Given a decision that requires less than half a second for evaluating the characteristics of the ... more Given a decision that requires less than half a second for evaluating the characteristics of the incoming pitch and generating a motor response, hitting a baseball potentially requires unique perception-action coupling to achieve high performance. We designed a rapid perceptual decision-making experiment modeled as a Go/No-Go task yet tailored to reflect a real scenario confronted by a baseball hitter. For groups of experts (Division I baseball players) and novices (non-players), we recorded electroencephalography (EEG) while they performed the task. We analyzed evoked EEG single-trial variability, contingent negative variation (CNV), and pre-stimulus alpha power with respect to the expert vs. novice groups. We found strong evidence for differences in inhibitory processes between the two groups, specifically differential activity in supplementary motor areas (SMA), indicative of enhanced inhibitory control in the expert (baseball player) group. We also found selective activity in th...
Background / Purpose: Hitting a baseball is often described as the most difficult thing to do in ... more Background / Purpose: Hitting a baseball is often described as the most difficult thing to do in sports. This rapid decision requires the batter to make a judgment in a fraction of a second based largely on the trajectory and spin of the ball. There has been relatively little investigation into the neural correlates of this rapid decision making process. Recently, our group identified neural correlates of pitch identification using EEG. These results showed the temporal evolution of the decision making process, however, due to the poor spatial resolution of EEG we were unable to draw strong conclusions about which brain regions are recruited. In our current study, we use simultaneous EEG/fMRI to identify regions that are activated as well as their temporal order during the process of baseball pitch identification. Here we present preliminary results from both traditional BOLD fMRI and single trial variability analyses. Main conclusion: Our work represents a first attempt to identify...
We develop a probabilistic network model over image spaces and demonstrate its broad utility in m... more We develop a probabilistic network model over image spaces and demonstrate its broad utility in mammographic image analysis, particularly with respect to computer-aided diagnosis. The model employs a multi-scale pyramid decomposition to factor images across scale and a network of tree-structured hidden variables to capture long-range spatial dependencies. This factoring makes the computation of the density functions local and tractable. The result is a hierarchical mixture of conditional probabilities, similar to a hidden Markov model on a tree. The model parameters are found with maximum likelihood estimation using the expectation-maximization algorithm. The utility of the model is demonstrated for three applications: (1) detection of mammographic masses for computer-aided diagnosis; (2) qualitative assessment of model structure through mammographic synthesis; and (3) compression of mammographic regions of interest. 2003 Elsevier Science B.V. All rights reserved.
Objective: EEG and functional MRI potentially provide complementary information about the brain, ... more Objective: EEG and functional MRI potentially provide complementary information about the brain, yet high quality EEG is difficult to acquire during MR scanning because of mutual artifacts induced. Many groups have circumvented this difficulty by interleaving MRI volume acquisitions with EEG. However, for event-related EEG/fMRI studies, particularly those directed at single-trial analysis, high-density, high-quality, uninterrupted EEG is necessary. We have
EEG alpha-band activity is generally thought to represent an inhibitory state related to decrease... more EEG alpha-band activity is generally thought to represent an inhibitory state related to decreased attention and play a role in suppression of task-irrelevant stimulus processing, but a competing hypothesis suggests an active role in processing task-relevant information - one in which phase dynamics are involved. Here we used simultaneous EEG-fMRI and a whole-brain analysis to investigate the effects of prestimulus alpha activity on the event-related BOLD response during an auditory oddball task. We separately investigated the effects of the posterior alpha rhythm's power and phase on activity related to task-relevant stimulus processing and also investigated higher-level decision-related processing. We found stronger decision-related BOLD activity in areas late in the processing stream when subjects were in the high alpha power state prior to stimulus onset, but did not detect any effect in primary sensory regions. Our phase analysis revealed correlates in the bilateral thalamu...
