首页|基于18F-AV 1451 PET/MR不同衰减校正方式对阿尔茨海默病诊断效能研究

基于18F-AV 1451 PET/MR不同衰减校正方式对阿尔茨海默病诊断效能研究

扫码查看
目的 分析基于18F-AV 1451 PET/MR的不同衰减校正方式对阿尔茨海默病(AD)的诊断效能。方法 将北部战区总医院自2021 年10 月至2022 年7 月收治的接受18F-AV 1451 PET/MR检查的16 例AD患者纳入AD组,另将同期14 例健康志愿者纳入NC组。两组均行18F-AV 1451 PET/MR扫描及神经心理学测试,所有图像均经基于Atlas和ZTE的两种衰减校正方式进行重建,使用Matlab 2018b的SPM 12 进行标准化,采用AAL模板提取脑区,选取嗅皮质、后扣带回、缘上回、角回、楔前叶、颞中回、颞极-颞中回、颞下回作为感兴趣区并获得各个脑区的平均标准化摄取值(SUV),以小脑为参考得到各个脑区的标准化摄取值比(SUVr)。回顾性分析两组研究对象的临床资料,比较Atlas和ZTE在不同脑区的相对差异(RD)及对AD的诊断效能。结果 Atlas和ZTE在嗅皮质、后扣带回、楔前叶、颞极-颞中回的SUV RD大于SUVr RD,在缘上回、角回、颞中回、颞下回的SUV RD小于SUVr RD,差异有统计学意义(P<0。05)。NC组Atlas和ZTE的SUVr分别为(1。08±0。10)和(1。10±0。08),AD组分别为(1。30±0。28)和(1。33±0。28),AD组Atlas和ZTE的SUVr均高于NC组,差异有统计学意义(P<0。05)。Atlas和ZTE诊断AD的受试者工作特征(ROC)曲线下面积分别为 0。777(0。609~0。944)和 0。830(0。679~0。982),最佳截断值分别为1。122 5 和1。186 8,灵敏度分别为81。2%和75。0%,特异度分别为71。4%和85。7%,差异有统计学意义(P<0。05)。结论 基于Atlas和ZTE的衰减校正方法均是进行18F-AV 1451 定量评估的可靠方法,两种重建方式对SUV、SUVr的影响在不同脑区呈现不同趋势,基于ZTE诊断AD的ROC曲线下面积略大于Atlas,最佳截断值和特异度也更高,但灵敏度稍低,临床工作中可以按需选择。
Diagnostic efficacy of different attenuation correction methods based on 18 F-AV 1451 PET/MR in Alzheimer's disease
Objective To analyze the diagnostic efficacy of different attenuation correction methods based on 18F-AV 1451 PET/MR in Alzheimer's disease(AD).Methods A total of 16 patients with AD who received 18F-AV 1451 PET/MR examination from Octo-ber 2021 to July 2022 in the General Hospital of Northern Theater Command were included in the AD group,and 14 healthy volunteers were included in the NC group during the same period.18F-AV 1451 PET/MR scanning and neuropsychological testing were performed in both groups.All images were reconstructed using two attenuation correction methods based on Atlas and ZTE,standardized using SPM 12 of Matlab 2018b,and brain regions were extracted using AAL template.The orhinal cortex,posterior cingulate gyrus,superior marginal gyrus,angular gyrus,anterior cuneus,middle temporal gyrus,temporal polar-middle temporal gyrus and inferior temporal gyrus were selected as regions of interest,and the average standard uptake value(SUV)of each brain region was obtained,and the standard uptake value ratio(SUVr)of each brain region was obtained with the cerebellum as reference.The clinical data of the two groups were retrospectively analyzed,and the relative difference(RD)of Atlas and ZTE in different brain regions and the diagnostic efficacy of AD were compared.Results The SUV RD of Atlas and ZTE was greater than SUVr RD in olfactional cortex,posterior cingulate gyrus,an-terior cuneus and pole-middle temporal gyrus,while the SUV RD of SUVr was smaller in superior marginal gyrus,angular gyrus,middle temporal gyrus and inferior temporal gyrus,with statistical significance(P<0.05).The SUVr of Atlas and ZTE in NC group was(1.08±0.10)and(1.10±0.08),and that of AD group was(1.30±0.28)and(1.33±0.28),respectively.The SUVr of Atlas and ZTE in AD group was higher than that in NC group,and the difference was statistically significant(P<0.05).The area under receiver operating characteristic(ROC)curves of Atlas and ZTE for the diagnosis of AD were 0.777(0.609-0.944)and 0.830(0.679-0.982),the optimal cut-off values were1.122 5 and 1.186 8,and the sensitivity were 81.2%and 75.0%,respectively.The specificity was 71.4%and85.7%,respectively,and the difference was statistically significant(P<0.05).Conclusion The attenuation correction methods based on Atlas and ZTE are both reliable methods for the quantitative evaluation of 18F-AV 1451.The influence of the two re-construction methods on SUV and SUVr presents different trends in different brain regions.The area under ROC curve of AD diagnosed by ZTE is slightly larger than that of Atlas,and the optimal cut-off value and specificity are also higher,however,the sensitivity is slightly lower.

Alzheimer's disease18F-AV 1451PET/MRImage reconstructionAttenuation correction

王明丽、阮超、陆国秀、刘华伟、石庆学、庞艳波、王治国

展开 >

大连医科大学北部战区总医院研究生培养基地,辽宁 沈阳 110016

北部战区总医院核医学科,辽宁 沈阳 110016

GE医疗中国分子影像事业部,北京 100176

北部战区总医院干部病房三科,辽宁 沈阳 110016

展开 >

阿尔茨海默病 18F-AV1451 PET/MR 图像重建 衰减校正

辽宁省重点研发资助计划辽宁省应用基础研究计划(联合计划)

2019JH2/103000102022JH2/101500011

2024

临床军医杂志
解放军沈阳军区卫生人员训练基地

临床军医杂志

CSTPCD
影响因子:0.465
ISSN:1671-3826
年,卷(期):2024.52(3)
  • 1
  • 20