Studies in classification, data analysis, and knowledge organization, 2017
Standard data mining procedures are sensitive to the presence of outlying measurements in the dat... more Standard data mining procedures are sensitive to the presence of outlying measurements in the data. Therefore, robust data mining procedures are highly desirable, which are resistant to outliers. This work has the aim to propose new robust classification procedures for high-dimensional data and algorithms for their efficient computation. Particularly, we use the idea of implicit weights assigned to individual observation to propose several robust regularized versions of linear discriminant analysis (LDA), suitable for data with the number of variables exceeding the number of observations. The approach is based on a regularized version of the minimum weighted covariance determinant (MWCD) estimator and represents a unique attempt to combine regularization and high robustness, allowing to down-weight outlying observations. Classification performance of new methods is illustrated on real fMRI data acquired in neuroscience research.
Epilepsy is a common neurological disorder, with one third of patients not responding to currentl... more Epilepsy is a common neurological disorder, with one third of patients not responding to currently available antiepileptic drugs. The proportion of pharmacoresistant epilepsies has remained unchanged for many decades. To cure epilepsy and control seizures requires a paradigm shift in the development of new approaches to epilepsy diagnosis and treatment. Contemporary medicine has benefited from the exponential growth of computational modeling, and the application of network dynamics theory to understanding and treating human brain disorders. In epilepsy, the introduction of these approaches has led to personalized epileptic network modeling that can explore the patient's seizure genesis and predict the functional impact of resection on its individual network's propensity to seize. The application of the dynamic systems approach to neurostimulation therapy of epilepsy allows designing stimulation strategies that consider the patient's seizure dynamics and long‐term fluctuations in the stability of their epileptic networks. In this article, we review, in a nontechnical fashion suitable for a broad neuroscientific audience, recent progress in personalized dynamic brain network modeling that is shaping the future approach to the diagnosis and treatment of epilepsy.
Abstract The statistical concept of Granger causality is defined by prediction improvement, i.e. ... more Abstract The statistical concept of Granger causality is defined by prediction improvement, i.e. the causing time series contains unique information about the future of the caused one. Recently we proposed extending this concept to bivariate diffusion processes by defining Granger causality for each point of the state space as the Granger causality of a process obtained by local linearisation. This provides a Granger causality map, well-defined at least in the vicinity of stable fixed points of the deterministic part of the dynamics. This extension has convenient properties, but carries several important limitations. In the current paper we show how the Granger causality of diffusion processes can be further generalized, incorporating in particular the concept of conditional causality. Moreover, we demonstrate the application potential to systems with a more complex attractor structure such as limit cycles or bistability of fixed points.
One of the interesting aspects of EEG data is the presence of temporally stable and spatially coh... more One of the interesting aspects of EEG data is the presence of temporally stable and spatially coherent patterns of activity, known as microstates, which have been linked to various cognitive and clinical phenomena. However, there is still no general agreement on the interpretation of microstate analysis. Various clustering algorithms have been used for microstate computation, and multiple studies suggest that the microstate time series may provide insight into the neural activity of the brain in the resting state. This study addresses two gaps in the literature. Firstly, by applying several state-of-the-art microstate algorithms to a large dataset of EEG recordings, we aim to characterise and describe various microstate algorithms. We demonstrate and discuss why the three “classically” used algorithms ((T)AAHC and modified K-Means) yield virtually the same results, while HMM algorithm generates the most dissimilar results. Secondly, we aim to test the hypothesis that dynamical micro...
ABSTRACTIntracranial EEG (iEEG) data is a powerful way to map brain function, characterized by hi... more ABSTRACTIntracranial EEG (iEEG) data is a powerful way to map brain function, characterized by high temporal and spatial resolution, allowing the study of interactions among neuronal populations that orchestrate cognitive processing. However, the statistical inference and analysis of brain networks using iEEG data faces many challenges related to its sparse brain coverage, and its inhomogeneity across patients. We review these challenges and develop a methodological pipeline for estimation of network structure not obtainable from any single patient, illustrated on the inference of the interaction among visual streams using a dataset of 27 human iEEG recordings from a visual experiment employing visual scene stimuli. 100 ms sliding window and multiple band-pass filtered signals are used to provide temporal and spectral resolution. For the connectivity analysis we showcase two connectivity measures reflecting different types of interaction between regions of interest (ROI): Phase Lock...
