Evaluation of Dixon Sequence on Hybrid PET/MR Compared with Contrast-Enhanced PET/CT for PET-Positive Lesions
Evaluation of Dixon Sequence on Hybrid PET/MR Compared with Contrast-Enhanced PET/CT for PET-Positive Lesions
复制标题
DOI：
10.1007/s13139-013-0240-6
复制
发表时间：
2014-03-01
影响因子：
1.3
通讯作者：
Chun, Kyung Ah
中科院分区：
文献类型：
--
作者：
Jeong, Ju Hye;Cho, Ihn Ho;Chun, Kyung Ah
目的混合正电子发射断层成像和磁共振（PET/MR）成像进行两点Dixon MR序列衰减校正。然而，混合PET/MR的MR数据应该提供解剖和形态学信息以及衰减图。我们评估了混合PET/MR的Dixon序列对肿瘤患者中PET阳性病变的解剖相关性，并与增强PET/计算机断层扫描（CT）进行了比较。方法12例患者采用单次注射双显像方案。经静脉注射造影剂PET/ ct (F-18-FDG注射403 +/- 45 MBq后85 +/- 13 min), PET/MR（注射后125 +/- 19 min）。PET/CT采用对比增强CT，混合PET/MR采用Dixon MR序列进行衰减校正和PET解剖分配。比较Dixon MR序列和增强CT对pet阳性病变的解剖相关性（视觉评分评分范围从0到3）。此外，对检测到的病变进行标准化摄取值（suv）评估以进行定量比较。结果混合PET/ mr和增强PET/CT在10例患者中发现55个病灶FDG摄取增加。共出现28个淋巴结、11个骨病变、3个真皮结节、3个胸膜增厚病变、2个甲状腺结节、1个胰腺、1个肝脏、1个卵巢、1个子宫、1个乳房、1个软组织和2个肺部病变。CT增强扫描对PET表现的解剖相关性表现最佳（增强CT， 2.64 +/- 0.70；同相，1.29 +/- 1.01；反相，1.29 +/- 1.15；水加权，1.71 +/- 1.07；脂肪加权，0.56 +/- 1.03）。对比增强CT和所有四张共登记的Dixon MR图像的评分有显著差异。定量评估显示，PET/MR混合成像测量的suv （SUVmean, 2.63 +/- 1.62; SUVmax, 4.30 +/- 2.88）与PET/CT增强成像（SUVmean, rho=0.93; SUVmax, rho=0.95）在PET阳性病变（SUVmean, rho=0.93; SUVmax, rho=0.95）之间具有高度相关性，尽管混合成像PET/MR比PET/CT增强成像显示的suv减少（平均减少；SUVmean, 32.44 +/- 15.64%; SUVmax, 35.16 +/- 12.59%）。结论尽管衰减校正方法不同，PET阳性病变的SUV在混合PET/MR和增强PET/CT之间具有良好的相关性。但衰减校正所获得的Dixon MR图像由于空间分辨率较低，无法提供PET图像的解剖信息。因此，可能需要额外的快速和更高分辨率的MR序列来获取解剖信息。
Purpose Hybrid positron emission tomography and magnetic resonance (PET/MR) imaging performs a two-point Dixon MR sequence for attenuation correction. However, MR data in hybrid PET/MR should provide anatomic and morphologic information as well as an attenuation map. We evaluated the Dixon sequence of hybrid PET/MR for anatomic correlation of PET-positive lesions compared with contrast-enhanced PET/computed tomography (CT) in patients with oncologic diseases.Methods Twelve patients underwent a single injection, dual imaging protocol. PET/CTwas performed with an intravenous contrast agent (85 +/- 13 min after F-18-FDG injection of 403 +/- 45 MBq) and then (125 +/- 19 min after injection) PET/MR was performed. Attenuation correction and anatomic allocation of PET were performed using contrast-enhanced CT for PET/CT and Dixon MR sequence for hybrid PET/MR. The Dixon MR sequence and contrast-enhanced CT were compared for anatomic correlation of PET-positive lesions (scoring scale ranging from 0 to 3 for visual ratings). Additionally, standardized uptake values (SUVs) for the detected lesions were assessed for quantitative comparison.Results Both hybrid PET/MRand contrast-enhanced PET/CT identified 55 lesions with increased FDG uptake in ten patients. In total, 28 lymph nodes, 11 bone lesions, 3 dermal nodules, 3 pleural thickening lesions, 2 thyroid nodules, 1 pancreas, 1 liver, 1 ovary, 1 uterus, 1 breast, 1 soft tissue and 2 lung lesions were present. The best performance was observed for anatomic correlation of PET findings by the contrast-enhanced CT scans (contrast-enhanced CT, 2.64 +/- 0.70; in-phase, 1.29 +/- 1.01; opposed-phase, 1.29 +/- 1.15; water-weighted, 1.71 +/- 1.07; fat weighted, 0.56 +/- 1.03). A significant difference was observed between the scores obtained from the contrast-enhanced CT and all four coregistered Dixon MR images. Quantitative evaluation revealed a high correlation between the SUVs measured with hybrid PET/MR (SUVmean, 2.63 +/- 1.62; SUVmax, 4.30 +/- 2.88) and contrast-enhanced PET/CT (SUVmean, 3.88 +/- 2.30; SUVmax, 6.53 +/- 4.04) in PET-positive lesions (SUVmean, rho=0.93; SUVmax, rho=0.95), although hybrid PET/MR presented a decrease of SUVs compared with contrast-enhanced PET/CT (mean reduction; SUVmean, 32.44 +/- 15.64 %; SUVmax, 35.16 +/- 12.59 %).Conclusions Despite different attenuation correction approaches, the SUV of PET-positive lesions correlated well between hybrid PET/MR and contrast-enhanced PET/CT. However Dixon MR images acquired for attenuation correction were insufficient to provide anatomic information of PET images because of low spatial resolution. Thus, additional MR sequence with fast and higher resolution may be necessary for anatomic information.