Localizing Seizure-Onset Zones in Presurgical Evaluation of Drug-Resistant Epilepsy by Electroencephalography/fMRI: Effectiveness of Alternative Thresholding Strategies

Hauf, Martinus; Jann, Kay; Schindler, Kaspar; Jacobs, Bart C.; Meyer, Klaus; Rummel, Christian; Mariani, Luigi; Koenig, Thomas; Wiest, Roland

doi:10.3174/ajnr.a3052

Cited by 17 publications

(17 citation statements)

References 48 publications

(60 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Different authors have chosen different statistical methods for its definition: from the global maxima of response (59), to the number of voxels within the cluster (70). Several methods have aim to separate regions of propagation from those involved in initiation such as electrical source imaging (ESI) (71, 72) of interictal spikes.…”

Section: Epileptogenic Network Abnormalitiesmentioning

confidence: 99%

Network Connectivity in Epilepsy: Resting State fMRI and EEGâ€“fMRI Contributions

2014

View full text Add to dashboard Cite

There is a growing body of evidence pointing toward large-scale networks underlying the core phenomena in epilepsy, from seizure generation to cognitive dysfunction or response to treatment. The investigation of networks in epilepsy has become a key concept to unlock a deeper understanding of the disease. Functional imaging can provide valuable information to characterize network dysfunction; in particular resting state fMRI (RS-fMRI), which is increasingly being applied to study brain networks in a number of diseases. In patients with epilepsy, network connectivity derived from RS-fMRI has found connectivity abnormalities in a number of networks; these include the epileptogenic, cognitive and sensory processing networks. However, in majority of these studies, the effect of epileptic transients in the connectivity of networks has been neglected. EEG–fMRI has frequently shown networks related to epileptic transients that in many cases are concordant with the abnormalities shown in RS studies. This points toward a relevant role of epileptic transients in the network abnormalities detected in RS-fMRI studies. In this review, we summarize the network abnormalities reported by these two techniques side by side, provide evidence of their overlapping findings, and discuss their significance in the context of the methodology of each technique. A number of clinically relevant factors that have been associated with connectivity changes are in turn associated with changes in the frequency of epileptic transients. These factors include different aspects of epilepsy ranging from treatment effects, cognitive processes, or transition between different alertness states (i.e., awake–sleep transition). For RS-fMRI to become a more effective tool to investigate clinically relevant aspects of epilepsy it is necessary to understand connectivity changes associated with epileptic transients, those associated with other clinically relevant factors and the interaction between them, which represents a gap in the current literature. We propose a framework for the investigation of network connectivity in patients with epilepsy that can integrate epileptic processes that occur across different time scales such as epileptic transients and disease duration and the implications of this approach are discussed.

show abstract

Section: Epileptogenic Network Abnormalitiesmentioning

confidence: 99%

Network Connectivity in Epilepsy: Resting State fMRI and EEGâ€“fMRI Contributions

2014

View full text Add to dashboard Cite

show abstract

“…The basic principle of fMRI lies in the statistical testing of changes in the blood oxygen level-dependent (BOLD) signal induced by either a given task or correlations with endogenous stimuli in the brain, as interictal epileptiform discharges (Hauf et al, 2012). Whilst the analysis of fMRI data is most frequently univariate, i.e., by paired categorical analysis using statistical parametric mapping, recent attempts have shifted toward understanding how multiple brain regions interact with one another.…”

Section: Introductionmentioning

confidence: 99%

Time Course Based Artifact Identification for Independent Components of Resting-State fMRI

Rummel¹,

Verma²,

Schöpf³

et al. 2013

Front. Hum. Neurosci.

