-
放射性核素平衡法门控心血池显像能够无创性反映左右心室功能参数,对冠心病的心功能评价、异常心电传导和心肌病的辅助诊断等均具有重要价值。该显像方法受人为干扰小、测量方法稳定性好、准确性高。除了心脏疾病,其在肿瘤患者化疗及靶向治疗可能导致的心脏毒性和不良反应引起的心室功能损伤的定期监测中具有重要的临床应用价值[1-2]。此外,该显像方法在左右心室机械收缩同步性评估中的价值也一直受到临床重视[3-4]。NaI-SPECT平衡法门控心血池平面显像(简称NaI-SPECT平面显像)最为经典,但受到房室重叠、本底勾画和低探测计数的制约,其测定心室功能的准确性受到影响[5]。SPECT断层技术在一定程度上克服了这些问题,而且在左心室射血分数(left ventricular ejection fraction, LVEF)及容积测量上与心脏磁共振的一致性和相关性较好[6-7],也成为有金属植入物患者(存在心脏磁共振禁忌)准确评估左、右心室情况的首选替代检查方法[8]。但NaI-SPECT平面显像存在时间和空间分辨率低、探测器放射性计数敏感性低和采集时间长等局限性。
近年来,随着技术的进步和设备的更迭,配置固体常温半导体材料碲锌镉(cadmium zinc telluride,CZT)探测器的SPECT(即CZT-SPECT)被发明和应用,从而实现了上述NaI-SPECT性能的极大提高[9]。高门控帧数、高灵敏度将极大提高心血池断层显像的采集信息量[10-11],并且CZT-SPECT实现了表模式(list model)采集,后期可以对全部采集数据进行任意时间段的截取和重组,实现多种重建模式,如获得多种采集时间、门控帧数和成像角度的重建图像[12]。现有的对其诊断效能及采集方案优化的研究均表明这项新兴技术具有很大的潜在临床价值[13-14],但总体而言,有关CZT-SPECT平衡法门控心血池断层显像(简称CZT-SPECT断层显像)技术方面的研究相对较少,国内尚无相关研究报道。本研究旨在通过对比分析传统NaI-SPECT平面显像与心脏专用CZT-SPECT断层显像及其重建平面显像(简称CZT-SPECT断层重建平面显像)所获得的左、右心室功能参数的相关性及其差异,进行心脏专用CZT-SPECT断层显像测定双心室功能的方法学研究并探讨其优势。
-
本研究共纳入患者58例,患者的一般资料见表1。所有患者均有心脏相关疾病,包含表1中的1种或多种。
一般资料 例(%)或 或M(Q1,Q3)$ \bar x\pm s $ 年龄(岁) 60.6±12.3 男性 38(65.5) 女性 20(34.5) 身高(cm) 167.3±8.5 体重(kg) 72.1±15.0 显像剂注射剂量(MBq) 802.9±66.6 LVEF(P)(%) 31.00(22.00,41.00) RVEF(P)(%) 39.78±12.16 危险因素 高血压 33(56.9) 高血脂 1(1.7) 糖尿病 8(13.8) 吸烟 26(44.8) 饮酒 22(37.9) 临床诊断 冠心病 20(34.5) 扩张型心肌病 19(32.8) 心力衰竭 20(34.5) 完全性左束支传导阻滞 20(34.5) 其他类型心律失常 10(17.2) 其他心脏疾病(瓣膜病、风湿性
心脏病、酒精性心肌病等)7(12.1) 注:LVEF(P)为NaI-SPECT平面显像获得的左心室射血分数;RVEF(P)为NaI-SPECT平面显像获得的右心室射血分数 表 1 纳入的58例行放射性核素平衡法门控心血池显像的 患者的一般资料
Table 1. General data on the 58 included patients of gated equilibrium radionuclide cardio angiography
-
LVEF(P)、LVEF(T)和LVEF(re)分别为31.00%(22.00%,41.00%)、27.00%(19.75%,38.50%)和30.50%(20.00%,38.50%)。LVEF(re)、LVEF(P)与LVEF(T)之间的差异均有统计学意义(Z=−2.645、−3.065,均P<0.05),LVEF(re)与LVEF(P)之间的差异无统计学意义(Z=−1.057,P>0.05);三者间比较,两两之间的相关性均为优秀,相对而言,LVEF(re)与LVEF(P)之间的相关性(r=0.946,P<0.001)稍优于LVEF(T)与LVEF(P)(r=0.919,P<0.001),见图1。
-
RVEF(P)、RVEF(T)和RVEF(re)分别为39.78%±12.16%、41.57%±15.18%和40.88%±13.19%。RVEF(P)、RVEF(T)和RVEF(re)两两之间的差异均无统计学意义(t=−1.949、−1.721、0.883,均P>0.05);三者间比较,两两之间的相关性均为优秀,相对而言,RVEF(re)与RVEF(P)之间相关性(r=0.929,P<0.001)优于RVEF(T)与RVEF(P)(r=0.892,P<0.001),见图2。
-
LVEF(5 min)、LVEF(4 min)、LVEF(3 min)分别为27.50%(21.00%,39.25%)、28.50%(21.75%,40.00%)和29.50%(23.00%,42.00%),与LVEF(10 min)[27.00%(19.75%,38.50%)]相比,差异均有统计学意义(Z=−2.798、−3.288、−3.995,均P<0.05),且相关性均为优秀(r=0.970、0.938、0.903,均P<0.001)。随着重建时间的缩短,LVEF呈稍升高趋势。
RVEF(5 min)、RVEF(4 min)、RVEF(3 min)分别为42.90%±14.68%、44.55%±16.01%、45.48%±15.24%,与RVEF(10 min)(41.57%±15.18%)之间的差异均有统计学意义(t=−2.187、−3.976、−5.154,均P<0.05),且相关性均为优秀(r=0.953、0.934、0.928,均P<0.001)。随着重建时间的缩短,RVEF呈稍升高趋势。
