F. Montravers et al., DIGITAL SUPERIMPOSITION OF CT AND POSITIVE SPECT TUMOR IMAGES - PHANTOM STUDY AND CLINICAL-APPLICATIONS, Clinical nuclear medicine, 22(3), 1997, pp. 151-157
Digital superimposition of SPECT and CT data was evaluated in a phanto
m and then applied to patient data. Seven patients were studied. Six p
atients had pheochromocytomas as evidenced by I-131 or I-123 MIBG loca
lization and one had ovarian cancer imaged by In-ill OC125 MoAb. Anato
mic or skin landmarks identified the level of each SPECT transaxial sl
ice. Both SPECT and CT image data sets were transferred to a minicompu
ter connected to an image processor. Afterwards, a scaled, rotated, an
d translated realignment was performed. Data for each modality were co
ded in different primary colors and then superimposed. Superimposition
of phantom data was checked for the absence of distortion of pinpoint
and large structures. For suspected tumor sites, superimposition of t
he patients' slices were allowed to check for matching SPECT and CT ab
normalities to localize a SPECT abnormality without a corresponding CT
lesion or to distinguish SPECT abnormalities from those seen on CT. I
n one case, the technique failed because of very low I-131 MIBG-tumor
uptake. The superimposition decreases false positives in SPECT and bot
h false negatives or false positives in CT.