First International IEEE EMBS Conference on Neural Engineering, 2003. Conference Proceedings., 2003
We describe a filter-based model of orientation processing in primary visual cortex (V1) and demo... more We describe a filter-based model of orientation processing in primary visual cortex (V1) and demonstrate that novelty in cortical "pinwheel" space can be used as a measure of perceptual salience. In the model, novelty is computed as the negative log likelihood of a pinwheel's activity relative to the population response. The population response is modeled using a mixture of Gaussians, enabling the representation of complex, multi-modal distributions. Hidden variables that are inferred in the mixture model can be viewed as grouping or "binding" pinwheels which have similar responses within the distribution space. Results are shown for several stimuli that illustrate well-known contextual effects related to perceptual salience, as well as results for a natural image.
First International IEEE EMBS Conference on Neural Engineering, 2003. Conference Proceedings., 2003
Blind source separation (BSS) has been proposed as a method to analyze multi-channel electroencep... more Blind source separation (BSS) has been proposed as a method to analyze multi-channel electroencephalography (EEG) data. A basic issue in applying BSS algorithms is the validity of the independence assumption. In this paper we investigate whether EEG can be considered to be a linear combination of independent sources. Linear BSS can be obtained under the assumptions of non-Gaussian, non-stationary, or non-white independent sources. If the linear independence hypothesis is violated these three different conditions will not necessarily lead to the same result. We show, using 64 channel EEG data, that different algorithms which incorporate the three different assumptions lead to the same results, thus supporting the linear independence hypothesis.
American Journal of Obstetrics and Gynecology, 1987
Maternal serum a-fetoprotein levels are higher in black women. Misinterpretation of maternal seru... more Maternal serum a-fetoprotein levels are higher in black women. Misinterpretation of maternal serum a-fetoprotein screening results can subject black gravid women to unnecessary invasive diagnostic procedures and their calculable risks. Screening errors for black women can result from the use of normative data bases established with maternal serum samples drawn from other racial groups or the use of such data bases in conjunction with a published correction factor. Because the incidence of open neural tube defects is lower for blacks than for others, excessive false positive results for blacks (estimated to be 8817 to 28,215 cases annually) would be a pernicious misapplication of maternal serum a-fetoprotein screening. We address the problem outlined above and recommend independently developed, valid, normative data bases. (AM J 0BSTET GYNECOL 1987;157:820-2.)
Aims-To process video slit lamp biomicroscopic fundus image sequences in order to generate wide f... more Aims-To process video slit lamp biomicroscopic fundus image sequences in order to generate wide field, high quality fundus image montages which might be suitable for photodocumentation. Methods-Slit lamp biomicroscopic fundus examination was performed on human volunteers with a contact or non-contact lens. A stock, charge coupled device camera permitted image capture and storage of the image sequence at 30 frames per second. Acquisition time was approximately 30 seconds. Individual slit lamp biomicroscope fundus image frames were aligned and blended with custom developed software.
Trabecular bone networks consist of two basic microstructural types: plates and rods. Although tr... more Trabecular bone networks consist of two basic microstructural types: plates and rods. Although trabecular rods represent only a small fraction of total bone volume, their existence has important roles in failure initiation and progression. The goal of this study was to quantitatively examine the contributions of trabecular rods in various orientations to the anisotropic elastic moduli of human vertebral trabecular bone. Twenty-one human vertebral trabecular bone specimens were scanned by micro computed tomography (µCT). A coordinate system of orthotropic axes representing the best elastic orthotropic symmetry was determined for each sample. Individual trabeculae segmentation (ITS), a 3D image analysis technique, was performed to identify each individual trabecular rod and determine its orientation in the orthotropic coordinate system. Next, three rod-removed images were created where longitudinal, oblique, or transverse trabecular rods were removed, respectively, from the original µCT images. The original and three categories of rod-removed images were then converted to finite element (FE) models for evaluation of their elastic moduli and anisotropy. Both the transverse and oblique rod-removal caused significant decreases in all six elastic moduli. However, the removal of longitudinal rods only caused significant changes in E 33 , G 23 , and G 31 but not in any transverse/in-plane elastic properties (E 11 , E 22 , and G 12 ). The analysis of covariance (ANCOVA) with repeated measures was applied to detect the moduli change in the different models caused by the effects beyond just bone volume loss. The results suggested that the loss of transverse rods induced a significant decrease in in-plane mechanical competence, which was greater than what could be explained only by the associated bone volume loss. In contrast, the reduction in the axial Young's modulus caused by the loss of transverse rods was proportional to the bone volume decrease. Furthermore, the loss of longitudinal rods affected the axial Young's modulus through both bone volume loss and architectural change. With aging, the reduction in in-plane mechanical competence would be magnified by the preferential loss of transverse rods. The predictive ability of bone mineral density, a surrogate of BV/TV in clinical measurements, may reduce more quickly for transverse mechanical properties than for the axial mechanical properties.