Nonstationarity of neural dynamics is a ubiquitous prop-erty that is crucial to understanding man... more Nonstationarity of neural dynamics is a ubiquitous prop-erty that is crucial to understanding many key phenom-ena of both healthy and diseased brain function, including circadian rhythms, dynamics of epileptic activ-ity as well as cognitive processing. Detecting switching of brain states has recently become of growing interest in the human brain neuroimaging community. However, from the data analysis/modelling perspective the task is quite challenging, and competing approaches exist [1]. One widely adopted approach is the use of clustering methods in the temporal domain to detect temporally contiguous clusters of time points with a similar struc-ture of some instantaneous property- e.g. neural activity or functional connectivity profile. While this approach may in principle help to explore the switching structure
European Journal of Physical and Rehabilitation Medicine, 2021
BACKGROUND Changes of white matter integrity in people with multiple sclerosis (MS) were document... more BACKGROUND Changes of white matter integrity in people with multiple sclerosis (MS) were documented following mainly motor/skill acquisitions physical therapy, while following neuroproprioceptive "facilitation, inhibition" (neurofacilitation) only by two pilot studies. Neurofacilitation has potential to induce white matter changes due to possibility to interfere with the neuronal tactility threshold, but stronger evidence is missing. AIM This study investigates whether neurofacilitation (three physical therapy types) induce white matter changes and if they relate to clinical improvement. DESIGN The Three-Arm Parallel-Group Exploratory Trial (NCT04355663). SETTING Each group underwent different kind of two months ambulatory therapy (Motor Program Activating Therapy, Vojta's reflex locomotion, and Functional Electric Stimulation in Posturally Corrected Position). POPULATION MS people with moderate disability. METHODS At baseline and after the program, participants underwent magnetic resonance diffusion tensor imaging (DTI) and clinical assessment. Fractional anisotropy maps obtained from DTI were further analyzed using tract-based spatial statistic exploring the mean values in the whole statistic skeleton. Moreover, additional exploratory analysis in 48 regions of white matter was done. RESULTS 92 people were recruited. DTI data from 61 were analysed. The neurofacilitation (irrespective type of therapy) resulted in significant improvement on the Berg Balance Scale (p=0.0089), mainly driven by the Motor Program Activating Therapy. No statistically significant change in the whole statistic skeleton was observed (only a trend for decrement of fractional anisotropy after Vojta's reflex locomotion). Additional exploratory analysis confirmed significant decrement of fractional anisotropy in the right anterior corona radiata. CONCLUSIONS Neurofacilitation improved balance without much evidence of white matter integrity changes in people with MS. CLINICAL REHABILITATION IMPACT The study results point to the importance of neuroproprioceptive "facilitation and inhibition" physical therapy in management of balance in people with multiple sclerosis and the potential to induce white matter changes due to possibility to interfere with the neuronal tactility threshold.
Biomedical Engineering Systems and Technologies, 2017
In bioinformatics, regularized linear discriminant analysis is commonly used as a tool for superv... more In bioinformatics, regularized linear discriminant analysis is commonly used as a tool for supervised classification problems tailor-made for high-dimensional data with the number of variables exceeding the number of observations. However, its various available versions are too vulnerable to the presence of outlying measurements in the data. In this paper, we exploit principles of robust statistics to propose new versions of regularized linear discriminant analysis suitable for high-dimensional data contaminated by (more or less) severe outliers. The work exploits a regularized version of the minimum weighted covariance determinant estimator, which is one of highly robust estimators of multivariate location and scatter. The performance of the novel classification methods is illustrated on real data sets with a detailed analysis of data from brain activity research.
