Table 4.
The 11 nonlinear trimodal regression analysis parameters were used to assess cardiovascular risks through machine learning algorithms. The evaluation metrics by cardiac pathophysiology were computed.
| Algorithm | Accuracy Mean [%] | Accuracy Max [%] | Sensitivity [%] | Specificity [%] | Recall [%] | Precision [%] | AUCROC | |
|---|---|---|---|---|---|---|---|---|
| CHD | GB | 75.9 | 77.7 | 70.0 | 81.7 | 70.0 | 79.3 | 0.864 |
| RF | 85.0 | 87.4 | 81.7 | 88.4 | 81.7 | 87.6 | 0.936 | |
| ADA-B | 79.5 | 82.2 | 74.9 | 84.1 | 74.9 | 82.4 | 0.873 | |
| CVD | GB | 73.1 | 75.7 | 67.1 | 79.1 | 67.1 | 76.2 | 0.834 |
| RF | 82.1 | 83.9 | 78.8 | 85.5 | 78.8 | 84.5 | 0.914 | |
| ADA-B | 70.2 | 77.0 | 63.3 | 77.2 | 63.3 | 73.5 | 0.766 | |
| CHF | GB | 88.6 | 90.3 | 85.0 | 92.1 | 85.0 | 91.5 | 0.962 |
| RF | 95.9 | 96.5 | 95.0 | 96.9 | 95.0 | 96.8 | 0.994 | |
| ADA-B | 94.0 | 95.4 | 92.1 | 95.8 | 92.1 | 95.7 | 0.987 |