Table 4 The state-of-the-art performances achieved in previous works about DMI prediction by radiomic-based binary models.
Paper | Image | Sample size | Methods | Performance |
|---|---|---|---|---|
Xiong, L. et al. (2023)14 | MRI | 154 | Ellipse fitting algorithm | AUC: 89% Accuracy: 87% |
Stanzione, A. (2021)15 | MRI | 54 | PyRadiomics – Random forest | AUC: 89% Accuracy: 87% |
Chen, X. (2020)16 | MRI | 530 | DL network | AUC: 89% Accuracy: 85% |
Zhun, X. (2021)17 | MRI | 79 | Geometric and texture features – SVM classifier | AUC: 92% Accuracy: 94% |
Ueno, Y. (2017)18 | MRI | 137 | Texture feauters – RF classifier | AUC: 84% Accuracy: 81% |
Liu, Xiaoling, et al. (2024)19 | US | 604 | EfficientNet-B6 | AUC: 81% Accuracy: 80% |
Our proposal (mod1) | US | 77 | Pre-trained Inception-V3 – SVM classifier | AUC: 91% Accuracy: 89% |
