ORIGINAL ARTICLES   

The Quarterly Journal of Nuclear Medicine 2003 June;47(2):90-100

Copyright © 2009 EDIZIONI MINERVA MEDICA

language: English

Evaluation of the clinical performances of a large NaI(Tl) crystal 3D PET scanner

Picchio M. ¹, Savi A. ¹, Lecchi M. ², Landoni C. ¹, Gianolli L. ¹, Brioschi M. ¹, Rossetti C. ³, Gilardi M. C. ¹, Fazio F. ¹

¹ Department of Nuclear Medicine Scientific Institute H. S. Raffaele University of Milano-Bicocca, IBFM-CNR, Milan, Italy 2 Institute of Radiological Sciences University of Milan, Milan, Italy 3 Department of Nuclear Medicine Hospital Niguarda Ca’ Granda, Milan, Italy

Aim. This ­study was ­aimed at assess­ing the clin­i­cal per­for­manc­es of a NaI(Tl) crys­tal 3D PET scan­ner, C-PET (­ADAC-UGM), ­using a mul­ti-­ring 2D BGO PET scan­ner (mul­ti-­ring PET), as a ref­er­ence.
Methods. Thirty-sev­en onco­log­i­cal ­patients ­were stud­ied in ­sequence ­with mul­ti-­ring PET and C-PET, with­in 30 ­days of a CT ­study. In ­order to ­assess the beha­vi­our of C-PET in rela­tion to acqui­si­tion ­count ­rate, ­patients ­were divid­ed ­into 3 ­groups accord­ing to the ­count ­rate at the ­time of the C-PET ­scan acqui­si­tion. Group A (n=21): 3000-5000 ­kcounts/sec (rec­om­mend­ed ­count ­rate ­range); Group B (n=8): <3000 Kcounts/sec and Group C (n=8): >5000 Kcounts/sec.
Results. The num­ber of ­lesions detect­ed by mul­ti-­ring PET and C-PET, clas­si­fied accord­ing to ­size, was com­pared. For Group A and Group B ­there was a ­good agree­ment ­between C-PET and mul­ti-­ring PET in ­terms of ­lesion detect­abil­ity (rel­a­tive sen­si­tiv­ity: 99.9% and 96.0%, respec­tive­ly), ­while for Group C the rel­a­tive sen­si­tiv­ity of C-PET was 61.9%.
Conclusion. Optimal per­for­manc­es of the C-PET scan­ner can ­thus be ­obtained at a ­count ­rate with­in or ­below the rec­om­mend­ed ­range. Despite a low­er ­lesion/back­ground con­trast result­ing ­from a ­high scat­ter and ran­dom ­noise, the sen­si­tiv­ity of C-PET in detect­ing hyper­met­a­bol­ic ­lesions is com­par­able to ­that of mul­ti-­ring PET. These find­ings are dis­cussed in rela­tion to the phys­i­cal per­for­mance of the two scan­ners and par­tic­u­lar­ly in rela­tion to the 3D vs 2D acqui­si­tion modal­ity.