Period-peak analysis of the EEG with microprocessor applications

Area of Science:

• Biomedical Engineering
• Neuroscience
• Signal Processing

Background:

• Clinical electroencephalogram (EEG) analysis is crucial for diagnosing neurological disorders.
• Existing EEG analysis algorithms may exhibit bias towards specific frequency spectrums.
• Manual interpretation of EEG tracings is time-consuming and subjective.

Purpose of the Study:

• To develop and evaluate a novel period-peak algorithm for automated background analysis of clinical EEG data.
• To create a time-domain EEG analysis method that aligns with manual interpretation principles.
• To assess the performance of the period-peak algorithm against existing EEG analysis techniques.

Main Methods:

• The period-peak algorithm operates in two modes: period analysis for major wave detection and peak detection for superimposed activity.
• Baseline crossings are utilized in the period analysis mode to identify major wave counts.
• Transitions to peak-detection mode occur when superimposed activity is present between major counts, enabling detection of simultaneous activities.

Main Results:

• The period-peak algorithm demonstrated capability in detecting the simultaneity of slow base-waves and faster superimposed activities in EEG.
• Preliminary comparative studies indicated that the period-peak algorithm exhibited less bias towards either end of the EEG spectrum.
• The algorithm was successfully implemented in assembly language on a microprocessor for real-time EEG data analysis.

Conclusions:

• The period-peak algorithm provides a robust and less biased approach to clinical EEG background analysis.
• Its time-domain nature and dual-mode operation enhance its ability to capture complex EEG signal characteristics.
• The successful implementation for on-line analysis suggests its potential for practical clinical application.