EEG-based outcome prediction after cardiac arrest with convolutional neural networks: Performance and visualization of discriminative features.

Jonas, Stefan; Rossetti, Andrea O; Oddo, Mauro; Jenni, Simon; Favaro, Paolo; Zubler, Frederic (2019). EEG-based outcome prediction after cardiac arrest with convolutional neural networks: Performance and visualization of discriminative features. Human brain mapping, 40(16), pp. 4606-4617. Wiley-Blackwell 10.1002/hbm.24724

[img] Text
Jonas, 2019, Human Brain Mapping.pdf - Published Version
Restricted to registered users only
Available under License Publisher holds Copyright.

Download (2MB)
[img]
Preview
Text
Jonas_Version_Boris.pdf - Accepted Version
Available under License Publisher holds Copyright.

Download (653kB) | Preview

Prognostication for comatose patients after cardiac arrest is a difficult but essential task. Currently, visual interpretation of electroencephalogram (EEG) is one of the main modality used in outcome prediction. There is a growing interest in computer-assisted EEG interpretation, either to overcome the possible subjectivity of visual interpretation, or to identify complex features of the EEG signal. We used a one-dimensional convolutional neural network (CNN) to predict functional outcome based on 19-channel-EEG recorded from 267 adult comatose patients during targeted temperature management after CA. The area under the receiver operating characteristic curve (AUC) on the test set was 0.885. Interestingly, model architecture and fine-tuning only played a marginal role in classification performance. We then used gradient-weighted class activation mapping (Grad-CAM) as visualization technique to identify which EEG features were used by the network to classify an EEG epoch as favorable or unfavorable outcome, and also to understand failures of the network. Grad-CAM showed that the network relied on similar features than classical visual analysis for predicting unfavorable outcome (suppressed background, epileptiform transients). This study confirms that CNNs are promising models for EEG-based prognostication in comatose patients, and that Grad-CAM can provide explanation for the models' decision-making, which is of utmost importance for future use of deep learning models in a clinical setting.

Item Type:

Journal Article (Original Article)

Division/Institute:

04 Faculty of Medicine > Department of Head Organs and Neurology (DKNS) > Clinic of Neurology

UniBE Contributor:

Zubler, Frédéric

Subjects:

600 Technology > 610 Medicine & health

ISSN:

1065-9471

Publisher:

Wiley-Blackwell

Language:

English

Submitter:

Chantal Kottler

Date Deposited:

22 Nov 2019 13:47

Last Modified:

05 Dec 2022 15:32

Publisher DOI:

10.1002/hbm.24724

PubMed ID:

31322793

Uncontrolled Keywords:

coma convolutional neural networks deep-learning electroencephalogram grad-CAM hypoxic ischemic encephalopathy interpretability prognostication

BORIS DOI:

10.7892/boris.134845

URI:

https://boris.unibe.ch/id/eprint/134845

Actions (login required)

Edit item Edit item
Provide Feedback