+1 Recommend
0 collections
      • Record: found
      • Abstract: found
      • Article: found
      Is Open Access

      An Algorithm Based on Deep Learning for Predicting In‐Hospital Cardiac Arrest


      Read this article at

          There is no author summary for this article yet. Authors can add summaries to their articles on ScienceOpen to make them more accessible to a non-specialist audience.



          In‐hospital cardiac arrest is a major burden to public health, which affects patient safety. Although traditional track‐and‐trigger systems are used to predict cardiac arrest early, they have limitations, with low sensitivity and high false‐alarm rates. We propose a deep learning–based early warning system that shows higher performance than the existing track‐and‐trigger systems.

          Methods and Results

          This retrospective cohort study reviewed patients who were admitted to 2 hospitals from June 2010 to July 2017. A total of 52 131 patients were included. Specifically, a recurrent neural network was trained using data from June 2010 to January 2017. The result was tested using the data from February to July 2017. The primary outcome was cardiac arrest, and the secondary outcome was death without attempted resuscitation. As comparative measures, we used the area under the receiver operating characteristic curve ( AUROC), the area under the precision–recall curve ( AUPRC), and the net reclassification index. Furthermore, we evaluated sensitivity while varying the number of alarms. The deep learning–based early warning system ( AUROC: 0.850; AUPRC: 0.044) significantly outperformed a modified early warning score ( AUROC: 0.603; AUPRC: 0.003), a random forest algorithm ( AUROC: 0.780; AUPRC: 0.014), and logistic regression ( AUROC: 0.613; AUPRC: 0.007). Furthermore, the deep learning–based early warning system reduced the number of alarms by 82.2%, 13.5%, and 42.1% compared with the modified early warning system, random forest, and logistic regression, respectively, at the same sensitivity.


          An algorithm based on deep learning had high sensitivity and a low false‐alarm rate for detection of patients with cardiac arrest in the multicenter study.

          Related collections

          Most cited references22

          • Record: found
          • Abstract: found
          • Article: not found

          First documented rhythm and clinical outcome from in-hospital cardiac arrest among children and adults.

          Cardiac arrests in adults are often due to ventricular fibrillation (VF) or pulseless ventricular tachycardia (VT), which are associated with better outcomes than asystole or pulseless electrical activity (PEA). Cardiac arrests in children are typically asystole or PEA. To test the hypothesis that children have relatively fewer in-hospital cardiac arrests associated with VF or pulseless VT compared with adults and, therefore, worse survival outcomes. A prospective observational study from a multicenter registry (National Registry of Cardiopulmonary Resuscitation) of cardiac arrests in 253 US and Canadian hospitals between January 1, 2000, and March 30, 2004. A total of 36,902 adults (> or =18 years) and 880 children (<18 years) with pulseless cardiac arrests requiring chest compressions, defibrillation, or both were assessed. Cardiac arrests occurring in the delivery department, neonatal intensive care unit, and in the out-of-hospital setting were excluded. Survival to hospital discharge. The rate of survival to hospital discharge following pulseless cardiac arrest was higher in children than adults (27% [236/880] vs 18% [6485/36,902]; adjusted odds ratio [OR], 2.29; 95% confidence interval [CI], 1.95-2.68). Of these survivors, 65% (154/236) of children and 73% (4737/6485) of adults had good neurological outcome. The prevalence of VF or pulseless VT as the first documented pulseless rhythm was 14% (120/880) in children and 23% (8361/36,902) in adults (OR, 0.54; 95% CI, 0.44-0.65; P<.001). The prevalence of asystole was 40% (350) in children and 35% (13 024) in adults (OR, 1.20; 95% CI, 1.10-1.40; P = .006), whereas the prevalence of PEA was 24% (213) in children and 32% (11,963) in adults (OR, 0.67; 95% CI, 0.57-0.78; P<.001). After adjustment for differences in preexisting conditions, interventions in place at time of arrest, witnessed and/or monitored status, time to defibrillation of VF or pulseless VT, intensive care unit location of arrest, and duration of cardiopulmonary resuscitation, only first documented pulseless arrest rhythm remained significantly associated with differential survival to discharge (24% [135/563] in children vs 11% [2719/24,987] in adults with asystole and PEA; adjusted OR, 2.73; 95% CI, 2.23-3.32). In this multicenter registry of in-hospital cardiac arrest, the first documented pulseless arrest rhythm was typically asystole or PEA in both children and adults. Because of better survival after asystole and PEA, children had better outcomes than adults despite fewer cardiac arrests due to VF or pulseless VT.
            • Record: found
            • Abstract: not found
            • Article: not found

            Rapid-response teams.

