Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
July 31, 2021Open Access

Random Forest classifiers combined with PCA dimensionality reduction achieved up to 94.3% ± 5.3% accuracy in classifying absence from tonic-clonic seizures using single timeseries EEG data.

View Full Paper
Ask AI
Bookmark
Share

Why the study?

Seizure classification typically relies on full-scalp or intracranial EEG, but embedded wearable systems are constrained to recordings from only a few electrodes.

Can machine learning classifiers accurately classify absence from tonic-clonic seizures using scalp EEG reduced to a single timeseries per hemisphere?

Comparison

Multiple classifiers using single-trace selection vs dimensionality reduction over hemispherical space

Key result

Random Forest classifiers combined with PCA dimensionality reduction achieved up to 94.3% ± 5.3% accuracy in classifying absence from tonic-clonic seizures using single timeseries EEG data.

Authors

SNSébastien NazeJTJianbin TangJKJames Kozloski

Discussion

Loading...

Member takes

Overview

Supports wearable EEG seizure classification; leaves open prospective validation before clinical adoption.

Structured PICO

Can machine learning classifiers accurately classify absence from tonic-clonic seizures using scalp EEG reduced to a single timeseries per hemisphere?

P
Population
EEG recordings from the Temple University Hospital dataset used to evaluate machine learning classifiers for distinguishing absence from tonic-clonic seizures.
I
Intervention
Machine learning classifiers (Random Forest, Decision Trees, Support Vector Machines) trained on EEG signals reduced to single timeseries per hemisphere (via PCA, averaging, or subset) using spectral power features.
C
Comparator
Comparison across different classifiers, preprocessing dimensionality reduction techniques, and sampling methods.
O
Outcome
Accuracy in classifying absence from tonic-clonic seizures.

Main Result

Effect estimate: 94.3% ± 5.3%

Reducing EEG signals to a single timeseries per hemisphere via PCA allows Random Forest classifiers to accurately distinguish seizure types, offering a viable approach for constrained wearable monitoring devices.

Limitations

  • Sacrifices temporal precision for spatial integration of the signals across the scalp
  • Lack of interpretability of abstract features learned by deep neural networks (though mitigated by using engineered features)
  • Requires further modeling of seizure dynamics to reproduce different classes synthetically

Cite This Study

Naze et al. (2021) studied Epilepsy (absence vs tonic-clonic seizures). Random Forest classifier with PCA dimensionality reduction vs. Support Vector Machines and other classifiers/preprocessing methods was evaluated on Classification accuracy for absence vs tonic-clonic seizures (94.3% ± 5.3%). Random Forest classifiers combined with PCA dimensionality reduction achieved up to 94.3% ± 5.3% accuracy in classifying absence from tonic-clonic seizures using single timeseries EEG data.

synapsesocial.com/papers/6a22b855cce2ba38c0cd6873https://doi.org/10.1101/2021.07.28.21261310
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1Features importance in seizure classification using scalp EEG reduced to single timeseries2021 · 6 citations
  2. 2Seizure prediction with spectral power of EEG using cost-sensitive support vector machines2011 · 409 citations
  3. 3Analysis of EEG signals for the detection of epileptic seizures using feature extraction2024
  4. 4Real-Time Epileptic Seizure Detection from Raw EEG Using Classical Machine Learning and Time-Domain Feature2026
  5. 5Classification of Pre-Clinical Seizure States Using Scalp EEG Cross-Frequency Coupling Features2018 · 69 citations