Advertisement

Reducing motion-correction-induced variability in 82rubidium myocardial blood-flow quantification

  • Alexis Poitrasson-RivièreEmail author
  • Jonathan B. Moody
  • Tomoe Hagio
  • Richard L. Weinberg
  • James R. Corbett
  • Venkatesh L. Murthy
  • Edward P. Ficaro
Original Article

Abstract

Background

Clinical use of myocardial blood flow (MBF) and flow reserve (MFR) is increasing. Motion correction is necessary to obtain accurate results but can introduce variability when performed manually. We sought to reduce that variability with an automated motion-correction algorithm.

Methods

A blinded randomized controlled trial of two technologists was performed on the motion correction of 100 dynamic 82Rb patient studies comparing manual motion correction with manual review and adjustment of automated motion correction. Inter-rater variability between technologists for MBF and MFR was the primary outcome with comparison made by analysis of the limits of agreement. Processing time was the secondary outcome.

Results

Limits of agreements between the two technologists decreased significantly for both MBF and MFR, going from [− 0.22, 0.22] mL/min/g and [− 0.31, 0.36] to [− 0.12, 0.15] mL/min/g and [− 0.15, 0.18], respectively (both P < .002). In addition, the average time spent on motion correcting decreased by 1 min per study from 5:21 to 4:21 min (P = .001).

Conclusions

In this randomized controlled trial, the use of automated motion correction significantly decreased inter-user variability and reduced processing time.

Keywords

PET Myocardial blood flow Image analysis 

Abbreviations

PET

Positron emission tomography

CT

Computed tomography

Rb

Rubidium

MBF

Myocardial blood flow

mfr

Myocardial flow reserve

LOA

Limits of agreement

CV

Coefficient of variation

Spanish Abstract

Antecedentes

La utilización clínica del flujo sanguíneo miocárdico (MBF por sus siglas en inglés) y de la reserva de flujo coronario (MFR por sus siglas en inglés) está en aumento. La corrección de movimiento es necesaria para obtener resultados exactos, pero puede introducir variabilidad cuando se realiza manualmente. Nosotros buscamos reducir esa variabilidad con un algoritmo automático de corrección de movimiento.

Métodos

Se realizó un ensayo controlado aleatorizado ciego de dos tecnólogos sobre la corrección de movimiento de 100 estudios dinámicos de pacientes de rubidio-82, comparando la corrección manual con la revisión y el ajuste de la corrección automática. La variabilidad interobservador entre los tecnólogos para MBF y MFR fue el resultado principal, con la comparación realizada por el análisis de los límites de concordancia. El tiempo de procesamiento fue el resultado secundario.

Resultados

Límites de concordancia entre los dos tecnólogos disminuyeron significativamente para MBF y MFR, de [− 0,22; 0,22] y [− 0,31; 0,36] a [− 0.12; 0,15] y [− 0,15; 0,18], respectivamente (P < ,002). Adicionalmente, el tiempo promedio de procesamiento disminuyo en 1 min por estudio, de 5:21 a 4:21 min (P = ,001).

Conclusiones

En este ensayo controlado aleatorizado, la utilización de corrección de movimiento automática disminuyo significativamente la variabilidad entre usarios y redujo el tiempo de procesamiento.

Chinese Abstract

背景

心肌血流量(MBF)和血流储备(MFR)的临床应用越来越多。使用运动校正可以得到更精准的结果,但手动校准可能会增加变异性。本文提出了一种通过自动校准算法来减少手动校正的变异性。

方法

两名技术人员通过单盲随机对照试验对100名行动态82Rb检查的患者进行了运动校正,比较通过人工检查进行的手动运动校正和自动运动校正的差异。 以MBF和MFR在技术人员之间的一致性界限为主要分析结果,以处理时间为次要结果。

结果

自动校准算法显著缩小MBF和MFR在两名技术人员之间的一致性界限,分别从[− 0.22, 0.22] mL/min/g和[− 0.31, 0.36]降到[− 0.12, 0.15] mL/min/g和[− 0.15, 0.18](均P < .002)。 此外,每次研究所花费的平均运动校正时间从5:21减少到4:21分钟(P = .001)。

结论

在这项随机对照试验中,使用自动运动校正显著降低了操作人员之间的可变性并缩短了处理时间。

French Abstract

Contexte

L’utilisation clinique du débit sanguin myocardique (DSM) et de la réserve de débit myocardique (RDM) est de plus en plus fréquente. La correction de mouvement durant la séquence d’images dynamiques est nécessaire pour obtenir des résultats justes, mais peut introduire de la variabilité lorsqu’elle est réalisée manuellement. Nous avons cherché à réduire cette variabilité en utilisant un algorithme de correction automatique de mouvement.

