Table 2.
Results of the clinical, image, and combination for the radiomics approach for predicting the good functional outcome (mRS ≤ 2).
| Methods | AUC 95% CI | F1-Score | Sensitivity | Specificity | PPV | NPV |
|---|---|---|---|---|---|---|
| Clinical experiment | ||||||
| RFC | 0.81 (0.79–0.82) | 0.69 (0.67–0.72) | 0.72 (0.68–0.76) | 0.75 (0.72–0.77) | 0.67 (0.63–0.71) | 0.79 (0.76–0.82) |
| SVM | 0.81 (0.80–0.83) | 0.71 (0.68–0.74) | 0.79 (0.75–0.82) | 0.70 (0.68–0.72) | 0.65 (0.61–0.69) | 0.82 (0.79–0.85) |
| LR | 0.81 (0.80–0.82) | 0.71 (0.68–0.73) | 0.77 (0.74–0.80) | 0.71 (0.69–0.73) | 0.65 (0.62–0.69) | 0.81 (0.78–0.84) |
| XGB | 0.81 (0.80–0.82) | 0.71 (0.68–0.74) | 0.77 (0.74–0.81) | 0.71 (0.70–0.72) | 0.66 (0.62–0.69) | 0.82 (0.79–0.84) |
| NN | 0.81 (0.80–0.82) | 0.69 (0.68–0.71) | 0.73 (0.66–0.80) | 0.74 (0.67–0.81) | 0.67 (0.60–0.73) | 0.79 (0.75–0.84) |
| Image experiment | ||||||
| RFC | 0.68 (0.65–0.70) | 0.50 (0.42–0.58) | 0.45 (0.33–0.57) | 0.77 (0.71–0.83) | 0.58 (0.53–0.62) | 0.66 (0.61–0.71) |
| SVM | 0.69 (0.66–0.71) | 0.60 (0.54–0.65) | 0.64 (0.58–0.71) | 0.64 (0.62–0.66) | 0.56 (0.50–0.62) | 0.72 (0.67–0.76) |
| LR | 0.68 (0.66–0.70) | 0.58 (0.53–0.63) | 0.60 (0.53–0.66) | 0.67 (0.65–0.69) | 0.56 (0.53–0.60) | 0.70 (0.65–0.74) |
| XGB | 0.67 (0.65–0.69) | 0.55 (0.52–0.58) | 0.56 (0.51–0.61) | 0.67 (0.63–0.71) | 0.55 (0.51–0.59) | 0.68 (0.64–0.72) |
| NN | 0.65 (0.59–0.71) | 0.49 (0.45–0.52) | 0.45 (0.37–0.52) | 0.72 (0.61–0.83) | 0.54 (0.48–0.61) | 0.65 (0.60–0.69) |
| Combination experiment | ||||||
| RFC | 0.80 (0.79–0.81) | 0.67 (0.64–0.70) | 0.66 (0.60–0.73) | 0.77 (0.72–0.82) | 0.67 (0.63–0.72) | 0.76 (0.73–0.80) |
| SVM | 0.79 (0.78–0.81) | 0.70 (0.67–0.73) | 0.78 (0.73–0.82) | 0.68 (0.66–0.71) | 0.64 (0.60–0.67) | 0.81 (0.78–0.84) |
| LR | 0.80 (0.78–0.81) | 0.70 (0.66–0.73) | 0.76 (0.72–0.80) | 0.70 (0.68–0.73) | 0.65 (0.60–0.69) | 0.80 (0.78–0.83) |
| XGB | 0.80 (0.78–0.81) | 0.69 (0.67–0.71) | 0.76 (0.72–0.79) | 0.69 (0.66–0.72) | 0.64 (0.61–0.67) | 0.80 (0.77–0.83) |
| NN | 0.78 (0.77–0.79) | 0.67 (0.65–0.68) | 0.64 (0.60–0.68) | 0.74 (0.70–0.75) | 0.66 (0.62–0.68) | 0.74 (0.68–0.76) |
Average over 5-fold cross-validation. RFC, random forest classifier; SVM, support vector machine; LR, logistic regression; XGB, gradient boosting; NN, neural networks. AUC, area under the curve; NPV, negative predictive value; PPV, positive predictive value. Values in bold indicate the best Sensitivity and Specificity values for a given experimental setup.