Marleen C Tjepkema-Cloostermans1, Catarina da Silva Lourenço2,3, Barry J Ruijter2, Selma C Tromp4, Gea Drost5, Francois H M Kornips6, Albertus Beishuizen7, Frank H Bosch8, Jeannette Hofmeijer2,9, Michel J A M van Putten10,2. 1. 1Department of Clinical Neurophysiology and Neurology, Medisch Spectrum Twente, Enschede, The Netherlands. 2Department of Clinical Neurophysiology, Technical Medical Centre, University of Twente, Enschede, The Netherlands. 3Biomedical Engineering, Universidade do Porto, Porto, Portugal. 4Department of Clinical Neurophysiology, St. Antonius Hospital, Nieuwegein, The Netherlands. 5Department of Neurology and Neurosurgery, University Medical Center Groningen, University of Groningen, Groningen, The Netherlands. 6Department of Neurology, VieCuri Medical Center, Venlo, The Netherlands. 7Intensive Care Center, Medisch Spectrum Twente, Enschede, The Netherlands. 8Department of Intensive Care, Rijnstate hospital, Arnhem, The Netherlands. 9Department of Neurology, Rijnstate hospital, Arnhem, The Netherlands. 2. Department of Clinical Neurophysiology, Technical Medical Centre, University of Twente, Enschede, The Netherlands. 3. Biomedical Engineering, Universidade do Porto, Porto, Portugal. 4. Department of Clinical Neurophysiology, St. Antonius Hospital, Nieuwegein, The Netherlands. 5. Department of Neurology and Neurosurgery, University Medical Center Groningen, University of Groningen, Groningen, The Netherlands. 6. Department of Neurology, VieCuri Medical Center, Venlo, The Netherlands. 7. Intensive Care Center, Medisch Spectrum Twente, Enschede, The Netherlands. 8. Department of Intensive Care, Rijnstate hospital, Arnhem, The Netherlands. 9. Department of Neurology, Rijnstate hospital, Arnhem, The Netherlands. 10. Department of Clinical Neurophysiology and Neurology, Medisch Spectrum Twente, Enschede, The Netherlands.
Abstract
OBJECTIVES: Visual assessment of the electroencephalogram by experienced clinical neurophysiologists allows reliable outcome prediction of approximately half of all comatose patients after cardiac arrest. Deep neural networks hold promise to achieve similar or even better performance, being more objective and consistent. DESIGN: Prospective cohort study. SETTING: Medical ICU of five teaching hospitals in the Netherlands. PATIENTS: Eight-hundred ninety-five consecutive comatose patients after cardiac arrest. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: Continuous electroencephalogram was recorded during the first 3 days after cardiac arrest. Functional outcome at 6 months was classified as good (Cerebral Performance Category 1-2) or poor (Cerebral Performance Category 3-5). We trained a convolutional neural network, with a VGG architecture (introduced by the Oxford Visual Geometry Group), to predict neurologic outcome at 12 and 24 hours after cardiac arrest using electroencephalogram epochs and outcome labels as inputs. Output of the network was the probability of good outcome. Data from two hospitals were used for training and internal validation (n = 661). Eighty percent of these data was used for training and cross-validation, the remaining 20% for independent internal validation. Data from the other three hospitals were used for external validation (n = 234). Prediction of poor outcome was most accurate at 12 hours, with a sensitivity in the external validation set of 58% (95% CI, 51-65%) at false positive rate of 0% (CI, 0-7%). Good outcome could be predicted at 12 hours with a sensitivity of 48% (CI, 45-51%) at a false positive rate of 5% (CI, 0-15%) in the external validation set. CONCLUSIONS: Deep learning of electroencephalogram signals outperforms any previously reported outcome predictor of coma after cardiac arrest, including visual electroencephalogram assessment by trained electroencephalogram experts. Our approach offers the potential for objective and real time, bedside insight in the neurologic prognosis of comatose patients after cardiac arrest.
