Table 1.
Quantitative results and p‐value obtained by comparison between proposed method and RF‐based method. For all tests, the planning CT image was taken as the ground truth image
| ME (HU) | MAE (HU) | PSNR (dB) | NCC | SNU | ||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Pelvis | Brain | Pelvis | Brain | Pelvis | Brain | Pelvis | Brain | Pelvis | Brain | |
| CBCT | 3 ± 29 | −2 ± 4 | 56.3 ± 19.7 | 23.8 ± 5.1 | 22.2 ± 3.4 | 32.3 ± 5.9 | 0.96 ± 0.02 | 0.98 ± 0.02 | 0.26 ± 0.16 | 0.15 ± 0.07 |
| RF | −5 ± 5 | 0 ± 1 | 17.7 ± 6.5 | 13.1 ± 2.0 | 28.0 ± 4.1 | 34.6 ± 1.7 | 0.95 ± 0.01 | 0.98 ± 0.02 | 0.16 ± 0.08 | 0.06 ± 0.05 |
| Proposed | −3 ± 4 | 0 ± 1 | 16.1 ± 4.5 | 13.0 ± 2.2 | 30.7 ± 3.7 | 37.5 ± 2.3 | 0.98 ± 0.01 | 0.99 ± 0.00 | 0.09 ± 0.05 | 0.05 ± 0.04 |
| P‐value (proposed vs. RF) | 0.19 | 0.17 | 0.24 | 0.78 | <0.001 | <0.001 | <0.001 | 0.12 | <0.001 | 0.04 |
RF, random forest; MAE, mean absolute error; PSNR, peak signal‐to‐noise ratio; NCC, normalized cross‐correlation; SNU, spatial nonuniformity; CBCT, cone‐beam computed tomography.