亚组分析结果显示,除LVEF(10 min)≥20%亚组,其余亚组中LVEF(5 min)与LVEF(10 min)之间的差异均无统计学意义(均P>0.05);LVEF(10 min)≥50%亚组中LVEF(3 min)与LVEF(10 min)之间的差异无统计学意义(P>0.05);在LVEF(10 min)各亚组中,LVEF(4 min)与LVEF(10 min)之间的差异均有统计学意义(均P<0.05)。在RVEF(10 min)各亚组中,RVEF(5 min)与RVEF(10 min)之间的差异均无统计学意义(均P>0.05),而RVEF(4 min)和RVEF(3 min)与RVEF(10 min)之间的差异均有统计学意义(均P<0.05)。见表2、表3。
重建不同采集时间获得的LVEF LVEF(10 min)≥20% LVEF(10 min)≥30% LVEF(10 min)≥40% LVEF(10 min)≥50% (n=44) (n=25) (n=14) (n=7) 检验值 P值 检验值 P值 检验值 P值 检验值 P值 LVEF(3 min) −3.043a 0.002 −2.529a 0.011 −2.456a 0.014 −2.287b 0.062 LVEF(4 min) −2.837a 0.005 −2.513a 0.012 −2.203a 0.028 −2.976b 0.025 LVEF(5 min) 2.837a 0.004 −1.597a 0.110 −1.429a 0.153 −1.804b 0.121 注:CZT为碲锌镉;SPECT为单光子发射计算机体层摄影术;LVEF(3 min)、LVEF(4 min)、LVEF(5 min)分别为由原始数据重建采集时间为3、4、5 min的数据获得的左心室射血分数;LVEF(10 min)为采集时间为10 min的原始数据获得的左心室射血分数;a表示Z值;b表示t值 表 2 CZT-SPECT断层显像患者重建不同采集时间获得的LVEF与原始数据获得的LVEF的各亚组之间的比较
Table 2. Comparison between subgroups of left ventricular ejection fraction obtained at different acquisition times and left ventricular ejection fraction obtained from raw data in patients with CZT-SPECT reconstruction
重建不同采集时间获得的RVEF RVEF(10 min)≥20% RVEF(10 min)≥30% RVEF(10 min)≥40% RVEF(10 min)≥50% (n=56) (n=44) (n=31) (n=16) 检验值 P值 检验值 P值 检验值 P值 检验值 P值 RVEF(3 min) −4.229a <0.001 −3.956a <0.001 −3.659a <0.001 −3.381b 0.004 RVEF(4 min) −3.760a <0.001 −3.327a 0.001 −3.724a <0.001 −4.015b 0.001 RVEF(5 min) −1.853a 0.064 −1.339a 0.180 −1.158a 0.247 −0.132b 0.897 注:CZT为碲锌镉;SPECT为单光子发射计算机体层摄影术;RVEF(3 min)、RVEF(4 min)、RVEF(5 min)分别为由原始数据重建采集时间为3、4、5 min的数据获得的右心室射血分数;RVEF(10 min)为采集时间为10 min的原始数据获得的右心室射血分数;a表示Z值;b表示t值 表 3 CZT-SPECT断层显像患者重建不同采集时间获得的RVEF与原始数据获得的RVEF的各亚组之间的比较
Table 3. Comparison between subgroups of right ventricular ejection fraction obtained at different acquisition times and right ventricular ejection fraction obtained from raw data in patients with CZT-SPECT reconstruction
碲锌镉SPECT平衡法门控心血池断层显像重建平面及快速显像方案的研究
Study of tomographic gated equilibrium radionuclide angiography on reconstructed-planar imaging and rapid acquisition mode with cadmium zinc telluride SPECT
-
摘要:
目的 通过对比分析传统NaI-SPECT平衡法门控心血池平面显像(简称NaI-SPECT平面显像)与心脏专用碲锌镉(CZT)SPECT(CZT-SPECT)平衡法门控心血池断层显像(简称CZT-SPECT断层显像)及其重建平面显像(简称CZT-SPECT断层重建平面显像)所获得的左、右心室功能参数的相关性及其差异,进行CZT-SPECT断层显像的方法学研究并探讨其优势。 方法 回顾性分析2021年8月至2022年11月在泰达国际心血管病医院行放射性核素平衡法门控心血池显像的患者58例[其中,男性38例、女性20例,年龄(60.6±12.3)岁],所有患者均于同日先后行NaI-SPECT平面显像和CZT-SPECT断层显像,重建CZT-SPECT断层显像数据成平面显像数据,比较和分析NaI-SPECT平面显像(P)、CZT-SPECT断层显像(T)及其重建平面显像(re)获得的左心室射血分数(LVEF)和右心室射血分数(RVEF)。重建3、4和5 min的CZT-SPECT断层显像采集数据,将获得的LVEF(3 min)、LVEF(4 min)、LVEF(5 min)和RVEF(3 min)、RVEF(4 min)、RVEF(5 min)与原始采集数据LVEF(10 min)、RVEF(10 min)进行比较。将LVEF(10 min)和RVEF(10 min)按≥20%、≥30%、≥40%和≥50%重新分为各亚组,并与上述重建数据分别进行分析。