Table 3 Classification performance of DeepAutoGlioma on external datasets
Methods | Performance measures (Average of 10 fold cross-validation) | |||||||
|---|---|---|---|---|---|---|---|---|
Accuracy [95% CI] | Precision [95% CI] | Recall [95% CI] | F1-score [95% CI] | FPR [95% CI] | Gmean [95% CI] | MCC [95% CI] | ||
LGG | ANN | 91.89% | 91.20% | 88.00% | 86.90% | 0.06 | 92.13% | 0.83 |
[91.85–91.93] | [91.15–91.25] | [87.94–88.06] | [86.83–86.97] | [0.02–0.10] | [92.09–92.17] | [0.74–0.92] | ||
CNN | 91.38% | 91.38% | 91.38% | 91.38% | 0 | 1 | 1 | |
[91.34–91.40] | [91.34–91.40] | [91.34–91.40] | [91.34–91.40] | [0–0] | [1–1] | [1–1] | ||
GBM | ANN | 84.10% | 74.48% | 79.48% | 76.15% | 0.06 | 90.55% | 0.72 |
[84.01–84.18] | [74.35–74.60] | [79.37–79.58] | [76.03–76.26] | [0.02–0.10] | [90.49–90.60] | [0.58–0.86] | ||
CNN | 86.41% | 79.87% | 83.33% | 81.02% | 0.05 | 92.92% | 0.76 | |
[86.33–86.49] | [79.75–79.99] | [83.23–83.43] | [80.90–81.13] | [0.01–0.09] | [92.87–92.97] | [0.62–0.90] | ||