Key result
An ensemble machine learning model accurately predicted in-hospital mortality following perioperative cardiac arrest with an AUC of 0.90, identifying prolonged CPR and high ASA physical status as top risk factors.
Why the study?
Perioperative cardiac arrest is associated with high mortality, prompting the study of its prognostic factors to mitigate risk through improved care management and planning.
Can machine learning algorithms accurately predict mortality prior to hospital discharge in patients with perioperative cardiac arrest?
Observational (n=150)
No
Can machine learning algorithms accurately predict mortality prior to hospital discharge in patients with perioperative cardiac arrest?
Answers are grounded in this paper and cite their sources.
An ensemble machine learning model can accurately predict mortality prior to hospital discharge in patients experiencing perioperative cardiac arrest, potentially aiding in personalized care.
Background Perioperative cardiac arrest (POCA) is associated with a high mortality rate. This work aims to study its prognostic factors for risk mitigation by means of care management and planning. Methods A database of 380,919 surgeries was reviewed and 150 POCAs were curated. The main outcome is the mortality prior to hospital discharge. Patient demographics, medical history, and clinical characteristics (anesthesia and surgery) were main features. Six ML algorithms were explored: LR, SVC, RF, GBM, AdaBoost, and VotingClassifier. The last one is an ensemble of the beginning five ones. k-fold cross-validation and bootstrapping minimized the prediction bias and variance, respectively. Explainers (SHAP and LIME) were used to interpret the predictions. Results The ensemble provided the most accurate and robust predictions (AUC=0.90 [95% CI, 0.78-0.98]) across various age groups. The risk factors were identified by order of importance. Surprisingly, the comorbidity of hypertension was found to have a protective effect on survival, which was reported by a recent study (Alnabelsi T. et. al. 2020) for the first time to our knowledge. Conclusions The validated ensemble classifier in aid of the explainers improved the predictive differentiation, deepening our understanding of POCA prognostication. It offers a holistic model-based approach for personalized anesthesia and surgical treatment.
No takes yet. Share an insight, caveat, or question.
Shang et al. (2022) conducted an observational in Perioperative cardiac arrest (POCA) (n=150). Ensemble machine learning model (VotingClassifier) vs. Logistic regression and other single machine learning models was evaluated on Prediction of in-hospital mortality (AUC 0.90, 95% CI 0.78-0.98). An ensemble machine learning model accurately predicted in-hospital mortality following perioperative cardiac arrest with an AUC of 0.90, identifying prolonged CPR and high ASA physical status as top risk factors.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: