Accuracy of image fusion of normal upper abdominal organs visualized with PET/CT

Abstract

Although PET/CT scanners have the potential for precise fused registration of structures visualized on both PET and CT, physiological motion during the acquisition of both studies may alter the appearance of organ shape, size or location. The degree of possible mismatch in abdominal organ size and position between PET and CT has not been evaluated. The aim of this study was to assess the consistency in location and measured size of upper abdominal organs with PET and CT using a combined PET/CT system. Forty-six consecutive inpatients who underwent clinical PET/CT scans for suspected cancer were evaluated. CT and PET images attenuation corrected by both CT and germanium-68 transmission scans were obtained, and we separately determined the location of the top and bottom (height), anterior and posterior margins (thickness), and right and left margins (width) for each organ, including liver, spleen, and bilateral kidneys, using CT and both sets of PET images. Differences between the two modalities in terms of location and measured organ size were investigated. In the upper margin of the liver and lower margin of the spleen, more than 10% of the cases showed a larger discrepancy (>20 mm) between CT-based and Ge-corrected PET-based measurements, although the differences in the positions of the edges were less than 10 mm in most cases. The center of the liver tended to be located cephalad and to the right of the body, and that of the spleen tended to be cephalad and posterior on PET, as compared with CT. Moreover, the center of both kidneys tended to be seen cephalad, posterior, and to the right on PET. The liver appeared slightly larger on PET than CT in thickness (CT vs CT-corrected PET vs Ge-corrected PET = 156 mm vs 162 mm vs 162 mm) and width (186 mm vs 189 mm vs 188 mm). By contrast, the spleen appeared slightly smaller on PET than CT in height (84 mm vs 77 mm vs 80 mm) and width (85 mm vs 81 mm vs 80 mm). A similar tendency was observed in the left kidney (105 mm vs 100 mm vs 99 mm in height, and 64 mm vs 59 mm vs 58 mm in width) and the right kidney (99 mm vs 93 mm vs 93 mm in height, and 64 mm vs 59 mm vs 60 mm in width). These differences between the two modalities were statistically significant ( P<0.05). In conclusion, minor mismatches in location and organ size were found to exist between CT and PET images, in part due to physiological motion. Although these differences could potentially affect the quality of the image registrations, they were generally of a modest nature.

Evaluation of Image Registration in PET/CT of the Liver and Recommendations for Optimized Imaging

“…Attenuation correction will be applied erroneously on PET at the location of dense objects (26,27) or where the position of a transition from low to high photon-attenuating tissue does not correspond on PET and CT (13,22). The diaphragmatic area is very susceptible to such errors due to the sharp tissue/air transition, combined with the risk for positional differences.…”

PET/CT scanner instrumentation, challenges, and solutions

“…Several organs, especially the liver and spleen, experience deviations both in volume and position during respiration. Nakamoto et al [30] showed that when using a PET/CT protocol calling for no breathholds, the liver in the PET image is on average slightly larger (1% -7% larger in some dimensions) than the CT and the spleen is slightly smaller. As expected, the greatest respiratory deviations occur in the upper margin of the liver and lower margin of the spleen.…”

Clinical evaluation of a breathing protocol for PET/CT

“…These devices allow for the fast and correct co-registration of anatomical and functional information in the same imaging session. Motion presents a challenge in the thoracic cage, and several studies have evaluated the influence of respiration on the quality of image co-registration [1,2,3,4,5]. It has been described that the acquisition of CT data during shallow tidal breathing (STB) can induce artifacts with a typical curvilinear shape cranial to both diaphragms on coronal PET and PET/CT images [6,7]; therefore, we started to acquire CT data during breath-hold (BH) in the normal expiration position.…”