Detection, Synthesis and Compression in Mammographic Image Analysis with a Hierarchical Image Pro... more Detection, Synthesis and Compression in Mammographic Image Analysis with a Hierarchical Image Probability Model C. Spence, L. Parra, and P. Sajda ... A Method for the Comparison of Biomechanical Breast Models C. Tanner, A. Degenhard, JA Schnabel, A. Castellano Smith, C. Hayes, LI Sonoda, MO Leach, R. Hose, DLG Hill, and DJ Hawkes ... Ipsilateral Multi-View CAD System for Mass Detection in Digital Mammography X. Sun, W. Qian, D. Song, and RA Clark ... HAMMER: Hierarchical Attribute Matching Mechanism for Elastic Registration D. Shen ...
In this short note we highlight the fact that linear blind source separation can be formulated as... more In this short note we highlight the fact that linear blind source separation can be formulated as a generalized eigenvalue decomposition under the assumptions of non-Gaussian, non-stationary, or non-white independent sources. The solution for the unmixing matrix is given by the generalized eigenvectors that simultaneously diagonalize the covariance matrix of the observations and an additional symmetric matrix whose form depends
Given a decision that requires less than half a second for evaluating the characteristics of the ... more Given a decision that requires less than half a second for evaluating the characteristics of the incoming pitch and generating a motor response, hitting a baseball potentially requires unique perception-action coupling to achieve high performance. We designed a rapid perceptual decision-making experiment modeled as a Go/No-Go task yet tailored to reflect a real scenario confronted by a baseball hitter. For groups of experts (Division I baseball players) and novices (non-players), we recorded electroencephalography (EEG) while they performed the task. We analyzed evoked EEG single-trial variability, contingent negative variation (CNV), and pre-stimulus alpha power with respect to the expert vs. novice groups. We found strong evidence for differences in inhibitory processes between the two groups, specifically differential activity in supplementary motor areas (SMA), indicative of enhanced inhibitory control in the expert (baseball player) group. We also found selective activity in th...
Background / Purpose: Hitting a baseball is often described as the most difficult thing to do in ... more Background / Purpose: Hitting a baseball is often described as the most difficult thing to do in sports. This rapid decision requires the batter to make a judgment in a fraction of a second based largely on the trajectory and spin of the ball. There has been relatively little investigation into the neural correlates of this rapid decision making process. Recently, our group identified neural correlates of pitch identification using EEG. These results showed the temporal evolution of the decision making process, however, due to the poor spatial resolution of EEG we were unable to draw strong conclusions about which brain regions are recruited. In our current study, we use simultaneous EEG/fMRI to identify regions that are activated as well as their temporal order during the process of baseball pitch identification. Here we present preliminary results from both traditional BOLD fMRI and single trial variability analyses. Main conclusion: Our work represents a first attempt to identify...