Méthodes

Un essai contrôlé aléatoire masqué a été réalisé pour vérifier l’efficacité de l’algorithme de correction automatique. Utilisant une population de 100 patients référés pour un protocole repos-effort dynamique au 82Rb, l’essai comparait la correction manuelle de mouvement à une rectification manuelle de la correction automatique pour deux opérateurs. La variabilité inter-opérateur du DSM et de la RDM, mesurée en termes d’intervalle de confiance (de niveau 0,95), constitue le résultat majeur de l’étude. Le temps de correction a aussi été analysé comme résultat secondaire.

Résultats

En utilisant l’algorithme de correction automatique, Les intervalles de confiance pour le DSM et la RDM sont significativement améliorés, allant de [− 0,22; 0,22] mL/min/g et [− 0,31; 0,36] à [− 0,12; 0,15] mL/min/g et [− 0,15; 0,18], respectivement (P < ,002). Par ailleurs, le temps moyen de correction de mouvement a diminué d’une minute par patient, passant de 5 min 21 s à 4 min 21 s (P = ,001).

Conclusions

Lors de cette étude, l’utilisation de la correction automatique de mouvement a significativement diminué la variabilité inter-opérateur, ainsi que le temps de correction.

Notes

Acknowledgements

The authors would like to thank Chanée Nelson and Kevin Fischio for their efforts in processing the datasets in this study.

Disclosure

A. Poitrasson-Rivière, J.B. Moody, and T. Hagio are employees of the INVIA Medical Imaging Solutions. R.L. Weinberg has no conflicts of interest to disclose. J.R. Corbett and E.P. Ficaro are owners of the INVIA Medical Imaging Solutions. V.L. Murthy receives research support and funding from the INVIA Medical Imaging Solutions; research grants and lecture honoraria from the Siemens Medical Imaging; an expert witness testimony payment on behalf of the Jubilant Draximage; and advisory board payments from the Curium and Ionetix. V.L. Murthy has stock in General Electric and Cardinal Health, and stock options in Ionetix.

Supplementary material

12350_2019_1911_MOESM1_ESM.docx (16 kb)
Supplementary material 1 (DOCX 16 kb)
12350_2019_1911_MOESM2_ESM.pptx (282 kb)
Supplementary material 2 (PPTX 281 kb)
12350_2019_1911_MOESM3_ESM.wav (44.1 mb)
Supplementary material 3 (WAV 45166 kb)