Studies in classification, data analysis, and knowledge organization, 2017
Standard data mining procedures are sensitive to the presence of outlying measurements in the dat... more Standard data mining procedures are sensitive to the presence of outlying measurements in the data. Therefore, robust data mining procedures are highly desirable, which are resistant to outliers. This work has the aim to propose new robust classification procedures for high-dimensional data and algorithms for their efficient computation. Particularly, we use the idea of implicit weights assigned to individual observation to propose several robust regularized versions of linear discriminant analysis (LDA), suitable for data with the number of variables exceeding the number of observations. The approach is based on a regularized version of the minimum weighted covariance determinant (MWCD) estimator and represents a unique attempt to combine regularization and high robustness, allowing to down-weight outlying observations. Classification performance of new methods is illustrated on real fMRI data acquired in neuroscience research.
Epilepsy is a common neurological disorder, with one third of patients not responding to currentl... more Epilepsy is a common neurological disorder, with one third of patients not responding to currently available antiepileptic drugs. The proportion of pharmacoresistant epilepsies has remained unchanged for many decades. To cure epilepsy and control seizures requires a paradigm shift in the development of new approaches to epilepsy diagnosis and treatment. Contemporary medicine has benefited from the exponential growth of computational modeling, and the application of network dynamics theory to understanding and treating human brain disorders. In epilepsy, the introduction of these approaches has led to personalized epileptic network modeling that can explore the patient's seizure genesis and predict the functional impact of resection on its individual network's propensity to seize. The application of the dynamic systems approach to neurostimulation therapy of epilepsy allows designing stimulation strategies that consider the patient's seizure dynamics and long‐term fluctuations in the stability of their epileptic networks. In this article, we review, in a nontechnical fashion suitable for a broad neuroscientific audience, recent progress in personalized dynamic brain network modeling that is shaping the future approach to the diagnosis and treatment of epilepsy.
Abstract The statistical concept of Granger causality is defined by prediction improvement, i.e. ... more Abstract The statistical concept of Granger causality is defined by prediction improvement, i.e. the causing time series contains unique information about the future of the caused one. Recently we proposed extending this concept to bivariate diffusion processes by defining Granger causality for each point of the state space as the Granger causality of a process obtained by local linearisation. This provides a Granger causality map, well-defined at least in the vicinity of stable fixed points of the deterministic part of the dynamics. This extension has convenient properties, but carries several important limitations. In the current paper we show how the Granger causality of diffusion processes can be further generalized, incorporating in particular the concept of conditional causality. Moreover, we demonstrate the application potential to systems with a more complex attractor structure such as limit cycles or bistability of fixed points.
One of the interesting aspects of EEG data is the presence of temporally stable and spatially coh... more One of the interesting aspects of EEG data is the presence of temporally stable and spatially coherent patterns of activity, known as microstates, which have been linked to various cognitive and clinical phenomena. However, there is still no general agreement on the interpretation of microstate analysis. Various clustering algorithms have been used for microstate computation, and multiple studies suggest that the microstate time series may provide insight into the neural activity of the brain in the resting state. This study addresses two gaps in the literature. Firstly, by applying several state-of-the-art microstate algorithms to a large dataset of EEG recordings, we aim to characterise and describe various microstate algorithms. We demonstrate and discuss why the three “classically” used algorithms ((T)AAHC and modified K-Means) yield virtually the same results, while HMM algorithm generates the most dissimilar results. Secondly, we aim to test the hypothesis that dynamical micro...
ABSTRACTIntracranial EEG (iEEG) data is a powerful way to map brain function, characterized by hi... more ABSTRACTIntracranial EEG (iEEG) data is a powerful way to map brain function, characterized by high temporal and spatial resolution, allowing the study of interactions among neuronal populations that orchestrate cognitive processing. However, the statistical inference and analysis of brain networks using iEEG data faces many challenges related to its sparse brain coverage, and its inhomogeneity across patients. We review these challenges and develop a methodological pipeline for estimation of network structure not obtainable from any single patient, illustrated on the inference of the interaction among visual streams using a dataset of 27 human iEEG recordings from a visual experiment employing visual scene stimuli. 100 ms sliding window and multiple band-pass filtered signals are used to provide temporal and spectral resolution. For the connectivity analysis we showcase two connectivity measures reflecting different types of interaction between regions of interest (ROI): Phase Lock...
Nonstationarity of neural dynamics is a ubiquitous prop-erty that is crucial to understanding man... more Nonstationarity of neural dynamics is a ubiquitous prop-erty that is crucial to understanding many key phenom-ena of both healthy and diseased brain function, including circadian rhythms, dynamics of epileptic activ-ity as well as cognitive processing. Detecting switching of brain states has recently become of growing interest in the human brain neuroimaging community. However, from the data analysis/modelling perspective the task is quite challenging, and competing approaches exist [1]. One widely adopted approach is the use of clustering methods in the temporal domain to detect temporally contiguous clusters of time points with a similar struc-ture of some instantaneous property- e.g. neural activity or functional connectivity profile. While this approach may in principle help to explore the switching structure
European Journal of Physical and Rehabilitation Medicine, 2021
BACKGROUND Changes of white matter integrity in people with multiple sclerosis (MS) were document... more BACKGROUND Changes of white matter integrity in people with multiple sclerosis (MS) were documented following mainly motor/skill acquisitions physical therapy, while following neuroproprioceptive "facilitation, inhibition" (neurofacilitation) only by two pilot studies. Neurofacilitation has potential to induce white matter changes due to possibility to interfere with the neuronal tactility threshold, but stronger evidence is missing. AIM This study investigates whether neurofacilitation (three physical therapy types) induce white matter changes and if they relate to clinical improvement. DESIGN The Three-Arm Parallel-Group Exploratory Trial (NCT04355663). SETTING Each group underwent different kind of two months ambulatory therapy (Motor Program Activating Therapy, Vojta's reflex locomotion, and Functional Electric Stimulation in Posturally Corrected Position). POPULATION MS people with moderate disability. METHODS At baseline and after the program, participants underwent magnetic resonance diffusion tensor imaging (DTI) and clinical assessment. Fractional anisotropy maps obtained from DTI were further analyzed using tract-based spatial statistic exploring the mean values in the whole statistic skeleton. Moreover, additional exploratory analysis in 48 regions of white matter was done. RESULTS 92 people were recruited. DTI data from 61 were analysed. The neurofacilitation (irrespective type of therapy) resulted in significant improvement on the Berg Balance Scale (p=0.0089), mainly driven by the Motor Program Activating Therapy. No statistically significant change in the whole statistic skeleton was observed (only a trend for decrement of fractional anisotropy after Vojta's reflex locomotion). Additional exploratory analysis confirmed significant decrement of fractional anisotropy in the right anterior corona radiata. CONCLUSIONS Neurofacilitation improved balance without much evidence of white matter integrity changes in people with MS. CLINICAL REHABILITATION IMPACT The study results point to the importance of neuroproprioceptive "facilitation and inhibition" physical therapy in management of balance in people with multiple sclerosis and the potential to induce white matter changes due to possibility to interfere with the neuronal tactility threshold.
Biomedical Engineering Systems and Technologies, 2017
In bioinformatics, regularized linear discriminant analysis is commonly used as a tool for superv... more In bioinformatics, regularized linear discriminant analysis is commonly used as a tool for supervised classification problems tailor-made for high-dimensional data with the number of variables exceeding the number of observations. However, its various available versions are too vulnerable to the presence of outlying measurements in the data. In this paper, we exploit principles of robust statistics to propose new versions of regularized linear discriminant analysis suitable for high-dimensional data contaminated by (more or less) severe outliers. The work exploits a regularized version of the minimum weighted covariance determinant estimator, which is one of highly robust estimators of multivariate location and scatter. The performance of the novel classification methods is illustrated on real data sets with a detailed analysis of data from brain activity research.
Uploads
Papers by Jaroslav Hlinka