计量资料的比较采用配对t检验(或Wilcoxon符号秩检验),相关性采用Pearson(或Spearman)相关性分析。 结果 LVEF(re)[30.50%(20.00%,38.50%)]、LVEF(P)[31.00%(22.00%,41.00%)]与LVEF(T)[27.00%(19.75%,38.50%)]之间的差异均有统计学意义(Z=−2.645、−3.065,均P<0.05),LVEF(re)与LVEF(P)之间的差异无统计学意义(Z=−1.057,P>0.05);RVEF(P)(39.78%±12.16%)、RVEF(T)(41.57%±15.18%)和RVEF(re)(40.88%±13.19%)之间两两比较,差异均无统计学意义(t=−1.949、−1.721、0.883,均P>0.05);3种显像方法获得的LVEF、RVEF比较,两两之间的相关性均为优秀(r=0.892~0.946,均P<0.001);LVEF(3 min)、LVEF(4 min)、LVEF(5 min)和RVEF(3 min)、RVEF(4 min)、RVEF(5 min)分别与LVEF(10 min)和RVEF(10 min)比较,差异均有统计学意义(Z=−2.798、−3.288、−3.995,t=−2.187、−3.976、−5.154,均P<0.05),且相关性均为优秀(r=0.903~0.970,均P<0.001)。亚组分析结果显示,除LVEF≥20%亚组外,其余各亚组中,LVEF(5 min)与LVEF(10 min)、RVEF(5 min)与RVEF(10 min)之间的差异均无统计学意义(Z=−1.853~−1.158,t=−1.804、−0.132,均P>0.05)。 结论 CZT-SPECT断层显像可通过重建获得平面显像数据,结合采集性能优势,可在获得可靠的数据用于后期处理和获得准确的测量结果的同时,进一步降低检查中的辐射剂量或缩短时间。 -
关键词:
- 门控血池显像 /
- 碲锌镉 /
- 体层摄影术,发射型计算机,单光子 /
- 心室射血分数
Abstract:Objective To analyze the correlation and difference of biventricular function from planar gated equilibrium radionuclide angiography with conventional sodium iodide SPECT (NaI-SPECT planar imaging) and tomographic gated equilibrium radionuclide angiography with cadmium zinc telluride SPECT (CZT-SPECT tomographic imaging), and reconstructed-planar imaging from CZT-SPECT tomographic imaging (CZT-SPECT re-planar imaging) to investigate the methodology and advantages of CZT-SPECT tomographic imaging. Methods A retrospective analysis was performed on 58 patients (38 males and 20 females aged (60.6±12.3) years), who underwent gated equilibrium radionuclide angiography in TEDA International Cardiovascular Hospital from August 2021 to November 2022. All patients were subjected to NaI-SPECT planar imaging and CZT-SPECT tomographic imaging on the same day. CZT-SPECT tomographic imaging data were reprocessed in re-planar form and at different acquisition times. The left ventricular ejection fraction (LVEF) and right ventricular ejection fraction (RVEF) of NaI-SPECT planar imaging (P), CZT-SPECT tomographic imaging (T), and CZT-SPECT re-planar imaging (re) were compared and analyzed, and the CZT-SPECT tomographic imaging acquisition data at 3, 4, and 5 min were reconstructed. The LVEF (3 min), LVEF (4 min), LVEF (5 min), and RVEF (3 min), RVEF (4 min), RVEF (5 min) were compared with the original acquisition data LVEF (10 min) and RVEF (10 min), respectively. Divide LVEF (10 min) and RVEF (10 min) into subgroups of ≥20%, ≥30%, ≥40%, and ≥50%, and analyze them separately with the reconstructed data mentioned above. Paired t-test (or Wilcoxon signed rank test) and Pearson (or Spearman) were used in analyzing differences and correlations among the data. Results The