Pub Date : 2014-06-04DOI: 10.1109/PRNI.2014.6858540
C. Rondinoni, C. Salmon, Jaicer Gonçalves Rolo, A. C. Santos
Previous findings suggest that temporal coherence between Blood-Oxygen-Level Dependent (BOLD) activation in certain areas is specifically related to the micro-structural organization of fascicles, i.e., the more organized the fibers, the more intense is the communication between areas. This assumption was considered in the analysis of functional and effective connectivity in patients with AD. Support Vector Machines for pattern classification (PRoNTo Toolbox-UCL) were applied to verify the usefulness of Granger-causality effective connectivity maps in correctly classifying patients and controls. Nineteen patients and eighteen healthy controls were recruited for the study and were scanned using DTI and resting state functional connectivity MRI (rs fc-MRI). Analysis of covariance with age as a confounding factor was applied to DTI data to identify areas related to disease progression. Granger mapping was used to identify brain areas related to differences of effective connectivity between groups. Maps were then input to feature extraction procedures. Models were specified with second-level masks and, after training, classifiers were validated by a leave-one-subject-out schedule. The main difference area between groups was found in the white matter below BA6, in the right hemisphere. Weight vector maps showed differences in areas related to attentional processing and auditory stimulus integration. Results point to an association between normal ageing and differences in effective connectivity related to AD. Our results show that degeneration of fibers is complementary to the degeneration of cortical cells, in accordance with the notion that AD is a network disease.
{"title":"Multimodal neuroimaging in Alzheimer's disease: Contributions of multi-voxel pattern analysis to the analysis of DTI and resting-state MRI","authors":"C. Rondinoni, C. Salmon, Jaicer Gonçalves Rolo, A. C. Santos","doi":"10.1109/PRNI.2014.6858540","DOIUrl":"https://doi.org/10.1109/PRNI.2014.6858540","url":null,"abstract":"Previous findings suggest that temporal coherence between Blood-Oxygen-Level Dependent (BOLD) activation in certain areas is specifically related to the micro-structural organization of fascicles, i.e., the more organized the fibers, the more intense is the communication between areas. This assumption was considered in the analysis of functional and effective connectivity in patients with AD. Support Vector Machines for pattern classification (PRoNTo Toolbox-UCL) were applied to verify the usefulness of Granger-causality effective connectivity maps in correctly classifying patients and controls. Nineteen patients and eighteen healthy controls were recruited for the study and were scanned using DTI and resting state functional connectivity MRI (rs fc-MRI). Analysis of covariance with age as a confounding factor was applied to DTI data to identify areas related to disease progression. Granger mapping was used to identify brain areas related to differences of effective connectivity between groups. Maps were then input to feature extraction procedures. Models were specified with second-level masks and, after training, classifiers were validated by a leave-one-subject-out schedule. The main difference area between groups was found in the white matter below BA6, in the right hemisphere. Weight vector maps showed differences in areas related to attentional processing and auditory stimulus integration. Results point to an association between normal ageing and differences in effective connectivity related to AD. Our results show that degeneration of fibers is complementary to the degeneration of cortical cells, in accordance with the notion that AD is a network disease.","PeriodicalId":133286,"journal":{"name":"2014 International Workshop on Pattern Recognition in Neuroimaging","volume":"140 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-06-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133108591","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2014-06-04DOI: 10.1109/PRNI.2014.6858539
Andreas Trier Poulsen, Simon Kamronn, L. Parra, L. K. Hansen
We propose a probabilistic generative multi-view model to test the representational universality of human information processing. The model is tested in simulated data and in a well-established benchmark EEG dataset.
