Premium
Performance evaluation of a high resolution dedicated breast PET scanner
Author(s) -
García Hernández Trinitat,
Vicedo González Aurora,
Ferrer Rebolleda Jose,
Sánchez Jurado Raúl,
Roselló Ferrando Joan,
Brualla González Luis,
Granero Cabañero Domingo,
del Puig Cozar Santiago Maria
Publication year - 2016
Publication title -
medical physics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.473
H-Index - 180
eISSN - 2473-4209
pISSN - 0094-2405
DOI - 10.1118/1.4945271
Subject(s) - scanner , medical physics , mammography , medical imaging , nuclear medicine , dosimetry , image resolution , medicine , computer science , breast cancer , radiology , artificial intelligence , cancer
Purpose: Early stage breast cancers may not be visible on a whole‐body PET scan. To overcome whole‐body PET limitations, several dedicated breast positron emission tomography (DbPET) systems have emerged nowadays aiming to improve spatial resolution. In this work the authors evaluate the performance of a high resolution dedicated breast PET scanner (Mammi‐PET, Oncovision). Methods: Global status, uniformity, sensitivity, energy, and spatial resolution were measured. Spheres of different sizes (2.5, 4, 5, and 6 mm diameter) and various 18 fluorodeoxyglucose ( 18 F‐FDG) activity concentrations were randomly inserted in a gelatine breast phantom developed at our institution. Several lesion‐to‐background ratios (LBR) were simulated, 5:1, 10:1, 20:1, 30:1, and 50:1. Images were reconstructed using different voxel sizes. The ability of experienced reporters to detect spheres was tested as a function of acquisition time, LBR, sphere size, and matrix reconstruction voxel size. For comparison, phantoms were scanned in the DbPET camera and in a whole body PET (WB‐PET). Two patients who just underwent WB‐PET/CT exams were imaged with the DbPET system and the images were compared. Results: The measured absolute peak sensitivity was 2.0%. The energy resolution was 24.0% ± 1%. The integral and differential uniformity were 10% and 6% in the total field of view (FOV) and 9% and 5% in the central FOV, respectively. The measured spatial resolution was 2.0, 1.9, and 1.7 mm in the radial, tangential, and axial directions. The system exhibited very good detectability for spheres ≥4 mm and LBR ≥10 with a sphere detection of 100% when acquisition time was set >3 min/bed. For LBR = 5 and acquisition time of 7 min the detectability was 100% for spheres of 6 mm and 75% for spheres of 5, 4, and 2.5 mm. Lesion WB‐PET detectability was only comparable to the DbPET camera for lesion sizes ≥5 mm when acquisition time was >3 min and LBR > 10. Conclusions: The DbPET has a good performance for its clinical use and shows an improved resolution and lesion detectability of small lesions compared to WB‐PET.