Table 6 The evaluation time of the M\(^2\)IQA method and the MOS on different datasets
From: Deep learning-driven multi-view multi-task image quality assessment method for chest CT image
| Â | Dataset | ||||||
|---|---|---|---|---|---|---|---|
Additional test set | LungCT-Diagnosis | CMB-LCA | |||||
M\(^2\)IQA | MOS | M\(^2\)IQA | MOS | M\(^2\)IQA | MOS | ||
|  | Tracheal carina | 0.00 ± 0.00 | 4.19 ± 1.24 | 0.00 ± 0.00 | 5.97 ± 1.61 | 0.00 ± 0.00 | 6.16 ± 1.16 |
|  | Bronchial beam | 14.67 ± 1.78 | 30.21 ± 3.68 | 3.77 ± 0.91 | 10.45 ± 2.21 | 5.21 ± 1.68 | 8.84 ± 2.60 |
|  | Rib | 6.75 ± 0.95 | 25.95 ± 4.00 | 1.79 ± 0.41 | 21.02 ± 5.78 | 2.52 ± 0.87 | 17.15 ± 2.68 |
|  | Arms position | 0.04 ± 0.03 | 0.93 ± 0.24 | N/A | N/A | N/A | N/A |
Evaluation time (s) (avg ± std) | Scan baseline position | 0.14 ± 0.15 | 1.36 ± 0.40 | N/A | N/A | N/A | N/A |
|  | Body position | 0.01 ± 0.00 | 3.64 ± 0.93 | 0.01 ± 0.00 | 4.36 ± 1.09 | 0.01 ± 0.00 | 5.05 ± 0.97 |
|  | Radiation protection | 0.03 ± 0.03 | 1.01 ± 0.31 | N/A | N/A | N/A | N/A |
|  | Artifact | 1.59 ± 0.78 | 6.08 ± 2.95 | 2.10 ± 1.11 | 6.91 ± 2.77 | 2.74 ± 1.19 | 7.57 ± 3.93 |
|  | Total | 23.02 ± 2.86 | 70.07 ± 9.07 | 7.67 ± 2.15 | 48.72 ± 9.44 | 10.48 ± 3.29 | 44.78 ± 6.39 |