{"title":"Bayesian correlated component analysis for inference of joint EEG activation","authors":"Andreas Trier Poulsen, Simon Kamronn, L. Parra, L. K. Hansen","doi":"10.1109/PRNI.2014.6858539","DOIUrl":"https://doi.org/10.1109/PRNI.2014.6858539","url":null,"abstract":"We propose a probabilistic generative multi-view model to test the representational universality of human information processing. The model is tested in simulated data and in a well-established benchmark EEG dataset.","PeriodicalId":133286,"journal":{"name":"2014 International Workshop on Pattern Recognition in Neuroimaging","volume":"77 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-06-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114975817","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2014-06-04DOI: 10.1109/PRNI.2014.6858524
Aapo Hyvärinen, J. Hirayama, M. Kawanabe
In many multivariate time series, the correlation structure is non-stationary, i.e. it changes over time. Analysis of such non-stationarities is of particular interest in neuroimaging, in which it leads to investigation of the dynamics of connectivity. A fundamental approach for such analysis is to estimate connectivities separately in short time windows, and use existing machine learning methods, such as principal component analysis (PCA), to summarize or visualize the changes in connectivity. Here, we use the PCA approach by Leonardi et al as the starting point and present two new methods. Our goal is to simplify interpretation of the results by finding components in the original data space instead of the connectivity space. First, we show how to further analyse the principal components of connectivity matrices by a tailor-made two-rank matrix approximation, in which the eigenvectors of the conventional low-rank approximation are transformed. Second, we show how to incorporate the two-rank constraint in the estimation of PCA itself to improve the results. We further provide an interpretation of the method in terms of estimation of a probabilistic generative model related to blind source separation methods and ICA. Preliminary experiments on magnetoencephalographic data reveal possibly meaningful non-stationarity patterns in power-to-power coherence of rhythmic sources (i.e. correlation of amplitudes).
{"title":"Dynamic connectivity factorization: Interpretable decompositions of non-stationarity","authors":"Aapo Hyvärinen, J. Hirayama, M. Kawanabe","doi":"10.1109/PRNI.2014.6858524","DOIUrl":"https://doi.org/10.1109/PRNI.2014.6858524","url":null,"abstract":"In many multivariate time series, the correlation structure is non-stationary, i.e. it changes over time. Analysis of such non-stationarities is of particular interest in neuroimaging, in which it leads to investigation of the dynamics of connectivity. A fundamental approach for such analysis is to estimate connectivities separately in short time windows, and use existing machine learning methods, such as principal component analysis (PCA), to summarize or visualize the changes in connectivity. Here, we use the PCA approach by Leonardi et al as the starting point and present two new methods. Our goal is to simplify interpretation of the results by finding components in the original data space instead of the connectivity space. First, we show how to further analyse the principal components of connectivity matrices by a tailor-made two-rank matrix approximation, in which the eigenvectors of the conventional low-rank approximation are transformed. Second, we show how to incorporate the two-rank constraint in the estimation of PCA itself to improve the results. We further provide an interpretation of the method in terms of estimation of a probabilistic generative model related to blind source separation methods and ICA. Preliminary experiments on magnetoencephalographic data reveal possibly meaningful non-stationarity patterns in power-to-power coherence of rhythmic sources (i.e. correlation of amplitudes).","PeriodicalId":133286,"journal":{"name":"2014 International Workshop on Pattern Recognition in Neuroimaging","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-06-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129847036","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2014-06-04DOI: 10.1109/PRNI.2014.6858522
S. Haufe, F. Meinecke, Kai Görgen, Sven Dähne, J. Haynes, B. Blankertz, F. Biessmann
Neuroimaging data are frequently analyzed with multivariate methods. Models expressing the data as a function of underlying factors related to the brain processes under study (signals) are called forward models, while models reversing this functional relationship are called backward models. Weigth vectors of backward models (called extraction filters) indicate the measurement channels informative with respect to isolating the signals. However, being a function of both signal and noise, significant weights may be observed at channels containing pure noise, while a proportion of signal-related channels may be given zero or insignificant weight. In contrast, forward model parameters (activation patterns) may exhibit significant weights only at signal-related channels, and are therefore interpretable with respect to the origin of the brain processes under study. It is sometimes incorrectly assumed that regularization (e.g., sparsification) of backward models makes extraction filters interpretable in the same sense. However, by transforming filters into patterns of corresponding forward models, as outlined here for the linear case, this can be indeed achieved. While these considerations hold for all types of data, the distinction between filters and patterns is particularly crucial for EEG and MEG data: only activation patterns can be localized to brain anatomy using customary inverse methods. We illustrate our theoretical results using a real EEG data example.