View full text Add to dashboard Cite

In functional magnetic resonance imaging (fMRI) coherent oscillations of the blood oxygen level-dependent (BOLD) signal can be detected. These arise when brain regions respond to external stimuli or are activated by tasks. The same networks have been characterized during wakeful rest when functional connectivity of the human brain is organized in generic resting-state networks (RSN). Alterations of RSN emerge as neurobiological markers of pathological conditions such as altered mental state. In single-subject fMRI data the coherent components can be identified by blind source separation of the pre-processed BOLD data using spatial independent component analysis (ICA) and related approaches. The resulting maps may represent physiological RSNs or may be due to various artifacts. In this methodological study, we propose a conceptually simple and fully automatic time course based filtering procedure to detect obvious artifacts in the ICA output for resting-state fMRI. The filter is trained on six and tested on 29 healthy subjects, yielding mean filter accuracy, sensitivity and specificity of 0.80, 0.82, and 0.75 in out-of-sample tests. To estimate the impact of clearly artifactual single-subject components on group resting-state studies we analyze unfiltered and filtered output with a second level ICA procedure. Although the automated filter does not reach performance values of visual analysis by human raters, we propose that resting-state compatible analysis of ICA time courses could be very useful to complement the existing map or task/event oriented artifact classification algorithms.

show abstract

“…In some patients, this voxel was localized at the border, but just outside the IOZ, while in others, it was found more remote. Hauf and colleagues (46) ascribed similar findings of distant fMRI peak activations to the effect of propagation. Second, deactivations were only infrequently found inside the IOZ, consistent with other reports (21, 53).…”

Section: Discussionmentioning

confidence: 80%

“…The setting of an appropriate statistical threshold in functional imaging is a critical point (6, 32, 46, 47). EEG-fMRI validation studies focused on sensitivity for detecting the IOZ, using “standard” statistical thresholds, but the specificity of these results is still largely unknown.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Sensitivity and Specificity of Interictal EEG-fMRI for Detecting the Ictal Onset Zone at Different Statistical Thresholds

et al. 2014

View full text Add to dashboard Cite

There is currently a lack of knowledge about electroencephalography (EEG)-functional magnetic resonance imaging (fMRI) specificity. Our aim was to define sensitivity and specificity of blood oxygen level dependent (BOLD) responses to interictal epileptic spikes during EEG-fMRI for detecting the ictal onset zone (IOZ). We studied 21 refractory focal epilepsy patients who had a well-defined IOZ after a full presurgical evaluation and interictal spikes during EEG-fMRI. Areas of spike-related BOLD changes overlapping the IOZ in patients were considered as true positives; if no overlap was found, they were treated as false-negatives. Matched healthy case-controls had undergone similar EEG-fMRI in order to determine true-negative and false-positive fractions. The spike-related regressor of the patient was used in the design matrix of the healthy case-control. Suprathreshold BOLD changes in the brain of controls were considered as false positives, absence of these changes as true negatives. Sensitivity and specificity were calculated for different statistical thresholds at the voxel level combined with different cluster size thresholds and represented in receiver operating characteristic (ROC)-curves. Additionally, we calculated the ROC-curves based on the cluster containing the maximal significant activation. We achieved a combination of 100% specificity and 62% sensitivity, using a Z-threshold in the interval 3.4–3.5 and cluster size threshold of 350 voxels. We could obtain higher sensitivity at the expense of specificity. Similar performance was found when using the cluster containing the maximal significant activation. Our data provide a guideline for different EEG-fMRI settings with their respective sensitivity and specificity for detecting the IOZ. The unique cluster containing the maximal significant BOLD activation was a sensitive and specific marker of the IOZ.

show abstract

Localizing Seizure-Onset Zones in Presurgical Evaluation of Drug-Resistant Epilepsy by Electroencephalography/fMRI: Effectiveness of Alternative Thresholding Strategies

Cited by 17 publications

References 48 publications

Network Connectivity in Epilepsy: Resting State fMRI and EEGâ€“fMRI Contributions

Network Connectivity in Epilepsy: Resting State fMRI and EEGâ€“fMRI Contributions

Time Course Based Artifact Identification for Independent Components of Resting-State fMRI

Sensitivity and Specificity of Interictal EEG-fMRI for Detecting the Ictal Onset Zone at Different Statistical Thresholds

Contact Info

Product

Resources

About