Removing cardiac interference from the electroencephalogram using a modified Pan-Tompkins algorithm and linear...

Area of Science:

• Neuroscience
• Biomedical Engineering
• Signal Processing

Background:

• Cardiac interference is a significant artifact in quantitative electroencephalograms (qEEG).
• Existing automated methods for cardiac artifact removal from EEG are limited by susceptibility to non-EEG artifacts or by altering clean EEG segments.
• ECG-based methods often assume cardiac periodicity or fail with ECG artifacts.

Purpose of the Study:

• To develop a robust, automated method for removing cardiac interference from EEG signals.
• To address limitations of current EEG and ECG-based artifact removal techniques.
• To improve the accuracy of qEEG for medical diagnoses.

Main Methods:

• A novel QRS-based regression method was developed to identify QRS peaks in the ECG without assuming periodicity.
• Artificial QRS signals were generated, and linear regression was applied to EEG channels against these signals.
• The method was tested on multi-channel EEGs from elderly subjects with cardiac and non-cardiac interference.

Main Results:

• The QRS-based regression method achieved an 80% correction rate for cardiac interference in EEG.
• The method effectively removed cardiac artifacts without altering uncorrupted EEG segments.
• The technique demonstrated robustness even in the presence of additional non-cardiac interference in the EEG.

Conclusions:

• The proposed QRS-based regression method offers an effective and robust solution for automated cardiac artifact removal from EEG.
• This advancement can enhance the reliability of qEEG for accurate medical diagnoses.
• The method overcomes limitations of existing techniques, particularly in complex artifactual conditions.