We develop a probabilistic network model over image spaces and demonstrate its broad utility in m... more We develop a probabilistic network model over image spaces and demonstrate its broad utility in mammographic image analysis, particularly with respect to computer-aided diagnosis. The model employs a multi-scale pyramid decomposition to factor images across scale and a network of tree-structured hidden variables to capture long-range spatial dependencies. This factoring makes the computation of the density functions local and tractable. The result is a hierarchical mixture of conditional probabilities, similar to a hidden Markov model on a tree. The model parameters are found with maximum likelihood estimation using the expectation-maximization algorithm. The utility of the model is demonstrated for three applications: (1) detection of mammographic masses for computer-aided diagnosis; (2) qualitative assessment of model structure through mammographic synthesis; and (3) compression of mammographic regions of interest. 2003 Elsevier Science B.V. All rights reserved.
Objective: EEG and functional MRI potentially provide complementary information about the brain, ... more Objective: EEG and functional MRI potentially provide complementary information about the brain, yet high quality EEG is difficult to acquire during MR scanning because of mutual artifacts induced. Many groups have circumvented this difficulty by interleaving MRI volume acquisitions with EEG. However, for event-related EEG/fMRI studies, particularly those directed at single-trial analysis, high-density, high-quality, uninterrupted EEG is necessary. We have
EEG alpha-band activity is generally thought to represent an inhibitory state related to decrease... more EEG alpha-band activity is generally thought to represent an inhibitory state related to decreased attention and play a role in suppression of task-irrelevant stimulus processing, but a competing hypothesis suggests an active role in processing task-relevant information - one in which phase dynamics are involved. Here we used simultaneous EEG-fMRI and a whole-brain analysis to investigate the effects of prestimulus alpha activity on the event-related BOLD response during an auditory oddball task. We separately investigated the effects of the posterior alpha rhythm's power and phase on activity related to task-relevant stimulus processing and also investigated higher-level decision-related processing. We found stronger decision-related BOLD activity in areas late in the processing stream when subjects were in the high alpha power state prior to stimulus onset, but did not detect any effect in primary sensory regions. Our phase analysis revealed correlates in the bilateral thalamu...
First International IEEE EMBS Conference on Neural Engineering, 2003. Conference Proceedings., 2003
We describe a filter-based model of orientation processing in primary visual cortex (V1) and demo... more We describe a filter-based model of orientation processing in primary visual cortex (V1) and demonstrate that novelty in cortical "pinwheel" space can be used as a measure of perceptual salience. In the model, novelty is computed as the negative log likelihood of a pinwheel's activity relative to the population response. The population response is modeled using a mixture of Gaussians, enabling the representation of complex, multi-modal distributions. Hidden variables that are inferred in the mixture model can be viewed as grouping or "binding" pinwheels which have similar responses within the distribution space. Results are shown for several stimuli that illustrate well-known contextual effects related to perceptual salience, as well as results for a natural image.
First International IEEE EMBS Conference on Neural Engineering, 2003. Conference Proceedings., 2003
Blind source separation (BSS) has been proposed as a method to analyze multi-channel electroencep... more Blind source separation (BSS) has been proposed as a method to analyze multi-channel electroencephalography (EEG) data. A basic issue in applying BSS algorithms is the validity of the independence assumption. In this paper we investigate whether EEG can be considered to be a linear combination of independent sources. Linear BSS can be obtained under the assumptions of non-Gaussian, non-stationary, or non-white independent sources. If the linear independence hypothesis is violated these three different conditions will not necessarily lead to the same result. We show, using 64 channel EEG data, that different algorithms which incorporate the three different assumptions lead to the same results, thus supporting the linear independence hypothesis.
American Journal of Obstetrics and Gynecology, 1987
Maternal serum a-fetoprotein levels are higher in black women. Misinterpretation of maternal seru... more Maternal serum a-fetoprotein levels are higher in black women. Misinterpretation of maternal serum a-fetoprotein screening results can subject black gravid women to unnecessary invasive diagnostic procedures and their calculable risks. Screening errors for black women can result from the use of normative data bases established with maternal serum samples drawn from other racial groups or the use of such data bases in conjunction with a published correction factor. Because the incidence of open neural tube defects is lower for blacks than for others, excessive false positive results for blacks (estimated to be 8817 to 28,215 cases annually) would be a pernicious misapplication of maternal serum a-fetoprotein screening. We address the problem outlined above and recommend independently developed, valid, normative data bases. (AM J 0BSTET GYNECOL 1987;157:820-2.)