{"title":"Parameter interpretation, regularization and source localization in multivariate linear models","authors":"S. Haufe, F. Meinecke, Kai Görgen, Sven Dähne, J. Haynes, B. Blankertz, F. Biessmann","doi":"10.1109/PRNI.2014.6858522","DOIUrl":"https://doi.org/10.1109/PRNI.2014.6858522","url":null,"abstract":"Neuroimaging data are frequently analyzed with multivariate methods. Models expressing the data as a function of underlying factors related to the brain processes under study (signals) are called forward models, while models reversing this functional relationship are called backward models. Weigth vectors of backward models (called extraction filters) indicate the measurement channels informative with respect to isolating the signals. However, being a function of both signal and noise, significant weights may be observed at channels containing pure noise, while a proportion of signal-related channels may be given zero or insignificant weight. In contrast, forward model parameters (activation patterns) may exhibit significant weights only at signal-related channels, and are therefore interpretable with respect to the origin of the brain processes under study. It is sometimes incorrectly assumed that regularization (e.g., sparsification) of backward models makes extraction filters interpretable in the same sense. However, by transforming filters into patterns of corresponding forward models, as outlined here for the linear case, this can be indeed achieved. While these considerations hold for all types of data, the distinction between filters and patterns is particularly crucial for EEG and MEG data: only activation patterns can be localized to brain anatomy using customary inverse methods. We illustrate our theoretical results using a real EEG data example.","PeriodicalId":133286,"journal":{"name":"2014 International Workshop on Pattern Recognition in Neuroimaging","volume":"22 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-06-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132310189","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2014-06-04DOI: 10.1109/PRNI.2014.6858509
Sreetama Basu, Wei Tsang Ooi, Daniel Racoceanu
Recent advances in neuroimaging has produced a spurt for automatic neuronal reconstruction algorithms for large scale data. A stochastic marked point process framework for unsupervised, automatic reconstruction of single neurons has been proposed. In this paper, we introduce improved priors modeling arborization patterns encountered in neurons for efficient detection of bifurcation junctions, terminal nodes, and intermediate points on neurite branches. These priors also enforce constraints for preserving the connectedness of the neuronal tree components in spite of imperfect labeling causing intensity inhomogeneity and discontinuities in branches. To demonstrate the effectiveness of the proposed priors, we performed neurite tracing on 3D light microscopy images of Olfactory Projection Fibre axons from the DIADEM data set and obtained good scores. We also analyzed the errors and their sources in the neurite tracing pipeline, in the hope of better integration of neuroimaging and automated tracing.
