DeepCRI: Real-time EEG-based Prognostication after Cardiac Arrest

Accurate prediction of neurological outcome after cardiac arrest is essential for guiding intensive care decisions. Electroencephalography (EEG) supports prognostication; however, interpretation relies on expert judgment and is often subjective and delayed.

We developed DeepCRI, a bedside-integrated deep learning system that produces continuously updated prognostic trajectories during the first 36 hours after arrest. DeepCRI uses time-dependent decision boundaries to define good-, poor-, and gray-zone regions over time, and applies a lock-in rule that fixes classification only after sustained, concordant high-confidence evidence within a compact temporal window, thereby preventing transient threshold crossings from driving decisions.

During model development in a cohort of 522 patients, DeepCRI achieved an area under the receiver operating characteristic curve (AUC) of 0.97 at 24 h, with low calibration error (ECE=0.049). Independent validation was performed in an internal (n=219) and an external cohort (n=167). In the internal validation, DeepCRI provided lock-in classifications in 81.7% of patients, achieving 100% specificity for poor outcome with a sensitivity of 49.5%, and 95.5% sensitivity for good outcome at 73.2% specificity; 18.3% remained in the gray zone. Performance in the external validation cohort was lower: 59.9% locked, and a single false predictions reduced poor-outcome specificity to 98.4%. Post hoc analysis indicated residual EMG artifacts contributed to this false poor-outcome prediction.

By embedding DeepCRI into routine ICU EEG infrastructure, we demonstrate the technical feasibility and clinical promise of continuous, real-time AI-driven prognostication for comatose patients after cardiac arrest.