Standard SPECT myocardial perfusion estimation from half-time acquisitions using deep convolutional residual neural networks

Abstract

Introduction

The purpose of this work was to assess the feasibility of acquisition time reduction in MPI-SPECT imaging using deep leering techniques through two main approaches, namely reduction of the acquisition time per projection and reduction of the number of angular projections.

Methods

SPECT imaging was performed using a fixed 90° angle dedicated dual-head cardiac SPECT camera. This study included a prospective cohort of 363 patients with various clinical indications (normal, ischemia, and infarct) referred for MPI-SPECT. For each patient, 32 projections for 20 seconds per projection were acquired using a step and shoot protocol from the right anterior oblique to the left posterior oblique view. SPECT projection data were reconstructed using the OSEM algorithm (6 iterations, 4 subsets, Butterworth post-reconstruction filter). For each patient, four different datasets were generated, namely full time (20 seconds) projections (FT), half-time (10 seconds) acquisition per projection (HT), 32 full projections (FP), and 16 half projections (HP). The image-to-image transformation via the residual network was implemented to predict FT from HT and predict FP from HP images in the projection domain. Qualitative and quantitative evaluations of the proposed framework was performed using a tenfold cross validation scheme using the root mean square error (RMSE), absolute relative error (ARE), structural similarity index, peak signal-to-noise ratio (PSNR) metrics, and clinical quantitative parameters.

Results

The results demonstrated that the predicted FT had better image quality than the predicted FP images. Among the generated images, predicted FT images resulted in the lowest error metrics (RMSE = 6.8 ± 2.7, ARE = 3.1 ± 1.1%) and highest similarity index and signal-to-noise ratio (SSIM = 0.97 ± 1.1, PSNR = 36.0 ± 1.4). The highest error metrics (RMSE = 32.8 ± 12.8, ARE = 16.2 ± 4.9%) and the lowest similarity and signal-to-noise ratio (SSIM = 0.93 ± 2.6, PSNR = 31.7 ± 2.9) were observed for HT images. The RMSE decreased significantly (P value < .05) for predicted FT (8.0 ± 3.6) relative to predicted FP (6.8 ± 2.7).

Conclusion

Reducing the acquisition time per projection significantly increased the error metrics. The deep neural network effectively recovers image quality and reduces bias in quantification metrics. Further research should be undertaken to explore the impact of time reduction in gated MPI-SPECT.

This is a preview of subscription content, log in to check access.

Figure 1
Figure 2
Figure 3
Figure 4
Figure 5
Figure 6
Figure 7

Abbreviations

SPECT:

Single-photon emission computed tomography

MPI:

Myocardial perfusion imaging

CAD:

Coronary artery disease

CNN:

Convolutional neural network

ResNet:

Residual network

FT-FP:

Full time and full projection, reference images

HP:

Half projection

HT:

Half time

FP-Prediction:

Full projection prediction

FT-Prediction:

Full time prediction

References

  1. 1.

    Iskandrian AE, Garcia EV. Nuclear cardiac imaging: Principles and applications. Oxford: Oxford University Press; 2008.

    Google Scholar 

  2. 2.

    Rahmim A, Zaidi H. PET versus SPECT: Strengths, limitations and challenges. Nucl Med Commun 2008;29:193-207.

    Article  Google Scholar 

  3. 3.

    Wu J, Liu C. Recent advances in cardiac SPECT instrumentation and imaging methods. Phys Med Biol 2019;64:06TR1.

    Google Scholar 

  4. 4.

    Ljungberg M, Pretorius PH. SPECT/CT: An update on technological developments and clinical applications. Br J Radiol 2018;91:20160402.

    Article  Google Scholar 

  5. 5.

    Rahmim A, Qi J, Sossi V. Resolution modeling in PET imaging: Theory, practice, benefits, and pitfalls. Med Phys 2013;40:064301-15.

    Article  Google Scholar 

  6. 6.

    Caobelli F, Kaiser SR, Thackeray JT, Bengel FM, Chieregato M, Soffientini A, et al. IQ SPECT allows a significant reduction in administered dose and acquisition time for myocardial perfusion imaging: Evidence from a phantom study. J Nucl Med 2014;55:2064-70.

    Article  Google Scholar 

  7. 7.

    Valenta I, Treyer V, Husmann L, Gaemperli O, Schindler MJ, Herzog BA, et al. New reconstruction algorithm allows shortened acquisition time for myocardial perfusion SPECT. Eur J Nucl Med Mol Imaging 2010;37:750-7.

    Article  Google Scholar 

  8. 8.