              • Record: found
              • Abstract: found
              • Article: not found

              Clinical antecedents to in-hospital cardiopulmonary arrest.

              While the outcome of in-hospital cardiopulmonary arrest has been studied extensively, the clinical antecedents of arrest are less well defined. We studied a group of consecutive general hospital ward patients developing cardiopulmonary arrest. Prospectively determined definitions of underlying pathophysiology, severity of underlying disease, patient complaints, and clinical observations were used to determine common clinical features. Sixty-four patients arrested 161 +/- 26 hours following hospital admission. Pathophysiologic alterations preceding arrest were classified as respiratory in 24 patients (38 percent), metabolic in 7 (11 percent), cardiac in 6 (9 percent), neurologic in 4 (6 percent), multiple in 17 (27 percent), and unclassified in 6 (9 percent). Patients with multiple disturbances had mainly respiratory (39 percent) and metabolic (44 percent) disorders. Fifty-four patients (84 percent) had documented observations of clinical deterioration or new complaints within eight hours of arrest. Seventy percent of all patients had either deterioration of respiratory or mental function observed during this time. Routine laboratory tests obtained before arrest showed no consistent abnormalities, but vital signs showed a mean respiratory rate of 29 +/- 1 breaths per minute. The prognoses of patients' underlying diseases were classified as ultimately fatal in 26 (41 percent), nonfatal in 23 (36 percent), and rapidly fatal in 15 (23 percent). Five patients (8 percent) survived to hospital discharge. Patients developing arrest on the general hospital ward services have predominantly respiratory and metabolic derangements immediately preceding their arrests. Their underlying diseases are generally not rapidly fatal. Arrest is frequently preceded by a clinical deterioration involving either respiratory or mental function. These features and the high mortality associated with arrest suggest that efforts to predict and prevent arrest might prove beneficial.

                Author and article information

                J Am Heart Assoc
                J Am Heart Assoc
                Journal of the American Heart Association: Cardiovascular and Cerebrovascular Disease
                John Wiley and Sons Inc. (Hoboken )
                26 June 2018
                03 July 2018
                : 7
                : 13 ( doiID: 10.1002/jah3.2018.7.issue-13 )
                : e008678
                [ 1 ] Department of Emergency Medicine Mediplex Sejong Hospital Incheon Korea
                [ 2 ] VUNO Seoul Korea
                [ 3 ] Department of Cardiology Mediplex Sejong Hospital Incheon Korea
                Author notes
                [*] [* ] Correspondence to: Joon‐myoung Kwon, MD, Department of Emergency medicine, Mediplex Sejong Hospital, 20, Gyeyangmunhwa‐ro, Gyeyang‐gu, Incheon 21080, Korea. E‐mail: kwonjm@ 123456sejongh.co.kr

                Dr Kwon and Mr Youngnam Lee contributed equally to this study.

                © 2018 The Authors. Published on behalf of the American Heart Association, Inc., by Wiley.

                This is an open access article under the terms of the http://creativecommons.org/licenses/by-nc/4.0/ License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited and is not used for commercial purposes.

                : 18 January 2018
                : 31 May 2018
                Page count
                Figures: 6, Tables: 4, Pages: 11, Words: 6467
                Original Research
                Original Research
                Resuscitation Science
                Custom metadata
                03 July 2018
                Converter:WILEY_ML3GV2_TO_NLMPMC version:version= mode:remove_FC converted:03.07.2018

                Cardiovascular Medicine
                artificial intelligence,cardiac arrest,deep learning,machine learning,rapid response system,resuscitation,cardiopulmonary arrest,cardiopulmonary resuscitation and emergency cardiac care,information technology


                Comment on this article