What does this research mean for the field?

A gradient boosting machine learning model improves the preoperative prediction of 30-day postsurgical mortality and intensive care unit admission compared to the traditional CARES risk calculator. Novelty: ClaimNovelty.METHODOLOGICAL. Consensus alignment: ConsensusAlignment.NEUTRAL.

April 8, 2019Open Access

Utilizing Machine Learning Methods for Preoperative Prediction of Postsurgical Mortality and Intensive Care Unit Admission

Q: What is the clinical evidence from this study?

Study design: Observational. Population: Patients undergoing noncardiac and nonneurological surgery (n=90785). Intervention: Gradient boosting machine learning model vs. Combined Assessment of Risk Encountered in Surgery (CARES) model. Primary outcome: Area under the precision-recall curve (AUPRC) for 30-day postsurgical mortality.

CCCalvin J. ChiewAgency for Science, Technology and Research NLNan LiuSouthwest Medical University TWTing Hway WongDuke-NUS Medical School

Key Result

A gradient boosting machine learning model improved the prediction of 30-day postsurgical mortality (AUPRC 0.23 vs 0.15) and ICU admission (AUPRC 0.38 vs 0.18) compared to the traditional CARES risk calculator.

Study Design

Type

Observational (n=90,785)

Multicenter

Structured PICO

Do machine learning models improve the preoperative prediction of 30-day postsurgical mortality and ICU admission compared to traditional risk models in patients undergoing noncardiac surgery?

Population

90,785 adults (aged 18 years and above) who underwent noncardiac and nonneurological surgery under general or regional anesthesia at Singapore General Hospital between January 1, 2012, and October 31, 2016. Median age 54 years, 46.3% male.

Intervention

Machine learning risk prediction models (random forest, adaptive boosting, gradient boosting, and support vector machine) trained on electronic medical record data

Comparator

Traditional risk models (Combined Assessment of Risk Encountered in Surgery [CARES] model and American Society of Anaesthesiologists-Physical Status [ASA-PS])

Outcome

30-day postsurgical mortality and need for intensive care unit (ICU) stay >24 hourshard clinical

Machine learning models, particularly gradient boosting, outperform traditional risk calculators like CARES and ASA-PS in predicting 30-day mortality and ICU admission after noncardiac surgery.

Main Result

Absolute Event Rate: 0.23% vs 0.15%

Limitations

Developed and internally validated using data from a single institution in Singapore, which might limit generalizability to other settings.
The choice of 10 classifiers for each ensemble was somewhat arbitrary.
Missing values were imputed with the observed median to reduce computational complexity, whereas more sophisticated imputation approaches could have improved predictive performance.
Machine learning approaches carry issues of interpretability and logistical challenges for clinical implementation.

Abstract

OBJECTIVE: To compare the performance of machine learning models against the traditionally derived Combined Assessment of Risk Encountered in Surgery (CARES) model and the American Society of Anaesthesiologists-Physical Status (ASA-PS) in the prediction of 30-day postsurgical mortality and need for intensive care unit (ICU) stay >24 hours. BACKGROUND: Prediction of surgical risk preoperatively is important for clinical shared decision-making and planning of health resources such as ICU beds. The current growth of electronic medical records coupled with machine learning presents an opportunity to improve the performance of established risk models. METHODS: All patients aged 18 years and above who underwent noncardiac and nonneurological surgery at Singapore General Hospital (SGH) between 1 January 2012 and 31 October 2016 were included. Patient demographics, comorbidities, preoperative laboratory results, and surgery details were obtained from their electronic medical records. Seventy percent of the observations were randomly selected for training, leaving 30% for testing. Baseline models were CARES and ASA-PS. Candidate models were trained using random forest, adaptive boosting, gradient boosting, and support vector machine. Models were evaluated on area under the receiver operating characteristic curve (AUROC) and area under the precision-recall curve (AUPRC). RESULTS: A total of 90,785 patients were included, of whom 539 (0.6%) died within 30 days and 1264 (1.4%) required ICU admission >24 hours postoperatively. Baseline models achieved high AUROCs despite poor sensitivities by predicting all negative in a predominantly negative dataset. Gradient boosting was the best performing model with AUPRCs of 0.23 and 0.38 for mortality and ICU admission outcomes respectively. CONCLUSIONS: Machine learning can be used to improve surgical risk prediction compared to traditional risk calculators. AUPRC should be used to evaluate model predictive performance instead of AUROC when the dataset is imbalanced.

Perguntar à IA

Bookmark

View Full Paper

Perguntar à IA

Bookmark

View Full Paper

Cite This Study

Chiew et al. (Mon,) conducted a observational in Patients undergoing noncardiac and nonneurological surgery (n=90,785). Gradient boosting machine learning model vs. Combined Assessment of Risk Encountered in Surgery (CARES) model was evaluated on Area under the precision-recall curve (AUPRC) for 30-day postsurgical mortality. A gradient boosting machine learning model improved the prediction of 30-day postsurgical mortality (AUPRC 0.23 vs 0.15) and ICU admission (AUPRC 0.38 vs 0.18) compared to the traditional CARES risk calculator.

synapsesocial.com/papers/6a17986a8008e5848e6ed494 https://doi.org/https://doi.org/10.1097/sla.0000000000003297

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: