Model | Accuracy | Sensitivity (95% CI) | Specificity (95% CI) | PPV (95% CI) | NPV (95% CI) |
Human | 0.58 | 0.56 (0.23 to 0.88) | 0.58 (0.41 to 0.75) | 0.28 (0.07 to 0.48) | 0.82 (0.66 to 0.98) |
Diameter stenosis | 0.58 | 0.67 (0.36 to 0.97) | 0.55 (0.37 to 0.72) | 0.30 (0.10 to 0.50) | 0.85 (0.69 to 1.00) |
Area stenosis | 0.60 | 0.56 (0.23 to 0.88) | 0.61 (0.44 to 0.78) | 0.29 (0.08 to 0.51) | 0.83 (0.67 to 0.98) |
QFR | 0.70 | 0.33 (0.03 to 0.64) | 0.81 (0.67 to 0.95) | 0.33 (0.03 to 0.64) | 0.81 (0.67 to 0.95) |
DL | 0.78 | 0.67 (0.36 to 0.97) | 0.81 (0.67 to 0.95) | 0.50 (0.22 to 0.78) | 0.89 (0.78 to 1.00) |
DL, deep learning; FCL, future culprit lesion; NPV, negative predictive value; PPV, positive predictive value; QFR, quantitative flow ratio.