TABLE 3.
Validation of inflection‐point algorithm on MV DRR images
| Patient | Inflection‐point algorithm (cm) | Manual measurement (cm) | Difference (mm) |
|---|---|---|---|
| P1 | 2.25 ± 0.08 | 2.20 ± 0.14 | 0.5 |
| P2 | 2.34 ± 0.08 | 2.38 ± 0.17 | 0.4 |
| P3 | 1.90 ± 0.12 | 1.94 ± 0.14 | 0.4 |
| P4 | 1.86 ± 0.06 | 1.91 ± 0.35 | 0.5 |
| P5 | 1.95 ± 0.07 | 1.74 ± 0.38 | 2.1 |
| P6 | 2.05 ± 0.13 | 1.97 ± 0.22 | 0.8 |
| P7 | 1.66 ± 0.19 | 1.62 ± 0.34 | 0.4 |
| P8 | 2.44 ± 0.07 | 2.11 ± 0.47 | 3.3 |
| P9 | 1.95 ± 0.12 | 1.78 ± 0.24 | 1.7 |
| P10 | 1.95 ± 0.05 | 1.81 ± 0.45 | 1.4 |
Results of the inflection‐point algorithm validation. The median chest wall position in a 31‐row region of interest was recorded with the median measurement obtained from 10 patients' MV DRR images. These were compared to the median of 10 manual measurements taken from the same region on the 10 patients' kV DRR images in Offline Review.