establishing simplified particle system models for each stage. We employ motion blur, particle blending, texture mapping, and other computer graphics techniques to achieve the final results. Our contributions include constructing physically-based empirical models to generate dust behavior and achieving simulation of the behavior in real time
ABSTRACT Traditionally, parallel coordinates plots involve linear interpolations between coordina... more ABSTRACT Traditionally, parallel coordinates plots involve linear interpolations between coordinate axes. This is a powerful idea because of the projective duality, which allows interpretation of the duals of geometric structures in the parallel coor- dinates domain based on their structure in the Cartesian/Euclidean domain. Moreover, when combined with saturation brushing traditional parallel coor- dinates allows for visualization of rather massive data sets. In this paper we propose some alternates to the linear interpolations, which complement the tra- ditional linear interpolations and have some additional attractive features.
Journal of Statistical Planning and Inference, Oct 31, 1992
Communications in Statistics Theory and Methods, Dec 31, 2004
Http Dx Doi Org 10 1080 03610928008827984, Jun 27, 2007
ABSTRACT
ABSTRACT Abstract In this paper, we propose a nonparametric method for data quantization so as to... more ABSTRACT Abstract In this paper, we propose a nonparametric method for data quantization so as to reduce massive data sets to more manageable sizes. We investigate the probabilistic foundation and demonstrate statistical results for the quantization process. We discuss optimal geometric quantization procedures and discuss the computational and storage complexity of these procedures.
Computational Statistics Data Analysis, Sep 1, 2009
A methodology is presented to construct an expectation robust algorithm for principal component r... more A methodology is presented to construct an expectation robust algorithm for principal component regression. The presented method is the first multivariate regression method which can resist outliers and which can cope with missing elements in the data simultaneously. Simulations and an example illustrate the good statistical properties of the method.
Technometrics, Apr 9, 2012
Consider the first order autoregressive model, Xt = a(t)Xt–1, + ∊t, t = 1,2, … The mean and covar... more Consider the first order autoregressive model, Xt = a(t)Xt–1, + ∊t, t = 1,2, … The mean and covariance structure is computed under the hypotheses that Xt, is not second order stationary and that ∊t is an uncorrelated sequence of random variables with E∊t, = 0 and var (∊t) = τt , not constant. Two applications are mentioned.
The Journal of the Acoustical Society of America, 1991
ABSTRACT
Http Dx Doi Org 10 1080 01621459 1991 10475028, Feb 27, 2012
Biomedical Optics 2016, 2016
Brain and Behavior, 2016
We have explored the potential prefrontal hemodynamic biomarkers to characterize subjects with Tr... more We have explored the potential prefrontal hemodynamic biomarkers to characterize subjects with Traumatic Brain Injury (TBI) by employing the multivariate machine learning approach and introducing a novel task-related hemodynamic response detection followed by a heuristic search for optimum set of hemodynamic features. To achieve this goal, the hemodynamic response from a group of 31 healthy controls and 30 chronic TBI subjects were recorded as they performed a complexity task. To determine the optimum hemodynamic features, we considered 11 features and their combinations in characterizing TBI subjects. We investigated the significance of the features by utilizing a machine learning classification algorithm to score all the possible combinations of features according to their predictive power. The identified optimum feature elements resulted in classification accuracy, sensitivity, and specificity of 85%, 85%, and 84%, respectively. Classification improvement was achieved for TBI subject classification through feature combination. It signified the major advantage of the multivariate analysis over the commonly used univariate analysis suggesting that the features that are individually irrelevant in characterizing the data may become relevant when used in combination. We also conducted a spatio-temporal classification to identify regions within the prefrontal cortex (PFC) that contribute in distinguishing between TBI and healthy subjects. As expected, Brodmann areas (BA) 10 within the PFC were isolated as the region that healthy subjects (unlike subjects with TBI), showed major hemodynamic activity in response to the High Complexity task. Overall, our results indicate that identified temporal and spatio-temporal features from PFC's hemodynamic activity are promising biomarkers in classifying subjects with TBI.
Genetic and Evolutionary Computation Conference, 2000
We are developing GA-based tools for use in constructing high quality approximations to continuou... more We are developing GA-based tools for use in constructing high quality approximations to continuous functions. In this paper we report on a GA-based method for adaptively select-ing e ective interpolation points. We evalu-ate our approach on a variety of test func-tions, and we compare are results to more traditional approaches. The results are quite promising and suggest directions for further
Proceedings 3rd IEEE International Workshop on Distributed Interactive Simulation and Real-Time Applications, 2000
Simulation of particle systems is time consuming. However, many particle system applications requ... more Simulation of particle systems is time consuming. However, many particle system applications require fast interactive animations. For example, simulation of physically realistic complex dust behaviors is very useful in training, education, art, advertising, and entertainment. There are no published models for real-time simulation of dust behavior generated by a traveling vehicle. In this paper we use particle systems, computational fluid dynamics, and behavioral simulation techniques to simulate dust behavior in real time. First, we analyze the forces and factors that affect dust generation and the behavior after dust particles are generated. Then, we construct physically-based empirical models to generate dust particles and control the behavior accordingly. We further simplify the numerical calculations by dividing dust behavior into three stages, and establishing simplified particle system models for each stage. We employ motion blur, particle blending, texture mapping, and other computer graphics techniques to achieve the final results. Our contributions include constructing physically-based empirical models to generate dust behavior and achieving simulation of the behavior in real time
ABSTRACT Traditionally, parallel coordinates plots involve linear interpolations between coordina... more ABSTRACT Traditionally, parallel coordinates plots involve linear interpolations between coordinate axes. This is a powerful idea because of the projective duality, which allows interpretation of the duals of geometric structures in the parallel coor- dinates domain based on their structure in the Cartesian/Euclidean domain. Moreover, when combined with saturation brushing traditional parallel coor- dinates allows for visualization of rather massive data sets. In this paper we propose some alternates to the linear interpolations, which complement the tra- ditional linear interpolations and have some additional attractive features.
Journal of Statistical Planning and Inference, Oct 31, 1992
Communications in Statistics Theory and Methods, Dec 31, 2004
Http Dx Doi Org 10 1080 03610928008827984, Jun 27, 2007
ABSTRACT
ABSTRACT Abstract In this paper, we propose a nonparametric method for data quantization so as to... more ABSTRACT Abstract In this paper, we propose a nonparametric method for data quantization so as to reduce massive data sets to more manageable sizes. We investigate the probabilistic foundation and demonstrate statistical results for the quantization process. We discuss optimal geometric quantization procedures and discuss the computational and storage complexity of these procedures.
Computational Statistics Data Analysis, Sep 1, 2009
A methodology is presented to construct an expectation robust algorithm for principal component r... more A methodology is presented to construct an expectation robust algorithm for principal component regression. The presented method is the first multivariate regression method which can resist outliers and which can cope with missing elements in the data simultaneously. Simulations and an example illustrate the good statistical properties of the method.
Technometrics, Apr 9, 2012
Consider the first order autoregressive model, Xt = a(t)Xt–1, + ∊t, t = 1,2, … The mean and covar... more Consider the first order autoregressive model, Xt = a(t)Xt–1, + ∊t, t = 1,2, … The mean and covariance structure is computed under the hypotheses that Xt, is not second order stationary and that ∊t is an uncorrelated sequence of random variables with E∊t, = 0 and var (∊t) = τt , not constant. Two applications are mentioned.
The Journal of the Acoustical Society of America, 1991
ABSTRACT
Http Dx Doi Org 10 1080 01621459 1991 10475028, Feb 27, 2012
Biomedical Optics 2016, 2016