My Editor’s Choice

Professor Ulf Ziemann,Editor-in-Chief, Clinical Neurophysiology

Professor Ulf Ziemann
Editor-in-Chief, Clinical Neurophysiology

The Editor’s Choice—

Automated EEG-based prediction of delayed cerebral ischemia after subarachnoid hemorrhage

Clinical Neurophysiology Volume 143 (November 2022)

Zheng WL, Kim JA, Elmer J, Zafar SF, Ghanta M, Moura Junior V, Patel A, Rosenthal E, Brandon Westover M. Automated EEG-based prediction of delayed cerebral ischemia after subarachnoid hemorrhage. Clinical Neurophysiology 2022; 143: 97–106

Delayed cerebral ischemia (DCI) is a leading cause of morbidity and mortality following aneurysmal subarachnoid hemorrhage (SAH). Several mechanisms contribute to DCI, including vasospasm, cortical spreading depolarization, impaired cerebral autoregulation, neuroinflammation and microthrombosis. Therefore, monitoring of vasospasm by transcranial ultrasound alone is likely insufficient for DCI risk prediction. Continuous EEG (cEEG) has emerged as an alternative monitoring technology, and several cEEG markers have been identified to indicate the imminent risk of DCI. However, cEEG data analysis requires time-consuming manual review. In the work by Zheng et al. in this volume of Clinical Neurophysiology, a machine-learning-based automated algorithm is proposed, tested and cross-validated on 113 moderate to severe SAH patients. It demonstrates that multiple cEEG features discriminate DCI from non-DCI patients: alpha-delta ratio, percent alpha variability, Shannon entropy, and epileptiform discharge burden. The model improved predictions by emphasizing the most informative features at a given time with an AUC-ROC of 0.73, by day 5 after SAH. The machine learning-based work is important because it enables rapid, automated, multi-featured cEEG assessment to increase the utility of cEEG for accurate DCI prediction.

Read editor’s choice collection of articles: