Table 1 The five different configurations of the training data
From: Three-stage segmentation of lung region from CT images using deep neural networks
| Training data | |
|---|---|
| Training images | Training labels |
| Ground truth images | Ground truth images |
| Otsu-based binary images | Otsu-based binary images |
| Foreground images | Foreground images |
| Grayscale images | Two-class k-means clustered images |
| Grayscale images | Three-class k-means clustered images |