Multi-channel EEG based automatic epileptic seizure detection using iterative filtering decomposition and Hidden...

Area of Science:

• Neurology
• Biomedical Engineering
• Signal Processing

Background:

• Epilepsy is a prevalent neurological disorder characterized by recurrent seizures.
• Electroencephalography (EEG) is a key non-invasive technique for neurological disorder analysis.
• Accurate seizure detection in EEG signals remains a critical clinical challenge.

Purpose of the Study:

• To develop and evaluate an iterative filtering-based decomposition method for improving EEG-based epilepsy seizure detection.
• To enhance the accuracy and reliability of automated seizure detection systems.
• To investigate the efficacy of combining time-domain and spectral features with probabilistic modeling for seizure classification.

Main Methods:

• An iterative filtering decomposition technique was applied to segment EEG signals into intrinsic mode functions.
• Feature extraction included 2-D power spectral density, dynamic mode decomposition power, variance, and Katz fractal dimension.
• A Hidden Markov Model (HMM) based probabilistic classifier was designed using these extracted features.

Main Results:

• The proposed method achieved high accuracy in seizure detection.
• Specifically, 99.60% accuracy was obtained on the online CHB-MIT surface EEG database.
• 99.74% accuracy was achieved on the All India Institute of Medical Science (AIIMS) Patna EEG database.

Conclusions:

• Iterative filtering decomposition of EEG signals is an effective strategy for enhancing seizure detection accuracy.
• The combination of advanced signal decomposition, feature engineering, and HMM provides a robust approach for epilepsy diagnosis.
• The validated high performance on independent datasets suggests clinical applicability.