Skip to main content
Fig. 10 | BioMedical Engineering OnLine

Fig. 10

From: Respiratory motion correction of PET using MR-constrained PET-PET registration

Fig. 10

Illustrating method robustness with scatter plots of \(\mathrm {SUV}_\mathrm {peak}\) recovery. These scatter plots show the improved robustness of the two MR-derived motion model based techniques (DC and IC) compared to the direct registration PET-PET registration technique (PT). Each plot compares distributions of \(\mathrm {SUV}_\mathrm {peak}\) for each lesion in positions 4–9 for all volunteers, both before correction and after correction by the IC, DC and PT methods. The dashed line represents \(y=x\). Any point on this line was unaffected by the correction attempt. Points above the line have had their SUVpeak increased (with 100 % corresponding to the motionless peak uptake), whereas points below the line show a reduction in SUVpeak. Note in particular that whilst none in the IC case are significantly below the line, 8 lesions dropped in SUVpeak with the PT method (16.7 % of the points shown)

Back to article page