Related Experiment Video
Updated: Nov 8, 2025
A Multimodal Imaging- and Stimulation-based Method of Evaluating Connectivity-related Brain Excitability in Patients with Epilepsy
Published on: November 13, 2016
EEG functional connectivity contributes to outcome prediction of postanoxic coma
Martín Carrasco-Gómez1, Hanneke M Keijzer2, Barry J Ruijter3
1Laboratory of Cognitive and Computational Neuroscience (LNCyC), Centre for Biomedical Technology, Universidad Politécnica de Madrid, Spain; Biomedical Research Networking Center in Bioengineering Biomaterials and Nanomedicine (CIBER-BBN), Madrid, Spain.
Functional connectivity measures in electroencephalography (EEG) significantly improve the prediction of poor neurological outcomes in comatose patients after cardiac arrest, especially when combined with other EEG features.
Area of Science:
- Neuroscience
- Medical Technology
- Clinical Prediction
Background:
- Coma after cardiac arrest presents challenges in predicting neurological outcomes.
- Electroencephalography (EEG) is crucial for monitoring brain activity in these patients.
- Existing EEG analysis methods may not fully capture the complexity of brain dysfunction.
Purpose of the Study:
- To evaluate the added value of EEG functional connectivity features for predicting outcomes in comatose patients post-cardiac arrest.
- To compare the predictive performance of functional connectivity features against non-coupling EEG features.
- To assess the combined predictive power of both feature sets.
Main Methods:
- A prospective, multicenter cohort study included 594 comatose patients following cardiac arrest.
- EEG functional connectivity metrics (coherence, phase locking value, mutual information) were computed at 12, 24, and 48 hours post-arrest.
- Machine learning models were trained using functional connectivity features, non-coupling EEG features, and a combination of both to predict neurological outcome (good vs. poor) at six months.
Main Results:
- The best functional connectivity-based classifier achieved 51% sensitivity at 100% specificity for predicting poor outcome at 12 hours.
- Non-coupling EEG features yielded 32% sensitivity at 100% specificity using data from 12 and 48 hours.
- Combining both feature sets improved prediction to 73% sensitivity at 100% specificity.
Conclusions:
- EEG functional connectivity measures enhance the accuracy of prediction models for poor outcomes in postanoxic coma.
- Early assessment of functional connectivity from EEG shows significant potential for improving prognostication in post-cardiac arrest patients.
