Bernhard Hellwig

Testing for directed influences among neural signals using partial directed coherence

One major challenge in neuroscience is the identification of interrelations between signals reflecting neural activity. When applying multivariate time series analysis techniques to neural signals, detection of directed relationships, which can be described in terms of Granger-causality, is of particular interest. Partial directed coherence has been introduced for a frequency domain analysis of linear Granger-causality based on modeling the underlying dynamics by vector autoregressive processes. We discuss the statistical properties of estimates for partial directed coherence and propose a significance level for testing for nonzero partial directed coherence at a given frequency. The performance of this test is illustrated by means of linear and non-linear model systems and in an application to electroencephalography and electromyography data recorded from a patient suffering from essential tremor.

Tremor-correlated cortical activity in essential tremor

BACKGROUND In patients with parkinsonian resting tremor, tremor-correlated activity in the contralateral sensorimotor cortex has been studied by both magnetoencephalography (MEG) and electroencephalography (EEG). In essential tremor, MEG failed to detect cortical involvement. The objective of this study was to investigate whether EEG recording can reveal tremor-correlated cortical activity in patients with essential tremor or enhanced physiological tremor. METHODS Seven patients with essential tremor and three patients with enhanced physiological tremor participated in the study. Unilateral postural tremor was activated by wrist extension on the right or on the left side. Electromyography (EMG) signals arising from the wrist extensor and flexor muscles, and a high-resolution EEG were recorded simultaneously. Coherences between the time series of the rectified tremor EMG and the EEG were estimated. FINDINGS In five of nine arms with essential tremor, we found highly significant coherences at the tremor frequency between the tremor EMG and the EEG. Isocoherence maps illustrating the topography of significant coherences over the scalp showed that the maximum coherences were located over the contralateral sensorimotor cortex. In the patients with enhanced physiological tremor, we were unable to detect consistent significant corticomuscular coherences at the tremor frequency. INTERPRETATION Using simultaneous EEG-EMG recordings, we showed that significant corticomuscular coherences at the tremor frequency can be found in essential tremor. This finding contrasts with a recent study based on MEG recordings. The results suggest that the sensorimotor cortex is involved in the generation of essential tremor, in a similar way to that previously shown in parkinsonian resting tremor.

Tremor-correlated cortical activity detected by electroencephalography.

OBJECTIVE In this study we investigated whether cortical activity related to Parkinsonian resting tremor can be detected by electroencephalography (EEG). METHODS Seven patients with idiopathic Parkinsons disease suffering from unilateral tremor participated in the study. Electromyography (EMG) signals arising from the wrist extensor and flexor muscles as well as a high resolution EEG were recorded simultaneously. Coherencies between EEG and EMG were calculated. RESULTS In all patients, we found highly significant coherencies at the tremor frequency or its first harmonic between the tremor EMG and contralateral EEG channels. There were no significant coherencies between the tremor EMG and ipsilateral EEG channels. Isocoherency maps illustrating the topography of the coherencies over the scalp showed that the maximum coherencies were situated over the cortical motor areas. In one case, a high coherency was also found over the parietal cortex. CONCLUSIONS The results show for the first time that tremor-correlated cortical activity can be detected by electroencephalography. The findings underline that motor areas of the cerebral cortex are involved in the neuronal network generating resting tremor in Parkinsons disease.

CROSS-SPECTRAL ANALYSIS OF TREMOR TIME SERIES

We discuss cross-spectral analysis and report applications for the investigation of human tremors. For the physiological tremor in healthy subjects, the analysis enables to determine the resonant contribution to the oscillation and allows to test for a contribution of reflexes to this tremor. Comparing the analysis of the relation between the tremor of both hands in normal subjects and subjects with a rare abnormal organization of certain neural pathways proves the involvement of central structures in enhanced physiological tremor. The relation between the left and the right side of the body in pathological tremor shows a specific difference between orthostatic and all other forms of tremor. An investigation of EEG and tremor in patients suffering from Parkinsons disease reveals the tremor-correlated cortical activity. Finally, the general issue of interpreting the results of methods designed for the analysis of bivariate processes when applied to multivariate processes is considered. We discuss and apply partial cross-spectral analysis in the frame of graphical models as an extention of bivariate cross-spectral analysis for the multivariate case.

[18F]FDG-PET is superior to [123I]IBZM-SPECT for the differential diagnosis of parkinsonism

Objective: Imaging of regional cerebral glucose metabolism with PET and striatal dopamine D2/D3 receptors (D2R) with SPECT improves the differential diagnosis of parkinsonism. We prospectively investigated 1) the diagnostic merits of these approaches in differentiating between Lewy body diseases (LBD; majority Parkinson disease [PD]) and atypical parkinsonian syndromes (APS); 2) the diagnostic value of [18F]fluorodeoxyglucose (FDG)-PET to differentiate among APS subgroups. Methods: Ninety-five of 107 consecutive patients with clinically suspected APS referred for imaging were recruited. [18F]FDG-PET scans were analyzed by visual assessment (including individual voxel-based statistical maps). Based on a priori defined disease-specific patterns, patients with putative APS were differentiated from LBD (first level) and allocated to the subgroups multiple system atrophy (MSA), progressive supranuclear palsy (PSP), or corticobasal degeneration (CBD) (second level). [123I] iodobenzamide (IBZM)-SPECT datasets were subjected to an observer-independent regions-of-interest analysis to assess striatal D2R availability. Movement disorder specialists made final clinical diagnoses after a median follow-up time of 12 months. Results: Seventy-eight patients with clinically verified APS (n = 44) or LBD (n = 34) were included in the statistical analysis. The area under the receiver operating characteristic curve for discrimination between APS and LBD was significantly larger for [18F]FDG-PET (0.94) than for [123I]IBZM-SPECT (0.74; p = 0.0006). Sensitivity/specificity of [18F]FDG-PET for diagnosing APS was 86%/91%, respectively. Sensitivity/specificity of [18F]FDG-PET in identifying APS subgroups was 77%/97% for MSA, 74%/95% for PSP, and 75%/92% for CBD. Conclusions: The diagnostic accuracy of [18F]FDG-PET for discriminating LBD from APS is considerably higher than for [123I]IBZM-SPECT. [18F]FDG-PET reliably differentiates APS subgroups.

