Table 2.
The impact of motion correction on the interpretation of cardiac 15O-water PET results in 170 scans with simulated motion, based on a 2.3 mL⋅min−1⋅g−1 threshold for MBFstress
| Pre-analysis approach | Post-analysis approach | ||||
|---|---|---|---|---|---|
| Changed (#) | True (#) | Changed (#) | True (#) | ||
| Global | Positive to negative | 5 | 4 | 2 | 2 |
| Negative to positive | 17 | 2 | 15 | 4 | |
| LAD | Positive to negative | 5 | 5 | 5 | 5 |
| Negative to positive | 11 | 10 | 12 | 10 | |
| RCA | Positive to negative | 14 | 4 | 2 | 2 |
| Negative to positive | 24 | 4 | 16 | 3 | |
| LCx | Positive to negative | 9 | 5 | 2 | 1 |
| Negative to positive | 15 | 6 | 12 | 4 | |
| Any | Positive to negative | 29 | 15 | 8 | 7 |
| Negative to positive | 40 | 18 | 38 | 16 | |
Based on changes in MBF globally or in any of the coronary territories, motion correction using the pre-analysis approach altered the interpretation of 69 scans, while using the post-analysis approach, the interpretation was altered in 46. The change in interpretation was considered true, if the post-motion correction interpretation agreed with the interpretation of the original motion-free scan