A continuous convolutional neural network: very early EEG most predictive for poor neurological outcome in postanoxic...

Our first objective was to characterize the predictive power of electroencephalography (EEG) over time after cardiac arrest (CA) using a continuous, convolutional neural network (CNN). Our second objective was to investigate the performance difference between a 9 and a 4-electrode-model in predicting poor neurological outcome.

We trained a CNN model using hourly 5-min EEG epochs from 366 postanoxic coma patients from the start of recording up to 72 h after CA. Primary outcome was best Cerebral Performance Category (CPC) within 6 months after CA, classified as good (CPC score 1-2) or poor (CPC score 3-5). Prediction was based on the average of predictions over all available hours, both independently and cumulatively. We report (1) model discrimination of poor neurological outcome up to 72 h after CA and (2) model performance difference between a 9 and 4-electrode configuration.

The 9-electrode cumulative CNN model reached an area under the receiver operating characteristic curve (AUC) of 82 % and a sensitivity of 71 % at 0 % false-positive rate (FPR) over all available hours. Highest predictive performance was observed before 14 h after CA, peaking at 8 h (AUC = 0.93, sensitivity at 0 % FPR = 0.81). Performance differences over hours were negligible between the 9 and 4-electrode model.

The CNN seems valuable for poor neurological outcome prediction in postanoxic coma patients, especially for very early (6-14 h after CA) EEG. A reduced electrode configuration is equally sufficient as the full set. Future studies could explore capturing nuanced temporal dependencies in EEG signals for prognostication.