Thalamic gray matter changes in unilateral Parkinsonian resting tremor: a voxel-based morphometric analysis of 3-dimensional magnetic resonance imaging

The thalamus is assumed to be involved in the generation of Parkinsonian tremor. Ten patients with tremor-dominant idiopathic Parkinsons disease (IPD) and strictly unilateral resting tremor were investigated by cerebral high-resolution 3-dimensional magnetic resonance imaging (MRI). MRI data were analyzed by an observer-independent morphometric technique, voxel-based morphometry (VBM). For VBM, MRI data were automatically normalized and segmented, then gray matter volumes were analyzed on a voxel-by-voxel basis in comparison to an age-matched control group using Statistical Parametric Mapping (SPM99). Highly significant structural changes, i.e. locally increased gray matter concentrations (P<0.001), were found in the nucleus ventralis intermedius (VIM) of the thalamus contralateral to the tremor side and were significantly covariant with tremor amplitudes. On the one hand, these changes were localized in close vicinity to a thalamic focal hypermetabolism as revealed by a previous positron emission tomography study in unilateral Parkinsonian tremor patients. On the other hand, the localization of the focal structural changes in VIM corresponds with the generally accepted target area of tremor surgery in IPD.

Dynamic synchronisation of central oscillators in essential tremor

OBJECTIVE Coherence analysis of electromyography (EMG) signals in essential tremor (ET) suggests that tremor in the right and left arm is induced by independent central oscillators. The sensorimotor cortex seems to be part of the tremor-generating neuronal network in ET. Here, we investigated using electroencephalography (EEG) whether the independence of central oscillators in ET is supported by the analysis of cortical activity. METHODS In 8 patients with ET, bilateral hand tremor was activated by wrist extension. EMGs from the wrist flexors and extensors were recorded simultaneously with an EEG. EEG-EMG coherence was estimated for 74 epochs of 60 s duration. RESULTS In 42.6% of the cases, EEG-EMG coherence at the tremor frequency existed only with the contralateral sensorimotor cortex. However, 21.6% of the tremor-EMGs were coherent with EEG activity over both the contralateral and ipsilateral sensorimotor cortex. Bilateral and exclusively contralateral EEG-EMG coherence could alternate within the same recording. Bilateral EEG-EMG coherence was associated with increased right-left EEG-EEG coherence, increased right-left EMG-EMG coherence as well as with increased tremor strength. CONCLUSIONS In ET, central oscillators in the right and left brain are not entirely independent of each other. They may dynamically synchronise, presumably by interhemispheric coupling via the corpus callosum.

Tremor-correlated neuronal activity in the subthalamic nucleus of Parkinsonian patients

Tremor in Parkinsons disease (PD) is generated by an oscillatory neuronal network consisting of cortex, basal ganglia and thalamus. The subthalamic nucleus (STN) which is part of the basal ganglia is of particular interest, since deep brain stimulation of the STN is an effective treatment for PD including Parkinsonian tremor. It is controversial if and how the STN contributes to tremor generation. In this study, we analyze neuronal STN activity in seven patients with Parkinsonian rest tremor who underwent stereotactic surgery for deep brain stimulation. Surface EMG was recorded from the wrist flexors and extensors. Simultaneously, neuronal spike activity was registered in different depths of the STN using an array of five microelectrodes. After spike-sorting, spectral coherence was analyzed between spike activity of STN neurons and tremor activity. Significant coherence at the tremor frequency was detected between EMG and neuronal STN activity in 76 out of 145 neurons (52.4%). In contrast, coherence in the beta band occurred only in 10 out of 145 neurons (6.9%). Tremor-coherent STN activity was widely distributed over the STN being more frequent in its dorsal parts (70.8-88.9%) than in its ventral parts (25.0-48.0%). Our results suggest that synchronous neuronal STN activity at the tremor frequency contributes to the pathogenesis of Parkinsonian tremor. The wide-spread spatial distribution of tremor-coherent spike activity argues for the recruitment of an extended network of subthalamic neurons for tremor generation.

Single-unit Analysis of Substantia Nigra Pars Reticulata Neurons in Freely Behaving Rats with Genetic Absence Epilepsy

Summary:  Purpose: The substantia nigra pars reticulata (SNpr) is assumed to be involved in the control of several kinds of epileptic seizures, an assumption based mostly on neuropharmacologic evidence. However, only very few neurophysiological recordings from the basal ganglia support neuropharmacologic data. We investigated the electrophysiologic activity of SNpr neurons in rats with genetic absence epilepsy.

Direct or indirect? Graphical models for neural oscillators.

Univariate and bivariate time series analysis techniques have enabled new insights into neural processes. However, these techniques are not feasible to distinguish direct and indirect interrelations in multivariate systems. To this aim multivariate times series techniques are presented and investigated by means of simulated as well as physiological time series. Pitfalls and limitations of these techniques are discussed.