SUV metrics | Current use in clinical settings | Major drawbacks or advantages | References | Statistically significant effects in current study |
---|---|---|---|---|
SUVmean | + | • PVE at the edges of target region | No | |
• Time consuming to outline target region | ||||
• Inclusion of nontarget tissue | ||||
SUVmax | + | • Susceptible to noise | No | |
• Random voxel location | ||||
• Independent of observer | ||||
• Easy to apply | ||||
SUVthreshold | + | • Inclusion of nontarget tissue | No | |
• Less sensitive to noise | ||||
• Easy to apply | ||||
• More reproducible than SUVmean | ||||
SUVfixed volume (placement adjusted in current study) | + | • Less sensitive to noise than SUVmax | Yes | |
• Limits PVE at edges of target region | ||||
• Relatively similar location of voxels between scans and between animals | ||||
mSUVthreshold | - | • Less sensitive to noise than SUVmax |  | Yes |
• Relatively easy to apply | ||||
• Limits PVE at edges of target region |