{"title":"Improved marked point process priors for single neurite tracing","authors":"Sreetama Basu, Wei Tsang Ooi, Daniel Racoceanu","doi":"10.1109/PRNI.2014.6858509","DOIUrl":"https://doi.org/10.1109/PRNI.2014.6858509","url":null,"abstract":"Recent advances in neuroimaging has produced a spurt for automatic neuronal reconstruction algorithms for large scale data. A stochastic marked point process framework for unsupervised, automatic reconstruction of single neurons has been proposed. In this paper, we introduce improved priors modeling arborization patterns encountered in neurons for efficient detection of bifurcation junctions, terminal nodes, and intermediate points on neurite branches. These priors also enforce constraints for preserving the connectedness of the neuronal tree components in spite of imperfect labeling causing intensity inhomogeneity and discontinuities in branches. To demonstrate the effectiveness of the proposed priors, we performed neurite tracing on 3D light microscopy images of Olfactory Projection Fibre axons from the DIADEM data set and obtained good scores. We also analyzed the errors and their sources in the neurite tracing pipeline, in the hope of better integration of neuroimaging and automated tracing.","PeriodicalId":133286,"journal":{"name":"2014 International Workshop on Pattern Recognition in Neuroimaging","volume":"709 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-06-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116968273","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2014-06-04DOI: 10.1109/PRNI.2014.6858527
J. Laton, J. V. Schependom, J. Gielen, J. Decoster, T. Moons, J. Keyser, M. Hert, G. Nagels
The diagnostic process for schizophrenia is mainly clinical and has to be performed by an experienced psychiatrist, relying mainly on clinical signs and symptoms. Current neurophysiological measurements can distinguish groups of healthy controls and groups of schizophrenia patients. Individual classification based on neurophysiological measurements only shows moderate accuracy. In this study, we wanted to examine whether it is possible to distinguish controls and patients individually with a good accuracy. To this end we used a combination of features from different test paradigms, in particular the auditory and visual P300 and the mismatch negativity. We selected 54 patients and 54 controls, matched for age and gender, from the data available at the UPC Kortenberg. The EEG-data were high- and low-pass filtered, epoched, artefacts were rejected and the epochs were averaged. Features (latencies and amplitudes of component peaks) were extracted from the averaged signals. The resulting dataset was used to train and test classification algorithms. Here we applied Naïve Bayes and Decision Tree (without and with AdaBoost). A combination of three evoked potentials allowed us to accurately classify individual subjects as either control or patient. For the three investigated classifiers a total accuracy of more than 80%, a sensitivity of above 82% and a specificity of at least 78% was found.
{"title":"In search of biomarkers for schizophrenia using electroencephalography","authors":"J. Laton, J. V. Schependom, J. Gielen, J. Decoster, T. Moons, J. Keyser, M. Hert, G. Nagels","doi":"10.1109/PRNI.2014.6858527","DOIUrl":"https://doi.org/10.1109/PRNI.2014.6858527","url":null,"abstract":"The diagnostic process for schizophrenia is mainly clinical and has to be performed by an experienced psychiatrist, relying mainly on clinical signs and symptoms. Current neurophysiological measurements can distinguish groups of healthy controls and groups of schizophrenia patients. Individual classification based on neurophysiological measurements only shows moderate accuracy. In this study, we wanted to examine whether it is possible to distinguish controls and patients individually with a good accuracy. To this end we used a combination of features from different test paradigms, in particular the auditory and visual P300 and the mismatch negativity. We selected 54 patients and 54 controls, matched for age and gender, from the data available at the UPC Kortenberg. The EEG-data were high- and low-pass filtered, epoched, artefacts were rejected and the epochs were averaged. Features (latencies and amplitudes of component peaks) were extracted from the averaged signals. The resulting dataset was used to train and test classification algorithms. Here we applied Naïve Bayes and Decision Tree (without and with AdaBoost). A combination of three evoked potentials allowed us to accurately classify individual subjects as either control or patient. For the three investigated classifiers a total accuracy of more than 80%, a sensitivity of above 82% and a specificity of at least 78% was found.","PeriodicalId":133286,"journal":{"name":"2014 International Workshop on Pattern Recognition in Neuroimaging","volume":"36 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-06-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129130242","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2014-06-04DOI: 10.1109/PRNI.2014.6858549
Sandro Vega-Pons, P. Avesani, M. Andric, U. Hasson
The analysis of human brain connectivity networks has become an increasingly prevalent task in neuroimaging. A few recent studies have shown the possibility of decoding brain states based on brain graph classification. Graph kernels have emerged as a powerful tool for graph comparison that allows the direct use of machine learning classifiers on brain graph collections. They allow classifying graphs with different number of nodes and therefore the inter-subject analysis without any kind of previous alignment of individual subject's data. Using whole-brain fMRI data, in this paper we present a method based on graph kernels that provides above-chance accuracy results for the inter-subject discrimination of two different types of auditory stimuli. We focus our research on determining whether this method is sensitive to the relational information in the data. Indeed, we show that the discriminative information is not only coming from topological features of the graphs like node degree distribution, but also from more complex relational patterns in the neighborhood of each node. Moreover, we investigate the suitability of two different graph representation methods, both based on data-driven parcellation techniques. Finally, we study the influence of noisy connections in our graphs and provide a way to alleviate this problem.
