Table 6 Effect of the attention modules on the model performance.
From: DLA-Net: dual lesion attention network for classification of pneumoconiosis using chest X-ray images
Method | Multi-class classification | ||||
|---|---|---|---|---|---|
ACC | SEN | SPE | F1 | AUC | |
Base | 0.8279 ± 2.55 | 0.7189 ± 3.04 | 0.8838 ± 1.93 | 0.7294 ± 2.92 | 0.8013 ± 2.23 |
Base + CA | 0.8425 ± 0.54 | 0.7356 ± 1.10 | 0.8900 ± 0.85 | 0.7447 ± 0.93 | 0.8128 ± 0.69 |
Base + SA | 0.8372 ± 1.62 | 0.7404 ± 2.06 | 0.8779 ± 1.88 | 0.7447 ± 2.24 | 0.8091 ± 1.88 |
Base + CA + SA (Parallelly) | 0.8520 ± 2.23 | 0.7556 ± 3.10 | 0.8934 ± 1.54 | 0.7616 ± 2.86 | 0.8245 ± 2.23 |
Base + CA + SA (Sequentially) | 0.8562 ± 1.61 | 0.7643 ± 1.44 | 0.8941 ± 1.81 | 0.7687 ± 1.53 | 0.8292 ± 1.53 |