Aims-To process video slit lamp biomicroscopic fundus image sequences in order to generate wide f... more Aims-To process video slit lamp biomicroscopic fundus image sequences in order to generate wide field, high quality fundus image montages which might be suitable for photodocumentation. Methods-Slit lamp biomicroscopic fundus examination was performed on human volunteers with a contact or non-contact lens. A stock, charge coupled device camera permitted image capture and storage of the image sequence at 30 frames per second. Acquisition time was approximately 30 seconds. Individual slit lamp biomicroscope fundus image frames were aligned and blended with custom developed software.
Trabecular bone networks consist of two basic microstructural types: plates and rods. Although tr... more Trabecular bone networks consist of two basic microstructural types: plates and rods. Although trabecular rods represent only a small fraction of total bone volume, their existence has important roles in failure initiation and progression. The goal of this study was to quantitatively examine the contributions of trabecular rods in various orientations to the anisotropic elastic moduli of human vertebral trabecular bone. Twenty-one human vertebral trabecular bone specimens were scanned by micro computed tomography (µCT). A coordinate system of orthotropic axes representing the best elastic orthotropic symmetry was determined for each sample. Individual trabeculae segmentation (ITS), a 3D image analysis technique, was performed to identify each individual trabecular rod and determine its orientation in the orthotropic coordinate system. Next, three rod-removed images were created where longitudinal, oblique, or transverse trabecular rods were removed, respectively, from the original µCT images. The original and three categories of rod-removed images were then converted to finite element (FE) models for evaluation of their elastic moduli and anisotropy. Both the transverse and oblique rod-removal caused significant decreases in all six elastic moduli. However, the removal of longitudinal rods only caused significant changes in E 33 , G 23 , and G 31 but not in any transverse/in-plane elastic properties (E 11 , E 22 , and G 12 ). The analysis of covariance (ANCOVA) with repeated measures was applied to detect the moduli change in the different models caused by the effects beyond just bone volume loss. The results suggested that the loss of transverse rods induced a significant decrease in in-plane mechanical competence, which was greater than what could be explained only by the associated bone volume loss. In contrast, the reduction in the axial Young's modulus caused by the loss of transverse rods was proportional to the bone volume decrease. Furthermore, the loss of longitudinal rods affected the axial Young's modulus through both bone volume loss and architectural change. With aging, the reduction in in-plane mechanical competence would be magnified by the preferential loss of transverse rods. The predictive ability of bone mineral density, a surrogate of BV/TV in clinical measurements, may reduce more quickly for transverse mechanical properties than for the axial mechanical properties.
Detection, Synthesis and Compression in Mammographic Image Analysis with a Hierarchical Image Pro... more Detection, Synthesis and Compression in Mammographic Image Analysis with a Hierarchical Image Probability Model C. Spence, L. Parra, and P. Sajda ... A Method for the Comparison of Biomechanical Breast Models C. Tanner, A. Degenhard, JA Schnabel, A. Castellano Smith, C. Hayes, LI Sonoda, MO Leach, R. Hose, DLG Hill, and DJ Hawkes ... Ipsilateral Multi-View CAD System for Mass Detection in Digital Mammography X. Sun, W. Qian, D. Song, and RA Clark ... HAMMER: Hierarchical Attribute Matching Mechanism for Elastic Registration D. Shen ...
In this short note we highlight the fact that linear blind source separation can be formulated as... more In this short note we highlight the fact that linear blind source separation can be formulated as a generalized eigenvalue decomposition under the assumptions of non-Gaussian, non-stationary, or non-white independent sources. The solution for the unmixing matrix is given by the generalized eigenvectors that simultaneously diagonalize the covariance matrix of the observations and an additional symmetric matrix whose form depends
Uploads
Papers by Paul Sajda