{"title":"Classification of inter-subject fMRI data based on graph kernels","authors":"Sandro Vega-Pons, P. Avesani, M. Andric, U. Hasson","doi":"10.1109/PRNI.2014.6858549","DOIUrl":"https://doi.org/10.1109/PRNI.2014.6858549","url":null,"abstract":"The analysis of human brain connectivity networks has become an increasingly prevalent task in neuroimaging. A few recent studies have shown the possibility of decoding brain states based on brain graph classification. Graph kernels have emerged as a powerful tool for graph comparison that allows the direct use of machine learning classifiers on brain graph collections. They allow classifying graphs with different number of nodes and therefore the inter-subject analysis without any kind of previous alignment of individual subject's data. Using whole-brain fMRI data, in this paper we present a method based on graph kernels that provides above-chance accuracy results for the inter-subject discrimination of two different types of auditory stimuli. We focus our research on determining whether this method is sensitive to the relational information in the data. Indeed, we show that the discriminative information is not only coming from topological features of the graphs like node degree distribution, but also from more complex relational patterns in the neighborhood of each node. Moreover, we investigate the suitability of two different graph representation methods, both based on data-driven parcellation techniques. Finally, we study the influence of noisy connections in our graphs and provide a way to alleviate this problem.","PeriodicalId":133286,"journal":{"name":"2014 International Workshop on Pattern Recognition in Neuroimaging","volume":"41 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-06-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128319528","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2014-06-04DOI: 10.1109/PRNI.2014.6858516
Elvis Dohmatob, Alexandre Gramfort, B. Thirion, G. Varoquaux
Learning predictive models from brain imaging data, as in decoding cognitive states from fMRI (functional Magnetic Resonance Imaging), is typically an ill-posed problem as it entails estimating many more parameters than available sample points. This estimation problem thus requires regularization. Total variation regularization, combined with sparse models, has been shown to yield good predictive performance, as well as stable and interpretable maps. However, the corresponding optimization problem is very challenging: it is non-smooth, non-separable and heavily ill-conditioned. For the penalty to fully exercise its structuring effect on the maps, this optimization problem must be solved to a good tolerance resulting in a computational challenge. Here we explore a wide variety of solvers and exhibit their convergence properties on fMRI data. We introduce a variant of smooth solvers and show that it is a promising approach in these settings. Our findings show that care must be taken in solving TV-ℓ1 estimation in brain imaging and highlight the successful strategies.
{"title":"Benchmarking solvers for TV-ℓ1 least-squares and logistic regression in brain imaging","authors":"Elvis Dohmatob, Alexandre Gramfort, B. Thirion, G. Varoquaux","doi":"10.1109/PRNI.2014.6858516","DOIUrl":"https://doi.org/10.1109/PRNI.2014.6858516","url":null,"abstract":"Learning predictive models from brain imaging data, as in decoding cognitive states from fMRI (functional Magnetic Resonance Imaging), is typically an ill-posed problem as it entails estimating many more parameters than available sample points. This estimation problem thus requires regularization. Total variation regularization, combined with sparse models, has been shown to yield good predictive performance, as well as stable and interpretable maps. However, the corresponding optimization problem is very challenging: it is non-smooth, non-separable and heavily ill-conditioned. For the penalty to fully exercise its structuring effect on the maps, this optimization problem must be solved to a good tolerance resulting in a computational challenge. Here we explore a wide variety of solvers and exhibit their convergence properties on fMRI data. We introduce a variant of smooth solvers and show that it is a promising approach in these settings. Our findings show that care must be taken in solving TV-ℓ1 estimation in brain imaging and highlight the successful strategies.","PeriodicalId":133286,"journal":{"name":"2014 International Workshop on Pattern Recognition in Neuroimaging","volume":"33 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-06-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132324583","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2014-06-04DOI: 10.1109/PRNI.2014.6858542
S. Schoenmakers, M. Gerven, T. Heskes
New computational models have made it possible to reconstruct perceived images from BOLD responses in visual cortex. We expand a linear Gaussian framework for percept decoding with Gaussian mixture models to better represent the prior distribution of images. In our setup, different mixture components correspond to different letter categories. Our framework not only leads to more accurate reconstructions, but also automatically infers semantic categories from low-level visual areas of the human brain.