differences among LVEF (re) (30.50% (20.00%, 38.50%)), LVEF (P) (31.00% (22.00%, 41.00%)), and LVEF (T) (27.00% (19.75%, 38.50%)) were statistically significant (Z=−2.645, −3.065; both P<0.05), whereas the differences between LVEF (re) and LVEF (P) were not statistically significant (Z=−1.057; P>0.05). No significant difference was found among RVEF (P) (39.78%±12.16%), RVEF (T) (41.57%±15.18%) and RVEF (re) (40.88%±13.19%; t=−1.949, −1.721, 0.883; all P>0.05). The LVEF and RVEF correlations obtained by the three imaging methods were excellent (r=0.892–0.946; all P<0.001). The differences between LVEF (3 min), LVEF (4 min), LVEF (5 min), and LVEF (10 min), and between RVEF (3 min), RVEF (4 min), RVEF (5 min), and RVEF (10 min) were statistically significant (Z=−2.798, −3.288, and −3.995; t=−2.187, −3.976, and −5.154; all P<0.05), and the correlations were excellent (r=0.903–0.970; all P<0.001). Subgroup analysis showed no significant difference between LVEF (5 min) and LVEF (10 min) and between RVEF (5 min) and RVEF (10 min) in all subgroups, except the LVEF≥20% subgroup (Z=−1.853 to −1.158; t=−1.804, −0.132; all P>0.05). Conclusions CZT-SPECT tomographic imaging can obtain planar datasets through reconstruction. Owing to its acquisition performance, CZT-SPECT tomographic imaging can further reduce the radiation dose or time for examination while obtaining reliable data for postprocessing and accurate measurement results. -
表 1 纳入的58例行放射性核素平衡法门控心血池显像的 患者的一般资料
Table 1. General data on the 58 included patients of gated equilibrium radionuclide cardio angiography
一般资料 例(%)或 或M(Q1,Q3)$ \bar x\pm s $ 年龄(岁) 60.6±12.3 男性 38(65.5) 女性 20(34.5) 身高(cm) 167.3±8.5 体重(kg) 72.1±15.0 显像剂注射剂量(MBq) 802.9±66.6 LVEF(P)(%) 31.00(22.00,41.00) RVEF(P)(%) 39.78±12.16 危险因素 高血压 33(56.9) 高血脂 1(1.7) 糖尿病 8(13.8) 吸烟 26(44.8) 饮酒 22(37.9) 临床诊断 冠心病 20(34.5) 扩张型心肌病 19(32.8) 心力衰竭 20(34.5) 完全性左束支传导阻滞 20(34.5) 其他类型心律失常 10(17.2) 其他心脏疾病(瓣膜病、风湿性
心脏病、酒精性心肌病等)7(12.1) 注:LVEF(P)为NaI-SPECT平面显像获得的左心室射血分数;RVEF(P)为NaI-SPECT平面显像获得的右心室射血分数 表 2 CZT-SPECT断层显像患者重建不同采集时间获得的LVEF与原始数据获得的LVEF的各亚组之间的比较
Table 2. Comparison between subgroups of left ventricular ejection fraction obtained at different acquisition times and left ventricular ejection fraction obtained from raw data in patients with CZT-SPECT reconstruction
重建不同采集时间获得的LVEF LVEF(10 min)≥20% LVEF(10 min)≥30% LVEF(10 min)≥40% LVEF(10 min)≥50% (n=44) (n=25) (n=14) (n=7) 检验值 P值 检验值 P值 检验值 P值 检验值 P值 LVEF(3 min) −3.043a 0.002 −2.529a 0.011 −2.456a 0.014 −2.287b 0.062 LVEF(4 min) −2.837a 0.005 −2.513a 0.012 −2.203a 0.028 −2.976b 0.025 LVEF(5 min) 2.837a 0.004 −1.597a 0.110 −1.429a 0.153 −1.804b 0.121 注:CZT为碲锌镉;SPECT为单光子发射计算机体层摄影术;LVEF(3 min)、LVEF(4 min)、LVEF(5 min)分别为由原始数据重建采集时间为3、4、5 min的数据获得的左心室射血分数;LVEF(10 min)为采集时间为10 min的原始数据获得的左心室射血分数;a表示Z值;b表示t值 表 3 CZT-SPECT断层显像患者重建不同采集时间获得的RVEF与原始数据获得的RVEF的各亚组之间的比较
