数字化PET/CT步进式短采集时间的临床应用可行性分析OACSTPCD
Analysis on the feasibility of short collection time of step by step mode of digital PET/CT
目的:分析数字化正电子发射断层显像与计算机断层显像(PET/CT)短采集时间对图像质量、病灶检出和半定量参数的影响,探讨步进式采集方式短采集时间的临床应用可行性.方法:回顾性分析63例受检者数字化PET/CT图像资料,所有图像均采用PET/CT扫描仪以步进式采集模式及按照列表模式(list-mode)扫描协议进行全身PET/CT检查获得,按照每床位的不同采集时间进行重建并将其分为4组:90 s/床位、60 s/床位、45 s/床位和30 s/床位组.以90 s/床位组为对照,比较各组间小病灶(PET图像较小的放射性浓聚灶且CT图像病灶短径<0.5 cm)在PET图像上的检出率,测量原发肿瘤病灶的客观数据,比较标准化摄取值(SUV)的最大值(SUVmax)、峰值(SUVpeak)、平均值(SUVmean)、噪声标准差(SD)、肿瘤代谢体积(MTV)和病灶总糖酵解量(TLG);计算并比较信噪比(SNR)、对比噪声比(CNR)和病灶本底比(TBR).4组图像数据组间采用统计学单因素方差分析(ANOVA检验),组内半定量参数采用Bland-Altman一致性分析.图像质量按照Likert 5分法进行主观评价,图像一致性评价采用Kappa检验.结果:4组图像质量主观评价均≥3分,4组图像质量比较一致性非常好(Kappa=0.8,P<0.05);与90 s/床位组比较,其余3组小病灶检出率均为100%;组间半定量指标SUVmax、SUVpeak、SUVmean、噪声SD、TLG、MTV差异均无统计学意义(P>0.05);4组间病灶18F氟代脱氧葡萄糖(18F-FDG)PET图像质量客观指标CNR、SNR及TBR差异均有统计学意义(F=31.741、34.199、12.021,P<0.001),60 s/床位组、45 s/床位组噪声与90 s/床位组无明显差异,30 s/床位组图像噪声有所增加;组内Bland-Altman一致性分析在95%可信任区间.结论:数字化PET/CT步进式采集模式短采集时间具有很好的临床实用性.
Objective:To analyze the effect of short collection time of digital positron emission tomography/computed tomography(PET/CT)on image quality,detected lesion and semi-quantitative parameters,and explore the feasibility of clinical application of short collection time of step by step collection mode.Methods:The digital PET/CT images of 63 subjects were retrospectively analyzed.All of them adopted PET/CT scan to conduct PET/CT examination for whole body according to the collection mode of step by step type and list-mode scan protocol.The reconstruction was implemented according to the different collection time of each bed,and they were divided into 90 s/bed group,60 s/bed group,45 s/bed group and 30 s/bed group.The 90s/bed was used as control group to compare the detection rate of the small lesion of each group("the shorter diameter of smaller radioactive concentration lesion of PET image and the lesion of CT image"was less than 0.5 cm)in PET images,and to measure the objective data of original tumor lesion,and to compare the maximum value(SUVmax),peak value(SUVpeak)and mean value(SUVmean)of standardized uptake value(SUV),noise(SD),metabolic tumor volume(MTV)and total lesion glycolysis(TLG).The signal-noise ratio(SNR),contrast noise ratio(CNR)and target-to-background ratio(TBR)were calculated and compared.The one-way analysis of variance(ANOVA test)was adopted to conduct comparison of inter groups for the image data of 4 groups.The Bland-Altman consistency test was adopted to analyze semi-quantitative parameters of intra group.The Likert 5 scores method was adopted to conduct subject evaluation for the quality of images.The Kappa test was adopted to evaluate the consistency of images.Results:The subjective scores of image quality of 4 groups were all≥3 points,and the consistency of image quality among the four groups was very well(Kappa=0.8,P<0.05).Compared with 90s/bed group,the detection rate of small lesions of other three groups were 100%.The differences of SUVmax,SUVpeak,SUVmean,SD,TLG and MTV of semi-quantitative indicators among 4 groups were not significant(P>0.05).There were significant differences in CNR,SNR and TBR of objective indicators of 18F-FDG PET image quality among 4 groups(F=31.741,34.199,12.021,P<0.001),respectively.There were no significant differences in noise between 60s/bed group and 90s/bed group,and between 45s/bed group and 90s/bed group,respectively.The noise of image of 30s/bed group appeared increase.The Bland-Altman consistent analysis of intra group basically was within 95%confidence interval(CI).Conclusions:The short collection time of step by step mode of digital PET/CT has very good clinical application.
刘建强;李肖品;李家俊;徼红阳;刘伏鹤;刘晶哲
清华大学第一附属医院核医学科 北京 100016
特种医学
正电子发射断层显像(PET)X射线计算机断层成像(CT)图像处理,计算机辅助脱氧葡萄糖
Positron emission tomography(PET)X-ray computed tomography(CT)Image processingComputer assistanceDeoxyglucose
《中国医学装备》 2024 (008)
7-12 / 6
评论