{"title":"Gaussian mixture models improve fMRI-based image reconstruction","authors":"S. Schoenmakers, M. Gerven, T. Heskes","doi":"10.1109/PRNI.2014.6858542","DOIUrl":"https://doi.org/10.1109/PRNI.2014.6858542","url":null,"abstract":"New computational models have made it possible to reconstruct perceived images from BOLD responses in visual cortex. We expand a linear Gaussian framework for percept decoding with Gaussian mixture models to better represent the prior distribution of images. In our setup, different mixture components correspond to different letter categories. Our framework not only leads to more accurate reconstructions, but also automatically infers semantic categories from low-level visual areas of the human brain.","PeriodicalId":133286,"journal":{"name":"2014 International Workshop on Pattern Recognition in Neuroimaging","volume":"25 5","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-06-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"113934877","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2014-06-04DOI: 10.1109/PRNI.2014.6858547
Alejandro Tabas-Diaz, E. Balaguer-Ballester, D. Pressnitzer, Anita Siebert, A. Rupp
Communication sounds are typically asymmetric in time and human listeners are highly sensitive to short-term temporal asymmetry. Nevertheless neurophysiological correlates of perceptual asymmetry remain largely elusive to current ap-proaches. Physiological recordings suggest that perceptual asymmetry is based on multiple scales of temporal integration within the auditory processing hierarchy. To test this hypothesis, we used magneto-encephalographic recordings to perform a model-driven analysis of auditory evoked fields (AEF) elicited by asymmetric sounds characterised by rising or decreasing envelopes (ramped and damped, respectively), using a hierarchical model of pitch perception with top-down modulation. We found a strong correlation between the perceived salience of ramped and damped stimuli and the AEFs, as quantified by the amplitude of the N100m component. Furthermore, the N100m magnitude is closely mirrored by a hierarchical model with stimulus-driven temporal integration windows of auditory nerve activity patterns. This strong correlation of AEFs, perception and modelling suggests that temporal asymmetry is processed in a hierarchical manner where integration windows are top-down modulated.
{"title":"Hierarchical processing of temporal asymmetry in human auditory cortex","authors":"Alejandro Tabas-Diaz, E. Balaguer-Ballester, D. Pressnitzer, Anita Siebert, A. Rupp","doi":"10.1109/PRNI.2014.6858547","DOIUrl":"https://doi.org/10.1109/PRNI.2014.6858547","url":null,"abstract":"Communication sounds are typically asymmetric in time and human listeners are highly sensitive to short-term temporal asymmetry. Nevertheless neurophysiological correlates of perceptual asymmetry remain largely elusive to current ap-proaches. Physiological recordings suggest that perceptual asymmetry is based on multiple scales of temporal integration within the auditory processing hierarchy. To test this hypothesis, we used magneto-encephalographic recordings to perform a model-driven analysis of auditory evoked fields (AEF) elicited by asymmetric sounds characterised by rising or decreasing envelopes (ramped and damped, respectively), using a hierarchical model of pitch perception with top-down modulation. We found a strong correlation between the perceived salience of ramped and damped stimuli and the AEFs, as quantified by the amplitude of the N100m component. Furthermore, the N100m magnitude is closely mirrored by a hierarchical model with stimulus-driven temporal integration windows of auditory nerve activity patterns. This strong correlation of AEFs, perception and modelling suggests that temporal asymmetry is processed in a hierarchical manner where integration windows are top-down modulated.","PeriodicalId":133286,"journal":{"name":"2014 International Workshop on Pattern Recognition in Neuroimaging","volume":"133 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-06-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116146738","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：