Table 3. Comparison between subgroups of right ventricular ejection fraction obtained at different acquisition times and right ventricular ejection fraction obtained from raw data in patients with CZT-SPECT reconstruction
重建不同采集时间获得的RVEF RVEF(10 min)≥20% RVEF(10 min)≥30% RVEF(10 min)≥40% RVEF(10 min)≥50% (n=56) (n=44) (n=31) (n=16) 检验值 P值 检验值 P值 检验值 P值 检验值 P值 RVEF(3 min) −4.229a <0.001 −3.956a <0.001 −3.659a <0.001 −3.381b 0.004 RVEF(4 min) −3.760a <0.001 −3.327a 0.001 −3.724a <0.001 −4.015b 0.001 RVEF(5 min) −1.853a 0.064 −1.339a 0.180 −1.158a 0.247 −0.132b 0.897 注:CZT为碲锌镉;SPECT为单光子发射计算机体层摄影术;RVEF(3 min)、RVEF(4 min)、RVEF(5 min)分别为由原始数据重建采集时间为3、4、5 min的数据获得的右心室射血分数;RVEF(10 min)为采集时间为10 min的原始数据获得的右心室射血分数;a表示Z值;b表示t值 -
[1] Totzeck M, Aide N, Bauersachs J, et al. Nuclear medicine in the assessment and prevention of cancer therapy-related cardiotoxicity: prospects and proposal of use by the European Association of Nuclear Medicine (EANM)[J]. Eur J Nucl Med Mol Imaging, 2023, 50(3): 792−812. DOI: 10.1007/s00259-022-05991-7. [2] Polomski EAS, Antoni ML, Jukema JW, et al. Nuclear medicine imaging methods of radiation-induced cardiotoxicity[J]. Semin Nucl Med, 2022, 52(5): 597−610. DOI: 10.1053/j.semnuclmed.2022.02.001. [3] Naya M, Manabe O, Koyanagawa K, et al. The role of nuclear medicine in assessments of cardiac dyssynchrony[J]. J Nucl Cardiol, 2018, 25(6): 1980−1987. DOI: 10.1007/s12350-017-1072-z. [4] Steelant B, Stankovic I, Roijakkers I, et al. The impact of infarct location and extent on LV motion patterns: implications for dyssynchrony assessment[J]. JACC Cardiovasc Imaging, 2016, 9(6): 655−664. DOI: 10.1016/j.jcmg.2015.07.021. [5] Tonge CM, Fernandez RC, Harbinson MT. Commentary: current issues in nuclear cardiology[J]. Br J Radiol, 2008, 81(964): 270−274. DOI: 10.1259/bjr/59260451. [6] Apert A, Canu M, Jankowski A, et al. Comparison of cadmium zinc telluride ECG-gated SPECT equilibrium radionuclide angiocardiography to magnetic resonance imaging to measure right ventricular volumes and ejection fraction in patients with cardiomyopathy[J]. J Nucl Cardiol, 2022, 29(4): 1647−1656. DOI: 10.1007/s12350-021-02653-3. [7] Xie BQ, Tian YQ, Zhang J, et al. Evaluation of left and right ventricular ejection fraction and volumes from gated blood-pool SPECT in patients with dilated cardiomyopathy: comparison with cardiac MRI[J]. J Nucl Med, 2012, 53(4): 584−591. DOI: 10.2967/jnumed.111.096057. [8] Lairez O, Delmas C, Fournier P, et al. Feasibility and accuracy of