References

  1. 1.
    Naya M, Murthy VL, Taqueti VR, Foster CR, Klein J, Garber M, et al. Preserved coronary flow reserve effectively excludes high-risk coronary artery disease on angiography. J Nucl Med 2014;55:248-55.  https://doi.org/10.2967/jnumed.113.121442.CrossRefPubMedPubMedCentralGoogle Scholar
  2. 2.
    Murthy VL, Naya M, Foster CR, Hainer J, Gaber M, Di Carli G, et al. Improved cardiac risk assessment with noninvasive measures of coronary flow reserve. Circulation 2011;124:2215-24.  https://doi.org/10.1161/CIRCULATIONAHA.111.050427.CrossRefPubMedPubMedCentralGoogle Scholar
  3. 3.
    Murthy VL, Bateman TM, Beanlands RS, Berman DS, Borges-Neto S, Chareonthaitawee P, et al. Clinical quantification of myocardial blood flow using pet: Joint position paper of the SNMMI cardiovascular council and the ASNC. J Nucl Cardiol 2017;59:273-93.  https://doi.org/10.1007/s12350-017-1110-x.CrossRefGoogle Scholar
  4. 4.
    Ziadi MC, Dekemp RA, Williams KA, Guo A, Chow BJ, Renaud JM. Impaired myocardial flow reserve on rubidium-82 positron emission tomography imaging predicts adverse outcomes in patients assessed for myocardial ischemia. J Am Coll Cardiol 2011;58:740-8.  https://doi.org/10.1016/j.jacc.2011.01.065.CrossRefGoogle Scholar
  5. 5.
    Farhad H, Dunet V, Bachelard K, Allenbach G, Kaufmann PA, Prior JO. Added prognostic value of myocardial blood flow quantitation in rubidium-82 positron emission tomography imaging. Eur Hear J 2013;14:1203-10.  https://doi.org/10.1093/ehjci/jet068.CrossRefGoogle Scholar
  6. 6.
    Klein R, Ocneanu A, Renaud JM, Ziadi MC, Beanlands RSB, deKemp RA. Consistent tracer administration profile improves test-retest repeatability of myocardial blood flow quantification with 82Rb dynamic PET imaging. J Nucl Cardiol 2016;25:929-41.  https://doi.org/10.1007/s12350-016-0698-6.CrossRefPubMedCentralGoogle Scholar
  7. 7.
    Efseaff M, Klein R, Ziadi MC, Beanlands RS, deKemp RA. Short-term repeatability of resting myocardial blood flow measurements using rubidium-82 PET imaging. J Nucl Cardiol 2012;19:997-1006.  https://doi.org/10.1007/s12350-012-9600-3.CrossRefPubMedGoogle Scholar
  8. 8.
    Manabe O, Yoshinaga K, Katoh C, Naya M, deKemp RA, Tamaki N. Repeatability of rest and hyperemic myocardial blood flow measurements with 82Rb dynamic PET. J Nucl Med 2009;50:68-71.  https://doi.org/10.2967/jnumed.108.055673.CrossRefPubMedGoogle Scholar
  9. 9.
    Klein R, Renaud JM, Ziadi MC, Thorn SL, Adler A, Beanlands RS, et al. Intra- and inter-operator repeatability of myocardial blood flow and myocardial flow reserve measurements using rubidium-82 pet and a highly automated analysis program. J Nucl Cardiol 2010;17:600-16.  https://doi.org/10.1007/s12350-010-9225-3.CrossRefPubMedGoogle Scholar
  10. 10.
    Lee BC, Moody JB, Poitrasson-Rivière A, Melvin AC, Weinberg RL, Corbett JR, et al. Blood pool and tissue phase patient motion effects on82rubidium PET myocardial blood flow quantification. J Nucl Cardiol 2018.  https://doi.org/10.1007/s12350-018-1256-1.CrossRefPubMedPubMedCentralGoogle Scholar
  11. 11.
    Lee BC, Moody JB, Poitrasson-Rivière A, Melvin AC, Weinberg RL, Corbett JR, et al. Automated dynamic motion correction using normalized gradient fields for 82rubidium PET myocardial blood flow quantification. J Nucl Cardiol 2018.  https://doi.org/10.1007/s12350-018-01471-4.CrossRefPubMedPubMedCentralGoogle Scholar
  12. 12.
    Ficaro EP, Lee BC, Kritzman JN, Corbett JR. Corridor4DM: The Michigan method for quantitative nuclear cardiology. J Nucl Cardiol 2007;14:455-65.  https://doi.org/10.1016/j.nuclcard.2007.06.006.CrossRefPubMedGoogle Scholar
  13. 13.
    Hamlett A, Ryan L, Serrano-Trespalacios P, Wolfinger R. Mixed models for assessing correlation in the presence of replication. J Air Waste Manag Assoc 2003;53:442-50.  https://doi.org/10.1080/10473289.2003.10466174.CrossRefPubMedGoogle Scholar
  14. 14.
    Silver NC, Hittner JB, May K. Testing dependent correlations with nonoverlapping variables: A Monte Carlo simulation. J Exp Educ 2004;73:53-69.  https://doi.org/10.3200/JEXE.71.1.53-70.CrossRefGoogle Scholar
  15. 15.
    Anagnostopoulos C, Almonacid A, El Fakhri G, Curillova Z, Sitek A, Roughton M, et al. Quantitative relationship between coronary vasodilator reserve assessed by 82Rb PET imaging and coronary artery stenosis severity. Eur J Nucl Med Mol Imaging 2008;35:1593-601.  https://doi.org/10.1007/s00259-008-0793-2.CrossRefPubMedPubMedCentralGoogle Scholar

Copyright information

© American Society of Nuclear Cardiology 2019

Authors and Affiliations

  • Alexis Poitrasson-Rivière
    • 1
    Email author
  • Jonathan B. Moody
    • 1
  • Tomoe Hagio
    • 1
  • Richard L. Weinberg
    • 2
  • James R. Corbett
    • 3
  • Venkatesh L. Murthy
    • 2
  • Edward P. Ficaro
    • 1
    • 2
  1. 1.INVIA Medical Imaging SolutionsAnn ArborUSA
  2. 2.Division of Cardiovascular Medicine, Department of Internal Medicine and Frankel Cardiovascular CenterUniversity of MichiganAnn ArborUSA
  3. 3.Division of Nuclear Medicine, Department of Radiology and Frankel Cardiovascular CenterUniversity of MichiganAnn ArborUSA

Personalised recommendations