Combined delay and graph embedding of epileptic discharges in EEG reveals complex and recurrent nonlinear dynamics

Erem, Burak; Hyde, Daniel C.; Peters, Jurriaan M.; Duffy, Frank H.; Brooks, Dana H.; Warfield, Simon K.

doi:10.1109/isbi.2015.7163884

Cited by 10 publications

(4 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In practice, however, the algorithm is readily applied to data with time-varying dynamics by assuming that a static manifold underlies the data, with the consequence of increasing the dimensionality of the reconstructed phase space. Although it may be possible to obtain a more parsimonious representation with an alternative approach, in our experience the algorithm can be quite sensitive to even small signal changes [28, 29]. Moreover, in general, it would be useful to have a method to validate that the changes in LE coordinates can be directly attributed to meaningful changes in the bioelectric signals themselves.…”

Section: Introductionmentioning

confidence: 99%

Extensions to a manifold learning framework for time-series analysis on dynamic manifolds in bioelectric signals

et al. 2016

View full text Add to dashboard Cite

This paper addresses the challenge of extracting meaningful information from measured bioelectric signals generated by complex, large scale physiological systems such as the brain or the heart. We focus on a combination of the well-known Laplacian Eigenmaps machine learning approach with dynamical systems ideas to analyze emergent dynamic behaviors. The method reconstructs the abstract dynamical system phase-space geometry of the embedded measurements and tracks changes in physiological conditions or activities through changes in that geometry. It is geared to extract information from the joint behavior of time traces obtained from large sensor arrays, such as those used in multiple-electrode ECG and EEG, and explore the geometrical structure of the low dimensional embedding of moving time-windows of those joint snapshots. Our main contribution is a method for mapping vectors from the phase space to the data domain. We present cases to evaluate the methods, including a synthetic example using the chaotic Lorenz system, several sets of cardiac measurements from both canine and human hearts, and measurements from a human brain.

show abstract

Section: Introductionmentioning

confidence: 99%

Extensions to a manifold learning framework for time-series analysis on dynamic manifolds in bioelectric signals

et al. 2016

View full text Add to dashboard Cite

show abstract

“…By including entire segments of data, we treat the IED-related network as a part of the brain states and relies on the EHMM to separate the states from others. The abnormality of the IED networks as studied previously (Costa et al, 2021;Erem et al, 2015) could then contribute to identify such abnormality from the normal resting state network.…”

Section: Discussionmentioning

confidence: 97%

Disentanglement of Resting State Brain Networks for Localizing Epileptogenic Zone in Focal Epilepsy

Bagić

2022

Preprint

View full text Add to dashboard Cite

Resting state electromagnetic recordings have been analyzed in epilepsy patients aiding presurgical evaluation. However, it has been rarely explored how pathological networks can be separated and thus used for epileptogenic focus localization purpose. We proposed here a resting state EEG/MEG analysis framework, to disentangle brain functional networks represented by electrophysiological oscillations. Firstly, by using an Embedded Hidden Markov Model (EHMM), we constructed a state space for resting state recordings consisting of brain states with different spatiotemporal patterns. After that, functional connectivity analysis along with graph theory were applied on the extracted brain states to quantify the network features of the extracted brain states, and we determine the source location of pathological states based on these features. The EHMM model was rigorously evaluated using computer simulations. Our simulation results revealed the proposed framework can extract brain states with high accuracy regarding both spatial and temporal profiles. We than validated the entire framework as compared with clinical ground truth in 10 patients with drug-resistant focal epilepsy who underwent MEG recordings. We segmented the resting state MEG recordings into a few brain states with diverse connectivity patterns and extracted pathological brain states by applying graph theory on the constructed functional networks. We showed reasonable localization results using the extracted pathological brain states in 6/10 patients, as compared to the invasive clinical findings. The framework can serve as an objective tool in extracting brain functional networks from noninvasive resting state electromagnetic recordings. It promises to aid presurgical evaluation guiding intracranial EEG electrodes implantation.

show abstract

“…One potential problem is that segments selected for analysis can be spread over several days of recording. Another potential problem is that a single, representative segment obtained by ensemble averaging may not be able to provide an adequate summary of the entire time course of discharges due to inter-epoch variability [2], which in turn may result in poor ESI of the sources at those time points. In this paper we propose a new ESI method, Dynamic Electrical Source Imaging (DESI), that may be able to overcome these limitations of conventional ESI by avoiding marking epochs containing individual discharges and ensemble averaging entirely.…”

Section: Introductionmentioning

confidence: 99%

“…However, actually improving the signal-to-noise ratio (SNR) depends on assumptions about what is signal and noise that may not always be valid. Ensemble averaging treats inter-epoch variations as noise and discards them, but in epilepsy they may contain relevant dynamic patterns [2], [4], and discarding them may affect the accuracy of ESI results. Other ESI approaches have modeled inter-epoch covariances statistically [5].…”

Section: Introductionmentioning

confidence: 99%

Dynamic Electrical Source Imaging (DESI) of Seizures and Interictal Epileptic Discharges Without Ensemble Averaging

Erem

Hyde

Peters

et al. 2017

IEEE Trans. Med. Imaging

View full text Add to dashboard Cite

We propose an algorithm for electrical source imaging of epileptic discharges that takes a data-driven approach to regularizing the dynamics of solutions. The method is based on linear system identification on short time segments, combined with a classical inverse solution approach. Whereas ensemble averaging of segments or epochs discards inter-segment variations by averaging across them, our approach explicitly models them. Indeed, it may even be possible to avoid the need for the time-consuming process of marking epochs containing discharges altogether. We demonstrate that this approach can produce both stable and accurate inverse solutions in experiments using simulated data and real data from epilepsy patients. In an illustrative example, we show that we are able to image propagation using this approach. We show that when applied to imaging seizure data, our approach reproducibly localized frequent seizure activity to within the margins of surgeries that led to patients’ seizure freedom. The same approach could be used in the planning of epilepsy surgeries, as a way to localize potentially epileptogenic tissue that should be resected.

show abstract

Combined delay and graph embedding of epileptic discharges in EEG reveals complex and recurrent nonlinear dynamics

Cited by 10 publications

References 9 publications

Extensions to a manifold learning framework for time-series analysis on dynamic manifolds in bioelectric signals

Extensions to a manifold learning framework for time-series analysis on dynamic manifolds in bioelectric signals

Disentanglement of Resting State Brain Networks for Localizing Epileptogenic Zone in Focal Epilepsy

Dynamic Electrical Source Imaging (DESI) of Seizures and Interictal Epileptic Discharges Without Ensemble Averaging

Contact Info

Product

Resources

About