CAJ | 학술논문

目的评价部分容积效应对单光子与符合线路显像中病灶/本底比值(L/B值)的影响.方法使用简易Jaszczak模型,内有6个高为7.0.cm,底面直径分别为2.8,1.8,1.3,0.8,0.8和0.8 cm的圆柱体作为病灶,分别用99TcmO4和18F-脱氧葡萄糖(FDG)进行2组对比度的实验.测量3种L/B值:L/B1为最大计数比值,L/B2为以病灶直径为感兴趣区(ROI)直径的平均计数比值,L/B3为以6个像素为ROI的平均计数比值.本底均为平均计数.比较3种L/B值与对比度的关系.采用SPSS 11.0软件进行配对t检验和直线相关、回归分析.结果单光子显像对比度为2:1和符合线路显像对比度为2:1及2.5:1时,直径0.8 cm病灶显示不清.除符合线路显像直径2.8 cm病灶的L/B值与对比度的差异无统计学意义外,其余病灶的L/B值均低于对比度;病灶直径越小,L/B值被低估的程度越明显,二者呈显著正相关,其回归方程分别为:单光子显像时Y2.8=0.086+0.689 X对比度、Y1.8=0.350+0.436X对比度、Y1.3=0.837+0.307 X对比度及Y0.8=0.564+0.199 X对比度符合线路显像时Y208=0.221+0.941 X对比度、Y1.8=0.426+0.647 X对比度H、Y<1.3>3=0.484+0.528X对比度H及Y0.8=0.429+0.318 X对比度.L/B1值最接近对比度.结论部分容积效应对单光子和符合线路的L/B值测量都有重要影响,对前者的影响更显著,可以用回归方程校正,用病灶最大计数计算L/B值更实用.
목적 평개부분용적효응대단광자여부합선로현상중병조/본저비치(L/B치)적영향.방법 사용간역Jaszczak모형,내유6개고위7.0.cm,저면직경분별위2.8,1.8,1.3,0.8,0.8화0.8 cm적원주체작위병조,분별용99TcmO4화18F-탈양포도당(FDG)진행2조대비도적실험.측량3충L/B치:L/B1위최대계수비치,L/B2위이병조직경위감흥취구(ROI)직경적평균계수비치,L/B3위이6개상소위ROI적평균계수비치.본저균위평균계수.비교3충L/B치여대비도적관계.채용SPSS 11.0연건진행배대t검험화직선상관、회귀분석.결과 단광자현상대비도위2:1화부합선로현상대비도위2:1급2.5:1시,직경0.8 cm병조현시불청.제부합선로현상직경2.8 cm병조적L/B치여대비도적차이무통계학의의외,기여병조적L/B치균저우대비도;병조직경월소,L/B치피저고적정도월명현,이자정현저정상관,기회귀방정분별위:단광자현상시Y2.8=0.086+0.689 X대비도、Y1.8=0.350+0.436X대비도、Y1.3=0.837+0.307 X대비도급Y0.8=0.564+0.199 X대비도부합선로현상시Y208=0.221+0.941 X대비도、Y1.8=0.426+0.647 X대비도H、Y<1.3>3=0.484+0.528X대비도H급Y0.8=0.429+0.318 X대비도.L/B1치최접근대비도.결론 부분용적효응대단광자화부합선로적L/B치측량도유중요영향,대전자적영향경현저,가이용회귀방정교정,용병조최대계수계산L/B치경실용.
Objective Partial volume effect is known to cause incorrect estimation of radiopharma-ceutical uptake. The aim of this study was to evaluate the influence of partial volume effect on the measure-ment of lesion to background ratio (L/B) in standard SPECT and coincidence circuit SPECT. Methods Using a simplified Jaszczak phantom containing 6 cylinders to simulate lesions with diameters of 2.8, 1.8,1.3, 0.8, 0.8 and 0.8 cm, standard SPECT with 999TcmO4- and coincidence SPECT with 18F-fluorodeox-yglucese (FDG) studies were performed. The average background counts and three types of L/B values were measured, namely L/B1 for maximal counts ratio; L/B2 for the average counts ratio with the size of re-gion of interest (ROI) equal to the lesion; and L/B3 for the average counts ratio of 6 pixels size in ROI.The L/B ratios were compared with the contrasts. SPSS 11.0 was used to calculate Paired Student's t test, linear correlation and regression. Results When the contrast was set to 2:1 in standard SPECT and 2:1 or 2.5:1 in coincidence SPECT, the lesions with 0.8 cm in diameter could not be detected. Except for the le-sion with 2.8 cm in diameter in coincidence SPECT, the L/B values were all significantly underestimated and there was strong positive correlation between the L/B values and the contrasts. The smaller the lesion size, the greater was the underestimation of the L/B values. The regression equations were Y2.8=0.086+0.689 Xcontrast, Y1.8=0.350+0.436 Xcontrast, Y103=0.837+0.307 Xcontrast Y0.8=0.564 + 0.199 Xcontrast for standard SPECT series, and Y2.8=0.221+0.941 X Y1.8=0.426 +0.647 XcontrastY1.3=0.484 +0.528 Xcontrast Y0.8=0.429+0.318 Xcontrast for coincidence SPECT series. The L/B values using the maxi-mum lesion radioactivity counts had the best results. Conclusions Significant partial volume effects on L/B ratios were detected in SPECT and coincidence SPECT, particularly the former. Calibration by regression equa-tion using the maximum L/B value may yield the best correction parameters.