Table 6 Comparison between diagnosis models.
From: A new ensemble heart attack diagnosis (EHAD) model using artificial intelligence techniques
Metrics | Recall (%) | Precision(%) | F1-score (%) | Accuracy (%) | Specificity(%) | MCC(%) | ROC-AUC | Time (Sec.) |
|---|---|---|---|---|---|---|---|---|
SVM | 84 | 84 | 84 | 82.42 | 80.49 | 64.49 | 88.54 | 0.02 |
ANN | 84 | 85.71 | 84.85 | 83.52 | 82.93 | 66.80 | 88.63 | 1.39 |
LSTM | 90 | 83.33 | 86.54 | 84.62 | 82.93 | 71.1 | 89.39 | 3.29 |
Gradient Boosting | 76 | 82.61 | 79.17 | 78.02 | 80.49 | 56.2 | 87.27 | 0.10 |
XGBoost | 80 | 83.33 | 81.63 | 80.22 | 80.44 | 60.3 | 87.22 | 20.84 |
Random Forest | 86 | 84.31 | 85.15 | 83.52 | 80.39 | 66.7 | 89.00 | 0.19 |
EHAD | 90 | 86.54 | 88.24 | 86.81 | 85.37 | 75.5 | 89.51 | 4.7 |
