Table 4

Experimental results of various indicators for model performance evaluation

ModelNSTEMIUA
AccuracyPrecisionRecallF-1 scoreAccuracyPrecisionRecallF-1 score
SVM0.91±0.015*0.90±0.002†0.92±0.012†0.94±0.015*0.90±0.003*0.89±0.002*0.73±0.011‡0.88 ±0.013*
XGB0.95±0.014*0.94±0.011*0.98±0.003*0.96±0.007*0.93±0.017*0.96±0.008*0.82±0.014*0.89±0.016*
RF0.88±0.009*0.87±0.012‡0.95±0.003*0.93±0.003*0.81±0.007*0.94±0.005*0.58±0.015*0.72±0.008‡
NB0.61±0.009*0.80±0.003‡0.63±0.002*0.71±0.003‡0.71±0.008*0.67±0.006‡0.54±0.013*0.42±0.006*
GBM0.91±0.013*§0.90±0.002†§0.89±0.013†§0.84±0.014†§0.90±0.011*§0.88±0.006*§0.68±0.015*§0.79±0.018†§
LR0.74±0.0160.81±0.0150.85±0.0120.75±0.0090.62±0.0140.63±0.0170.73±0.0120.69±0.013
  • *Compared with the LR model, p<0.01.

  • †Compared with the LR model, p<0.05.

  • ‡Compared with the LR model, p>0.05.

  • §Compared with the XGB model, p<0.05.

  • GBM, gradient boosting machines; LR, logistic regression; NB, naïve Bayesian; NSTEMI, non-ST-elevation myocardial infarction; RF, random forest; SVM, support vector machine; UA, unstable angina pectoris; XGBoost, extreme gradient boosting.