Key result
Machine-learning model using preoperative EMR data predicts 30-day mortality after non-cardiac surgery with ~0.93 AUROC.
Why the study?
Machine-learning techniques are useful for clinical prediction models, and the study aimed to construct a prediction model for postoperative 30-day mortality using automatically extracted electronic preoperative evaluation sheets.
Can a machine-learning prediction model based on preoperative electronic medical records accurately predict 30-day mortality after non-cardiac surgery?
Observational (n=339,725)
Yes
Can a machine-learning prediction model based on preoperative electronic medical records accurately predict 30-day mortality after non-cardiac surgery?
A machine-learning model using automatically extracted preoperative EMR data can accurately predict 30-day mortality after non-cardiac surgery.
Answers are grounded in this paper and cite their sources.
May facilitate automated preoperative risk stratification; extends ML models but leaves open prospective validation before clinical adoption.
BACKGROUND: Machine-learning techniques are useful for creating prediction models in clinical practice. This study aimed to construct a prediction model of postoperative 30-day mortality based on an automatically extracted electronic preoperative evaluation sheet. METHODS: We used data from 276,341 consecutive adult patients who underwent non-cardiac surgery between January 2011 and December 2020 at a tertiary center for model development and internal validation, and another dataset from 63,384 patients between January 2011 and October 2021 at another center for external validation. Postoperative 30-day mortality was 0.16%. We developed an extreme gradient boosting (XGB) prediction model using only variables from preoperative evaluation sheets. RESULTS: The model yielded an area under the curve of 0.960 and an area under the precision and recall curve of 0.216, which were 0.932 and 0.122, respectively, in the external validation set. The optimal threshold calculated by Youden's J statistic had a sensitivity of 0.885 and specificity of 0.914. In an additional analysis with balanced distribution, the model showed a similar predictive value. CONCLUSION: We presented a machine-learning prediction model for 30-day mortality after non-cardiac surgery using preoperative variables automatically extracted from electronic medical records and validated the model in a multi-center setting. Our model may help clinicians predict postoperative outcomes.
No takes yet. Share an insight, caveat, or question.
Choi et al. (2022) conducted an observational in Non-cardiac surgery (n=339,725). Extreme gradient boosting (XGB) prediction model vs. Baseline logistic regression model was evaluated on Prediction of 30-day mortality (AUROC in external validation) (AUROC 0.932, 95% CI 0.919-0.945). An extreme gradient boosting machine-learning model using automatically extracted preoperative electronic medical record data accurately predicted 30-day mortality after non-cardiac surgery, achieving an AUROC of 0.932 in external validation.
Synapse has enriched 10 related papers. Consider them for comparative context: