Table 3 ILD database: performance of the proposed method using five configurations of training data and three configurations of the segmentation system
From: Three-stage segmentation of lung region from CT images using deep neural networks
Training data | Dice score | |||
|---|---|---|---|---|
Training images | Training labels | Three-stage system | Without CNNs | Without Contrast Enhancement |
Ground truth images | Ground truth images | 0.87 | 0.85 | 0.81 |
Otsu-based binary images | Otsu-based binary images | 0.92 | 0.91 | 0.83 |
Foreground images | Foreground images | 0.88 | 0.84 | 0.79 |
Grayscale images | Two-class k-means clustered images | 0.93 | 0.87 | 0.85 |
Grayscale images | Three-class k-means clustered images | 0.95 | 0.86 | 0.87 |