OBJECTIVES: Visual assessment of the electroencephalogram by experienced clinical neurophysiologists allows reliable outcome prediction of approximately half of all comatosepatients after cardiac arrest. Deep neural networks hold promise to achieve similar or even better performance, being more objective and consistent. DESIGN: Prospective cohort study. SETTING: Medical ICU of five teaching hospitals in the Netherlands. PATIENTS: Eight-hundred ninety-five consecutive comatosepatients after cardiac arrest. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: Continuous electroencephalogram was recorded during the first 3 days after cardiac arrest. Functional outcome at 6 months was classified as good (Cerebral Performance Category 1-2) or poor (Cerebral Performance Category 3-5). We trained a convolutional neural network, with a VGG architecture (introduced by the Oxford Visual Geometry Group), to predict neurologic outcome at 12 and 24 hours after cardiac arrest using electroencephalogram epochs and outcome labels as inputs. Output of the network was the probability of good outcome. Data from two hospitals were used for training and internal validation (n = 661). Eighty percent of these data was used for training and cross-validation, the remaining 20% for independent internal validation. Data from the other three hospitals were used for external validation (n = 234). Prediction of poor outcome was most accurate at 12 hours, with a sensitivity in the external validation set of 58% (95% CI, 51-65%) at false positive rate of 0% (CI, 0-7%). Good outcome could be predicted at 12 hours with a sensitivity of 48% (CI, 45-51%) at a false positive rate of 5% (CI, 0-15%) in the external validation set. CONCLUSIONS: Deep learning of electroencephalogram signals outperforms any previously reported outcome predictor of coma after cardiac arrest, including visual electroencephalogram assessment by trained electroencephalogram experts. Our approach offers the potential for objective and real time, bedside insight in the neurologic prognosis of comatosepatients after cardiac arrest.
Authors: Anoop Mayampurath; Raffi Hagopian; Laura Venable; Kyle Carey; Dana Edelson; Matthew Churpek Journal: Crit Care Med Date: 2022-02-01 Impact factor: 9.296
Authors: Wei-Long Zheng; Edilberto Amorim; Jin Jing; Ona Wu; Mohammad Ghassemi; Jong Woo Lee; Adithya Sivaraju; Trudy Pang; Susan T Herman; Nicolas Gaspard; Barry J Ruijter; Marleen C Tjepkema-Cloostermans; Jeannette Hofmeijer; Michel J A M van Putten; M Brandon Westover Journal: IEEE Trans Biomed Eng Date: 2022-04-21 Impact factor: 4.756
Authors: Wei-Long Zheng; Edilberto Amorim; Jin Jing; Wendong Ge; Shenda Hong; Ona Wu; Mohammad Ghassemi; Jong Woo Lee; Adithya Sivaraju; Trudy Pang; Susan T Herman; Nicolas Gaspard; Barry J Ruijter; Jimeng Sun; Marleen C Tjepkema-Cloostermans; Jeannette Hofmeijer; Michel J A M van Putten; M Brandon Westover Journal: Resuscitation Date: 2021-10-24 Impact factor: 5.262
Authors: Gregory B Rehm; Irene Cortés-Puch; Brooks T Kuhn; Jimmy Nguyen; Sarina A Fazio; Michael A Johnson; Nicholas R Anderson; Chen-Nee Chuah; Jason Y Adams Journal: Crit Care Explor Date: 2021-01-08
Authors: Alexis N Simpkins; Katharina M Busl; Edilberto Amorim; Carolina Barnett-Tapia; Mackenzie C Cervenka; Monica B Dhakar; Mark R Etherton; Celia Fung; Robert Griggs; Robert G Holloway; Adam G Kelly; Imad R Khan; Karlo J Lizarraga; Hannah G Madagan; Chidinma L Onweni; Humberto Mestre; Alejandro A Rabinstein; Clio Rubinos; Dawling A Dionisio-Santos; Teddy S Youn; Lisa H Merck; Carolina B Maciel Journal: Neurocrit Care Date: 2020-09-21 Impact factor: 3.210