Table 2 The comparison results of the five evaluation sub-parts under five different detection models
From: Deep learning-driven multi-view multi-task image quality assessment method for chest CT image
Sub-parts | Evaluation metrics | YOLOv8 (avg ± std) | YOLOv7 (avg ± std) | RetinaNet (avg ± std) | CenterNet (avg ± std) | Faster R-CNN (avg ± std) |
|---|---|---|---|---|---|---|
Artifact | Precision | 1.00 ± 0.00 | 1.00 ± 0.00 | 0.99 ± 0.00 | 0.00 ± 0.00 | 0.95 ± 0.01 |
Sensitivity | 1.00 ± 0.00 | 1.00 ± 0.00 | 0.98 ± 0.02 | 0.00 ± 0.00 | 1.00 ± 0.00 | |
F1-score | 1.00 ± 0.00 | 1.00 ± 0.00 | 0.99 ± 0.01 | 0.00 ± 0.00 | 0.97 ± 0.01 | |
Arms position | Precision | 0.99 ± 0.00 | 0.96 ± 0.00 | 0.96 ± 0.00 | 0.00 ± 0.00 | 0.88 ± 0.00 |
Sensitivity | 0.96 ± 0.00 | 0.91 ± 0.00 | 0.89 ± 0.00 | 0.00 ± 0.00 | 0.91 ± 0.00 | |
F1-score | 0.97 ± 0.00 | 0.94 ± 0.00 | 0.92 ± 0.00 | 0.00 ± 0.00 | 0.90 ± 0.00 | |
Radiation protection | Precision | 0.99 ± 0.02 | 0.99 ± 0.02 | 1.00 ± 0.00 | 0.50 ± 0.71 | 0.88 ± 0.04 |
Sensitivity | 0.97 ± 0.01 | 0.90 ± 0.04 | 0.52 ± 0.52 | 0.01 ± 0.02 | 0.95 ± 0.03 | |
F1-score | 0.98 ± 0.00 | 0.94 ± 0.01 | 0.60 ± 0.47 | 0.02 ± 0.03 | 0.92 ± 0.03 | |
Rib | Precision | 0.90 ± 0.08 | 0.77 ± 0.01 | 0.79 ± 0.13 | 0.25 ± 0.50 | 0.16 ± 0.19 |
Sensitivity | 0.82 ± 0.11 | 0.71 ± 0.15 | 0.74 ± 0.14 | 0.01 ± 0.02 | 0.14 ± 0.18 | |
F1-score | 0.85 ± 0.06 | 0.73 ± 0.01 | 0.76 ± 0.13 | 0.02 ± 0.03 | 0.14 ± 0.17 | |
Bronchial beam | Precision | 0.90 ± 0.09 | 0.63 ± 0.10 | 0.86 ± 0.07 | 0.60 ± 0.43 | 0.37 ± 0.04 |
Sensitivity | 0.92 ± 0.01 | 0.62 ± 0.09 | 0.88 ± 0.03 | 0.36 ± 0.25 | 0.44 ± 0.15 | |
F1-score | 0.91 ± 0.05 | 0.62 ± 0.10 | 0.87 ± 0.03 | 0.44 ± 0.29 | 0.40 ± 0.09 |