Data Modeling Using Vital Sign Dynamics for In-hospital Mortality Classification in Patients with Acute Coronary Syndrome
2
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
Objectives: This study compared feature selection by machine learning or expert recommendation in the performance ofclassification models for in-hospital mortality among patients with acute coronary syndrome (ACS) who underwent percutaneouscoronary intervention (PCI). Methods: A dataset of 1,123 patients with ACS who underwent PCI was analyzed. Afterassigning 80% of instances to the training set through random splitting, we performed feature scaling and resampling withthe synthetic minority over-sampling technique and Tomek link method. We compared two feature selection methods: recursivefeature elimination with cross-validation (RFECV) and selection by interventional cardiologists. We used five simplemodels: support vector machine (SVM), random forest, decision tree, logistic regression, and artificial neural network. Theperformance metrics were accuracy, recall, and the false-negative rate, measured with 10-fold cross-validation in the trainingset and validated in the test set. Results: Patients’ mean age was 66.22 ± 12.88 years, and 33.63% had ST-elevation ACS. Fifteenof 34 features were selected as important with the RFECV method, while the experts chose 11 features. All models withfeature selection by RFECV had higher accuracy than the models with expert-chosen features. In the training set, the randomforest model had the highest accuracy (0.96 ± 0.01) and recall (0.97 ± 0.02). After validation in the test set, the SVM modeldisplayed the highest accuracy (0.81) and a recall of 0.61. Conclusions: Models with feature selection by RFECV had higheraccuracy than those with feature selection by experts in identifying patients with ACS at high risk for in-hospital mortality.
Abstract
Objectives: This study compared feature selection by machine learning or expert recommendation in the performance ofclassification models for in-hospital mortality among patients with acute coronary syndrome (ACS) who underwent percutaneouscoronary intervention (PCI). Methods: A dataset of 1,123 patients with ACS who underwent PCI was analyzed. Afterassigning 80% of instances to the training set through random splitting, we performed feature scaling and resampling withthe synthetic minority over-sampling technique and Tomek link method. We compared two feature selection methods: recursivefeature elimination with cross-validation (RFECV) and selection by interventional cardiologists. We used five simplemodels: support vector machine (SVM), random forest, decision tree, logistic regression, and artificial neural network. Theperformance metrics were accuracy, recall, and the false-negative rate, measured with 10-fold cross-validation in the trainingset and validated in the test set. Results: Patients’ mean age was 66.22 ± 12.88 years, and 33.63% had ST-elevation ACS. Fifteenof 34 features were selected as important with the RFECV method, while the experts chose 11 features. All models withfeature selection by RFECV had higher accuracy than the models with expert-chosen features. In the training set, the randomforest model had the highest accuracy (0.96 ± 0.01) and recall (0.97 ± 0.02). After validation in the test set, the SVM modeldisplayed the highest accuracy (0.81) and a recall of 0.61. Conclusions: Models with feature selection by RFECV had higheraccuracy than those with feature selection by experts in identifying patients with ACS at high risk for in-hospital mortality.
참고문헌 (22)
- "Cardiovascular diseases (CVDs)" (2021)[URL]
- Srivanichakorn S, "Morbidity and mortality situation of non-communicable diseases(diabetes type 2 and cardiovascular diseases)in Thailand during 2010-2014", Dis Control J, 43(4), pp.379~390 (2017)[DOI]
- Srimahachota S, "Thai Registry in Acute Coronary Syndrome(TRACS) : an extension of Thai Acute Coronary Syndrome registry(TACS)group : lower in-hospital but still high mortality at one-year", J Med Assoc Thai, 95(4), pp.508~518 (2012)
- Cei M, "In-hospital mortality and morbidity of elderly medical patients can be predicted at admission by the modified early warning score : a prospective study", Int J Clin Pract, 63(4), pp.591~595 (2009)[DOI]
- Goldhill DR, "A physiologically-based early warning score for ward patients : the association between score and outcome", Anaesthesia, 60(6), pp.547~553 (2005)[DOI]
- Groarke JD, "Use of an admission early warning score to predict patient morbidity and mortality and treatment success", Emerg Med J, 25(12), pp.803~806 (2008)[DOI]
- Alam N, "The impact of the use of the Early Warning Score(EWS)on patient outcomes : a systematic review", Resuscitation, 85, pp.587~594 (2014)[DOI]
- Bloch E, "Machine learning models for analysis of vital signs dynamics : a case for sepsis onset prediction", J Healthc Eng, 2019, pp.5930379~ (2019)[DOI]
- Kim SY, "A deep learning model for real-time mortality prediction in critically ill children", Crit Care, 23(1), pp.279~ (2019)[DOI]
- Rojas JC, "Predicting intensive care unit readmission with machine learning using electronic health record data", Ann Am Thorac Soc, 15(7), pp.846~853 (2018)[DOI]
- Kwon JM, "An algorithm based on deep learning for predicting in-hospital cardiac arrest", J Am Heart Assoc, 7(13), pp.e008678~ (2018)[DOI]
- Radhachandran A, "Prediction of short-term mortality in acute heart failure patients using minimal electronic health record data", BioData Min, 14(1), pp.23~ (2021)[DOI]
- Chehab O, "Predictors of in-hospital mortality in patients admitted with acute myocardial infarction in a developing country", Cardiol Res, 9(5), pp.293~299 (2018)[DOI]
- Granger CB, "Predictors of hospital mortality in the global registry of acute coronary events", Arch Intern Med, 163(19), pp.2345~2353 (2003)[DOI]
- Pitsavos C, "Association of creatinine clearance and in-hospital mortality in patients with acute coronary syndromes : the GREECS study", Circ J, 71(1), pp.9~14 (2007)[DOI]
- Santopinto JJ, "Creatinine clearance and adverse hospital outcomes in patients with acute coronary syndromes : findings from the global registry of acute coronary events(GRACE)", Heart, 89(9), pp.1003~1008 (2003)[DOI]
- Bittl JA, "Percutaneous coronary interventions in the diabetic patient: where do we stand?", Circ Cardiovasc Interv, 8(4), pp.e001944~ (2015)[DOI]
- Terkelsen CJ, "Mortality rates in patients with STelevation vs. non-ST-elevation acute myocardial infarction:observations from an unselected cohort", Eur Heart J, 26(1), pp.18~26 (2005)[DOI]
- Dai Y, "Atrial fibrillation in patients hospitalized with acute myocardial infarction : analysis of the china acute myocardial infarction(CAMI)registry", BMC Cardiovasc Disord, 17(1), pp.2~ (2017)[DOI]
- Seib CD, "Ensemble machine learning for the prediction of patient-level outcomes following thyroidectomy", Am J Surg, 222(2), pp.347~353 (2021)[DOI]
- Alam MZ, "Feature-ranking-based ensemble classifiers for survivability prediction of intensive care unit patients using lab test data", Inform Med Unlocked, 22, pp.100495~ (2021)[DOI]
- Liu J, "Predicting mortality of patients with acute kidney injury in the ICU using XGBoost model", PLoS One, 16(2), pp.e0246306~ (2021)[DOI]