Modular framework for detection of inter-ictal spikes in iEEG.

Area of Science:

• Neuroscience
• Biomedical Engineering
• Computational Biology

Background:

• Epilepsy is a neurological disorder characterized by recurrent seizures.
• Pharmaco-resistant epilepsy affects a significant portion of patients, necessitating advanced diagnostic tools.
• Accurate detection of inter-ictal spikes in intracranial EEG (iEEG) is crucial for understanding epilepsy and guiding treatment.

Purpose of the Study:

• To present a novel, modular detection framework for inter-ictal spikes in iEEG recordings.
• To develop a flexible system adaptable to varying detection quality and computational constraints.
• To improve the sensitivity, precision, and efficiency of epilepsy spike detection.

Main Methods:

• A three-module detection framework was developed: a first-level detector, a second-level feature extractor, and a third-level detection classifier.
• The framework operates as a "three-slot system," allowing easy substitution of modules for customization.
• The system's performance was evaluated using complex, real-world iEEG datasets.

Main Results:

• The proposed modular framework achieved highly sensitive and precise detection of inter-ictal spikes.
• The system demonstrated significant reduction in computation time compared to existing methods.
• The modular design allowed for adaptability in detection quality and computational cost.

Conclusions:

• The novel modular approach offers an effective and efficient solution for inter-ictal spike detection in iEEG.
• This framework provides a flexible platform for advancing epilepsy diagnostics and research.
• The system's adaptability and performance make it a valuable tool for clinical and research settings.