Simultaneous spatio-temporal matching pursuit decomposition of evoked brain responses in MEG

Two-Stage Atomic Decomposition of Multichannel EEG and the Previously Undetectable Sleep Spindles

We propose a two-step procedure for atomic decomposition of multichannel EEGs, based upon multivariate matching pursuit and dipolar inverse solution, from which atoms representing relevant EEG structures are selected according to prior knowledge. We detect sleep spindles in 147 polysomnographic recordings from the Montreal Archive of Sleep Studies. Detection is compared with human scorers and two state-of-the-art algorithms, which find only about a third of the structures conforming to the definition of sleep spindles and detected by the proposed method. We provide arguments supporting the thesis that the previously undetectable sleep spindles share the same properties as those marked by human experts and previously applied methods, and were previously omitted only because of unfavorable local signal-to-noise ratios, obscuring their visibility to both human experts and algorithms replicating their markings. All detected EEG structures are automatically parametrized by their time and frequency centers, width duration, phase, and spatial location of an equivalent dipolar source within the brain. It allowed us, for the first time, to estimate the spatial gradient of sleep spindles frequencies, which not only confirmed quantitatively the well-known prevalence of higher frequencies in posterior regions, but also revealed a significant gradient in the sagittal plane. The software used in this study is freely available.

Reducing the Number of MEG/EEG Trials Needed for the Estimation of Brain Evoked Responses: A Bootstrap Approach

OBJECTIVE

A common problem in magnetoencephalographic (MEG) and electroencephalographic (EEG) experimental paradigms relying on the estimation of brain evoked responses is the lengthy time of the experiment, which stems from the need to acquire a large number of repeated recordings. Using a bootstrap approach, we aim at reliably reducing the number of these repeated trials.

METHODS

To this end, we assessed five variants of non-parametric bootstrapping based on the classical signal-plus-noise model constituting the foundation of signal averaging in MEG/EEG. We explain which of these approaches should and which should not be used for the aforementioned purpose, and why.

RESULTS

We present results for two advocated bootstrap variants applied to auditory MEG data. The ensuing trial-averaged magnetic fields served as input to the estimation of cortical source generators, with spatio-temporal matching pursuit as an example of an inverse solution technique. We propose, for a wide range of trial numbers, a general framework to evaluate the statistical properties of the parameter estimates for source locations and related time courses.

CONCLUSION

The proposed bootstrap framework offers a systematic approach to reduce the number of trials required to estimate the evoked response. The general validity of our findings is neither bound to any particular type of MEG/EEG data nor to any specific source localization method.

SIGNIFICANCE

Practical implications of this work relate to the optimization of acquisition time of MEG/EEG experiments, thus reducing stress for the subjects (especially for patients) and minimizing related artifacts.

Can interhemispheric desynchronization of cerebral blood flow anticipate upcoming vasospasm in aneurysmal subarachnoid haemorrhage patients?

BACKGROUND

Asymmetry of cerebral autoregulation (CA) was demonstrated in patients after aneurysmal subarachnoid haemorrhage (aSAH). A classical method for CA assessment requires simultaneous measurement of both arterial blood pressure (ABP) and cerebral blood flow velocity (CBFV). In this study, we have proposed a cerebral blood flow asymmetry index based only on CBFV and analysed its association with the occurrence of vasospasm after aSAH.

NEW METHOD

The phase shifts (PS) between slow oscillations in left and right CBFV (side-to-side PS) and between ABP and CBFV (CBFV-ABP PS) were estimated using multichannel matching pursuit (MMP) and cross-spectral analysis.

RESULTS

We retrospectively analysed data collected from 45 aSAH patients (26 with vasospasm). Data were analysed up to 7th day after aSAH unless the vasospasm was detected earlier. A progressive asymmetry, manifested by a gradual increase in side-to-side PS on consecutive days after aSAH, was observed in patients who developed vasospasm (R_adj² = 0.14, p = 0.009). In these patients, early side-to-side PS was more positive than in patients without vasospasm (2.8° ± 5.6° vs -1.7° ± 5.7°, p = 0.011). No such a difference was found in CBFV-ABP PS. Patients with positive side-to-side PS were more likely to develop vasospasm than patients with negative side-to-side PS (21/7 vs 5/12, p = 0.0047).

COMPARISON WITH EXISTING METHOD

MMP, in contrast to the spectral approach, accounts for non-stationarity of analysed signals. MMP applied to the PS estimation reflects the cerebral blood flow asymmetry in aSAH better than the spectral analysis.

CONCLUSIONS

Changes in side-to-side PS might be helpful to identify patients who are at risk of vasospasm.