Table 6 Performance metrics of proposed models compared with earlier research using Chi-square FS
| Â | Methods | Database | Results (%) | ||||
|---|---|---|---|---|---|---|---|
AUROC | Accuracy | Sensitivity | Precision | Specificity | |||
Cai, J et al. [43] | Deep Learning | LIDC | 88.1 | 84.6 | 83.7 | – | 85.2 |
Wang et al. [44] | Deep Learning | LIDC | 92.75 | 85.23 | 92.79 | 84.56 | 72.89 |
Proposed model 10 | Radiomics + FGSVM | LIDC | 91.1 | 86.5 | 95.7 | 87.2 | 61.5 |
Proposed model 11 | Radiomics + MGSVM | LIDC | 91.0 | 90.4 | 93.0 | 94.1 | 82.4 |
Proposed model 12 | Radiomics + CGSVM | LIDC | 91.0 | 86.9 | 92.2 | 90.1 | 72.3 |
Proposed model 13 | Radiomics + Decision Tree | LIDC | 88.3 | 85.7 | 92.6 | 88.5 | 67.0 |
Proposed model 14 | Radiomics + Ensemble Boosted Trees | LIDC | 92.3 | 87.7 | 92.3 | 91.0 | 75.0 |
Proposed Model 15 | Radiomics + Ensemble Bagged Trees | LIDC | 92.9 | 89.1 | 93.5 | 91.8 | 77.2 |
Proposed model 16 | Radiomics + Ensemble RUSBoosted Trees | LIDC | 92.5 | 86.1 | 86.9 | 93.7 | 84.0 |
Proposed model 17 | Radiomics + Ensemble Subspace Discriminant | LIDC | 91.1 | 86.0 | 94.8 | 87.2 | 62.1 |
Proposed model 18 | Radiomics + Ensemble Subspace KNN | LIDC | 89.6 | 86.5 | 93.0 | 89.1 | 68.9 |