On Quantitative Biomarkers of VNS Therapy Using EEG and ECG Signals

Area of Science:

• Neurology
• Biomedical Engineering
• Signal Processing

Background:

• Epilepsy is a neurological disorder characterized by recurrent seizures.
• Medically refractory epilepsy often requires advanced treatment strategies.
• Neuromodulation, specifically Vagus nerve stimulation (VNS), is an emerging therapy for seizure management.

Purpose of the Study:

• To objectively evaluate the effectiveness of Vagus nerve stimulation (VNS) in reducing seizure severity.
• To assess the impact of automated VNS delivery on seizure control in medically refractory epilepsy.
• To identify quantitative biomarkers for predicting VNS therapy response.

Main Methods:

• Analysis of electroencephalographic (EEG) and electrocardiographic (ECG) signals from 16 patients with VNS therapy.
• Quantitative feature extraction to measure EEG spatial synchronization and heart rate changes during seizures.
• Application of unsupervised fuzzy-c-mean classification to differentiate pre- and post-VNS treatment seizures.

Main Results:

• VNS-stimulated seizures showed reduced ictal spread and cardiovascular impact compared to pre-treatment seizures.
• A classification accuracy of 85.85% was achieved using combined EEG-ECG features to distinguish pre- and post-VNS treatment seizures.
• The developed quantitative features demonstrated discriminative ability for VNS therapy responsiveness.

Conclusions:

• Timely delivery of VNS is crucial for reducing seizure severity and improving seizure control in epilepsy patients.
• Quantitative EEG-ECG features show promise as biomarkers for predicting long-term VNS therapy response.
• Automated VNS holds potential for enhanced seizure management.