Figure 4.

Counting rates expected on lesions (Signal) versus the ones expected on healthy tissue. The dashed line shows the bisector, representing the case in which the probe gives the exact same count over signal than over healthy tissue, having thus no sensitivity at all. In addition to the usual division among local lesions and lymph nodes, points are also classified (“Good” or “Bad”) according to the ROC sensitivity test described in the text, having fixed a probing time of 3 s and requiring $AUC>0.95$. The rates correspond to the application case described in the text (injection of 210 MBq of ⁶⁸Ga-PSMA $150\min$ before the surgery).