Characterising risk of in-hospital mortality following cardiac arrest using machine learning: A retrospective international registry study
PLoS Medicine Dec 06, 2018
Nanayakkara S, et al. - Using in-hospital data available within the first 24 hours of admission following cardiac arrest, researchers developed more accurate risk-prediction models with a combination of demographic, physiological, and biochemical information utilizing both logistic regression (LR) and machine learning (ML) techniques. For this investigation, patient-level data were obtained from the Australian and New Zealand Intensive Care Society (ANZICS) Adult Patient Database for patients who had experienced cardiac arrest within 24 hours prior to admission to an intensive care unit from January 2006 to December 2016. Without the use of pre-hospital data, they found that ML approaches significantly increased predictive mortality discrimination following cardiac arrest vs existing disease severity scores and LR. These findings can improve individual prognosis, provide information and prove useful for hospital-level risk adjustment in the management of cardiac arrest.
To read this article Sign up / Login for free
- Stay on top of policy changes & daily news that matter to you as a healthcare professional
- Update yourself with quick summaries from reputed journals in your specialty
- Earn from paid surveys and influence new developments in healthcare
- Participate in daily quizzes to improve skills and compete with peers
- Save yourself time, as we bring you these and more useful features all at one place!
You get a FREE lifetime membership, with a simple one-time sign up process!