Table 2 Supervised machine learning model performance.
Models with multivariate predictors | ||||||
|---|---|---|---|---|---|---|
Measurement (Value, 95% CI) | LDA | CART | kNN | SVM | Random Forest | Naïve Bayes |
Accuracy | 0.68 (0.56–0.78) | 0.72 (0.60–0.81) | 0.63 (0.51–0.74) | 0.62 (0.51–0.74) | 0.72 (0.60–0.81) | 0.65 (0.54–0.76) |
Sensitivity | 0.62 (0.41–0.80) | 0.58 (0.37–0.77) | 0.42 (0.23–0.63) | 0.50 (0.30–0.70) | 0.58 (0.37–0.77) | 0.65 (0.44–0.83) |
Specificity | 0.71 (0.57–0.83) | 0.79 (0.65–0.89) | 0.79 (0.65–0.89) | 0.69 (0.55–0.81) | 0.79 (0.65–0.89) | 0.65 (0.51–0.78) |
AUC-ROC | 0.78 | 0.70 | 0.83 | – | 0.88 | 0.77 |
Models with multivariate predictors + Pit-SCHEME score | ||||||
|---|---|---|---|---|---|---|
Measurement (Value, 95% CI) | LDA | CART | kNN | SVM | Random Forest | Naïve Bayes |
Accuracy | 0.81 (0.71–0.89) | 0.79 (0.68–0.87) | 0.75 (0.63–0.84) | 0.79 (0.68–0.87) | 0.85 (0.75–0.92) | 0.83 (0.72–0.90) |
Sensitivity | 0.74 (0.54–0.89) | 0.56 (0.35–0.75) | 0.56 (0.35–0.75) | 0.56 (0.35–0.75) | 0.78 (0.58–0.91) | 0.63 (0.42–0.81) |
Specificity | 0.85 (0.72–0.94) | 0.92 (0.80–0.98) | 0.88 (0.75–0.95) | 0.92 (0.80–0.98) | 0.92 (0.80–0.98) | 0.94 (0.83–0.99) |
AUC-ROC | 0.90 | 0.77 | 0.88 | – | 0.97 | 0.81 |