gated blood pool SPECT equilibrium radionuclide ventriculography for the assessment of left and right ventricular volumes and function in patients with left ventricular assist devices[J]. J Nucl Cardiol, 2018, 25(2): 625−634. DOI: 10.1007/s12350-016-0670-5. [9] Einstein AJ, Blankstein R, Andrews H, et al. Comparison of image quality, myocardial perfusion, and left ventricular function between standard imaging and single-injection ultra-low-dose imaging using a high-efficiency SPECT camera: the MILLISIEVERT study[J]. J Nucl Med, 2014, 55(9): 1430−1437. DOI: 10.2967/jnumed.114.138222. [10] Bailly M, Le Rouzic G, Metrard G. Gated tomographic radionuclide angiography using CZT gamma camera in patients receiving cardiotoxic chemotherapy: going faster and less irradiating[J]. Clin Nucl Med, 2019, 44(6): 472−474. DOI: 10.1097/RLU.0000000000002569. [11] Tissot H, Roch V, Morel O, et al. Left ventricular ejection fraction determined with the simulation of a very low-dose CZT-SPECT protocol and an additional count-calibration on planar radionuclide angiographic data[J]. J Nucl Cardiol, 2019, 26(5): 1539−1549. DOI: 10.1007/s12350-019-01619-w. [12] Liu YH, Fazzone-Chettiar R, Sandoval V, et al. New approach for quantification of left ventricular function from low-dose gated bloodpool SPECT: validation and comparison with conventional methods in patients[J]. J Nucl Cardiol, 2021, 28(3): 939−950. DOI: 10.1007/s12350-019-01823-8. [13] Hansen NL, Haarmark C, Zerahn B. Ventricular peak emptying and filling rates measured by gated tomographic radionuclide angiography using a cadmium-zinc-telluride SPECT camera in chemotherapy-naïve cancer patients[J]. J Nucl Cardiol, 2020, 27(4): 1193−1201. DOI: 10.1007/s12350-019-01756-2. [14] Rydberg J, Andersen J, Haarmark C, et al. The influence of anthropometric and basic circulatory variables on count rate in cadmium-zinc-telluride SPECT gated radionuclide angiography[J]. J Nucl Cardiol, 2019, 26(6): 1974−1980. DOI: 10.1007/s12350-018-1402-9. [15] Wells RG, Marvin B, Kovalski G, et al. Planar radionuclide angiography with a dedicated cardiac SPECT camera[J]. J Nucl Cardiol, 2013, 20(3): 358−366. DOI: 10.1007/s12350-013-9674-6. [16] Duvall WL, Guma-Demers KA, George T, et al. Radiation reduction and faster acquisition times with SPECT gated blood pool scans using a high-efficiency cardiac SPECT camera[J]. J Nucl Cardiol, 2016, 23(5): 1128−1138. DOI: 10.1007/s12350-015-0214-4. [17] Chen YC, Ko CL, Yen RF, et al. Comparison of biventricular ejection fractions using cadmium-zinc-telluride SPECT and planar equilibrium radionuclide angiography[J]. J Nucl Cardiol, 2016, 23(3): 348−361. DOI: 10.1007/s12350-015-0367-1.