A Multi-channel EEG Data Analysis for Poor Neuro-prognostication in Comatose Patients with Self and Cross-channel Attention Mechanism

TL;DR

This paper presents a deep learning framework utilizing multi-channel EEG data with attention mechanisms to improve the prediction of poor neurological outcomes in comatose patients, emphasizing high specificity and low false positive rates.

Contribution

It introduces a hybrid deep learning approach with self and cross-channel attention mechanisms for EEG-based prognosis, achieving high specificity in outcome prediction.

Findings

Achieved a score of 0.57 on validation data.

Utilized multi-channel EEG with attention mechanisms.

Focused on high specificity with false positive rate below 0.05.

Abstract

This work investigates the predictive potential of bipolar electroencephalogram (EEG) recordings towards efficient prediction of poor neurological outcomes. A retrospective design using a hybrid deep learning approach is utilized to optimize an objective function aiming for high specificity, i.e., true positive rate (TPR) with reduced false positives (< 0.05). A multi-channel EEG array of 18 bipolar channel pairs from a randomly selected 5-minute segment in an hour is kept. In order to determine the outcome prediction, a combination of a feature encoder with 1-D convolutional layers, learnable position encoding, a context network with attention mechanisms, and finally, a regressor and classifier blocks are used. The feature encoder extricates local temporal and spatial features, while the following position encoding and attention mechanisms attempt to capture global temporal dependencies. Results: The proposed framework by our team, OUS IVS, when validated on the challenge hidden validation data, exhibited a score of 0.57.