Deep Learning for outcome prediction of postanoxic coma
Electroencephalography (EEG) is increasingly used to assist in outcome prediction for patients with a postanoxic coma after cardiac arrest. Current literature shows that neurological outcome is invariably poor if the EEG remains iso-electric or low-voltage at 24 h after cardiac arrest or if it shows burst-suppression with identical bursts; such patterns are observed in approximately 30-50% of patients. Return of continuous EEG rhythms within 12 h after cardiac arrest predicts good neurological outcome with sensitivities in the range of 30 to 50% at specificities near 100%. In previous work, we reported on the Cerebral Recovery Index to assist in the visual assessment of the EEG. In this paper, we explore a deep learning approach, using a convolutional neural network for outcome prediction in patients with a postanoxic encephalopathy. Using EEGs from 287 patients at 12 h after cardiac arrest and 399 patients at 24 h after cardiac arrest, we trained and validated a convolutional neural network with raw EEG data (18 channels, longitudinal bipolar montage). As the outcome measure, we used the Cerebral Performance Category scale (CPC), dichotomized between good (CPC score 1-2) and poor outcome (CPC score 3-5). Using 5 minute artifact-free epochs from the continuous EEG recordings partitioned into 10 s snippets, we trained the convolutional neural network using 80% of the patients. Validation was performed with EEGs from the remaining 20% of patients. Outcome prediction was most accurate at 12 h after cardiac arrest, with a sensitivity of 58% at a specificity of 100% for the prediction of poor outcome. Good neurological outcome could be predicted at 12 h after cardiac arrest with a sensitivity of 58% at a specificity of 97%. In conclusion, we present a classifier for the prediction of neurological outcome after cardiac arrest, based on a convolutional neural network, providing reliable and objective prognostic information.
KeywordsElectroencephalography EEG monitoring ICU prognostication postanoxic encephalopathy Cerebral Recovery Index cardiac arrest
Unable to display preview. Download preview PDF.
- 1. Tjepkema-Cloostermans MC, Hofmeijer J, Trof RJ, Blans MJ, Beishuizen A, Putten MJAM. Electroencephalogram predicts outcome in patients with postanoxic coma during mild therapeutic hypothermia Critical Care Medicine. 2015;43.Google Scholar
- 2. Cloostermans Marleen C., Meulen Fokke B., Eertman Carin J., Hom HaroldW., Putten Michel J. A. M.. Continuous electroencephalography monitoring for early prediction of neurological outcome in postanoxic patients after cardiac arrest Critical Care Medicine. 2012;40:1.Google Scholar
- 3. Hofmeijer J, Beernink T.M.J., Bosch F.H., Beishuizen A, Tjepkema-Cloostermans M.C., Putten M.J.A.M.. Early EEG contributes to multimodal outcome prediction of postanoxic coma Neurology. 2015:1–7.Google Scholar
- 4. Spalletti M., Carrai R., Scarpino M., et al. Single electroencephalographic patterns as specific and time-dependent indicators of good and poor outcome after cardiac arrest Clinical Neurophysiology. 2016;127:2610–2617.Google Scholar
- 5. Sivaraju Adithya, Gilmore Emily J., Wira Charles R., et al. Prognostication of post-cardiac arrest coma: early clinical and electroencephalographic predictors of outcome Intensive Care Medicine. 2015;41:1264–1272.Google Scholar
- 6. Tjepkema-Cloostermans MC, Meulen FB, Meinsma G, Putten MJAM. A Cerebral Recovery Index (CRI) for early prognosis in patients after cardiac arrest Critical Care. 2013;17:R252.Google Scholar
- 7. Tjepkema-Cloostermans Marleen C., Hofmeijer Jeannette, Beishuizen Albertus, et al. Cerebral Recovery Index: reliable help for prediction of neurological outcome after cardiac arrest Critical Care Medicine. 2017;in press.Google Scholar
- 8. Chouard Tanguy, Venema Liesbeth. Machine intelligence Nature. 2015;521:435.Google Scholar
- 9. Bengio Y, Courville A, Vincent P. Representation Learning: A Review and New Perspectives Pattern Analysis and Machine Intelligence, IEEE Transactions on. 2013;35:1798–1828.Google Scholar
- 10. LeCun Y, Bengio Y, Hinton G. Deep learning Nature. 2015;13:35–35.Google Scholar
- 11.Kingma Diederik P., Ba Jimmy Lei. Adam: a Method for Stochastic Optimization International Conference on Learning Representations 2015. 2015:1–15.Google Scholar
- 12. Sandroni Claudio, Cavallaro Fabio, Callaway Clifton W, et al. Predictors of poor neurological outcome in adult comatose survivors of cardiac arrest: A systematic review and meta-analysis. Part 2: Patients treated with therapeutic hypothermia Resuscitation. 2013;84:1324–1338.Google Scholar
- 13. Lasko TA, Denny JC, Levy MA. Computational phenotype discovery using unsupervised feature learning over noisy, sparse, and irregular clinical data PLoS One 8(6): e66341. 2013.Google Scholar
- 14. Che Z, Kale D, Li W, Bahadori MT,, Liu Y. Deep Computational Pheno- typing Proceedings of the 21th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. 2015:507–516.Google Scholar
- 15. Johansen Alexander Rosenberg, Jin Jing, Maszczyk Tomasz, Dauwels Justin, Cash Sydney S,WestoverMBrandon. Epileptiform spike detection via convolutional neural networks IEEE ICASSP. 2016:754–758.Google Scholar
- 16. Jinbo Sun Wei Qin. Dakun Tan Rui Zhao. Sleep spindle detection using deep learning: A validation study based on crowdsourcing. Conf Proc IEEE Eng Med Biol Soc. 2015 Aug;2015:2828-31. doi: 10.1109/EMBC.2015.7318980. 2015.
- 17. Berger Hans. U¨ ber das Elektroenkephalogramm des Menschen. Archiv f¨ur Psychiatrie und Nervenkrankheiten. 1929;87:527–570.Google Scholar