In a report from News Medical, it was determined that a new method of EEG or electroencephalography recordings might “detect abnormalities in brain regions even before a seizure takes place.” Developed by KAUST “biostatistician Hernando Ombao and colleagues Yuan Wang of the University of Wisconsin-Madison and Moo K. Chung of the University of California, Irvine,” their new approach would offer a “clinically useful tool for seizure localization.”
The team aimed at improving understanding of epileptic seizures, and by doing so created a statistical approach that gleaned far more in-depth information out of the current method of employing EEG data to develop “new insight into how these signals originate and spread.”
The experts agreed that visual inspections of EEG patients, before and during seizure events has been quite effective at determining the area of the brain that might best benefit from surgical treatment, but that it was limited for patients with more difficult to treat epilepsy activity.
This new approach, according to the team of experts that designed and perfected it, “stems from a field of mathematics that analyzes large and complex datasets by studying shape representations of the data and its interactions.” It is through the analyses of the shapes that patterns within the data can be detected. Known as “a topological data analysis framework,” they sought to learn more than ever from EEGs made both before and during seizure events.
By “removing noise” from the recordings, and seeking clearer signals, they were able to plot a series of shapes “that directly relate to the signals in the recordings.” They worked towards pyramidal graphs or shapes, also called persistence landscapes, to represent the “signals coming from each electrode placed on the scalp.” These were able to deliver a more accurate chart for the point of origin of seizures.
A pyramid graph is viewed as ideal when data is “organized in some kind of hierarchal way,” according to experts in statistic. The data may be organized from most to least important, older to newer, specific to least specific or from least to most. It does not always take the form of a classic pyramid, but instead is representative of hierarchical data points that can map out a diversity of behaviors or patterns.
As the report noted, the analysis of one patient’s EEG recording indicated that “the seizure originated from a region in or around the electrodes measuring signals from the left temporal lobe of the brain. It then spread to the right temporal lobe.” And upon testing the strength and validity of the data, they determined that even with noise around the signals, the data was quite sensitive and robust.
Because of this characteristic of the data, the team is going to focus on a much broader selection of EEG recordings in order to validate findings in a more clinical manner. The intention is to use these same statistical methods to assess the impact of shocks to the brain (including during stroke or injury) as well as on the communications pathways between “brain regions and nerve cell populations.”
Topological Features to Assess Seizure Activity
Mr. Ombao, in commenting on the study said that “Epileptologists should enhance their toolboxes of data analysis by adding methods like this one that capture topological features as part of their assessment of seizure foci in more challenging cases of epilepsy.”
Clearly, there is more to be done with this type of data gathering and analyses, but the idea that predictive results might come from data evaluations of this kind is extremely promising. Those with the most difficult to treat forms of epilepsy may benefit tremendously from further enhancements or refinements.