    DePuey EG, Bommireddipalli S, Clark J, Thompson L, Srour Y. Wide beam reconstruction “quarter-time” gated myocardial perfusion SPECT functional imaging: A comparison to “full-time” ordered subset expectation maximum. J Nucl Cardiol 2009;16:736-52.

    Article  Google Scholar 

  9. 9.

    Faes L, Wagner SK, Fu DJ, Liu X, Korot E, Ledsam JR, et al. Automated deep learning design for medical image classification by health-care professionals with no coding experience: A feasibility study. Lancet Digit Health 2019;1:e232-42.

    Article  Google Scholar 

  10. 10.

    Ding Y, Sohn JH, Kawczynski MG, Trivedi H, Harnish R, Jenkins NW, et al. A deep learning model to predict a diagnosis of Alzheimer disease by using (18)F-FDG PET of the brain. Radiology 2019;290:456-64.

    Article  Google Scholar 

  11. 11.

    Sanaat A, Arabi H, Mainta I, Garibotto V, Zaidi H. Projection-space implementation of deep learning-guided low-dose brain PET imaging improves performance over implementation in image-space. J Nucl Med 2020 (in press).

  12. 12.

    Shiri I, Leung K, Ghafarian P, Geramifar P, Oveisi M, Ay MR, et al. HiResPET: High resolution PET image generation using deep convolution encoder decoder network [abstract]. J Nucl Med 2019;60:1368.

    Google Scholar 

  13. 13.

    Zhu B, Liu JZ, Cauley SF, Rosen BR, Rosen MS. Image reconstruction by domain-transform manifold learning. Nature 2018;555:487-92.

    CAS  Article  Google Scholar 

  14. 14.

    Zhong Z, Kim Y, Plichta K, Allen BG, Zhou L, Buatti J, et al. Simultaneous cosegmentation of tumors in PET-CT images using deep fully convolutional networks. Med Phys 2019;46:619-33.

    PubMed  PubMed Central  Google Scholar 

  15. 15.

    Xie T, Zaidi H. Estimation of the radiation dose in pregnancy: An automated patient-specific model using convolutional neural networks. Eur Radiol 2019;29:6805-15.

    Article  Google Scholar 

  16. 16.

    Shiri I, Ghafarian P, Geramifar P, Leung KH, Ghelichoghli M, Oveisi M, et al. Direct attenuation correction of brain PET images using only emission data via a deep convolutional encoder-decoder (Deep-DAC). Eur Radiol 2019;29:6867--79.

    Article  Google Scholar 

  17. 17.

    Arabi H, Zeng G, Zheng G, Zaidi H. Novel adversarial semantic structure deep learning for MRI-guided attenuation correction in brain PET/MRI. Eur J Nucl Med Mol Imaging 2019;46:2746-59.

    Article  Google Scholar 

  18. 18.

    Ma L, Ma C, Liu Y, Wang X. Thyroid diagnosis from SPECT images using convolutional neural network with optimization. Comput Intell Neurosci 2019;2019:6212759.

    PubMed  PubMed Central  Google Scholar 

  19. 19.

    Betancur J, Commandeur F, Motlagh M, Sharir T, Einstein AJ, Bokhari S, et al. Deep learning for prediction of obstructive disease from fast myocardial perfusion SPECT: A multicenter study. JACC Cardiovasc Imaging 2018;11:1654-63.

    Article  Google Scholar 

  20. 20.

    Betancur J, Hu L-H, Commandeur F, Sharir T, Einstein AJ, Fish MB, et al. Deep learning analysis of upright-supine high-efficiency spect myocardial perfusion imaging for prediction of obstructive coronary artery disease: A multicenter study. J Nucl Med 2019;60:664-70.

    Article  Google Scholar 

  21. 21.

    Wang T, Lei Y, Tang H, He Z, Castillo R, Wang C, et al. A learning-based automatic segmentation and quantification method on left ventricle in gated myocardial perfusion SPECT imaging: A feasibility study. J Nucl Cardiol 2019. https://doi.org/10.1007/s12350-019-01594-2.

    Article  PubMed  Google Scholar 

  22. 22.

    Dietze MM, Branderhorst W, Kunnen B, Viergever MA, de Jong HW. Accelerated SPECT image reconstruction with FBP and an image enhancement convolutional neural network. EJNMMI Phys 2019;6:14.

    Article  Google Scholar 

  23. 23.

    Ramon AJ, Yang Y, Pretorius PH, Johnson KL, King MA, Wernick MN. Initial investigation of low-dose spect-MPI via deep learning. 2018 IEEE Nuclear Science Symposium and Medical Imaging Conference Proceedings (NSS/MIC); 2018. p. 1-3.

  24. 24.

    Song C, Yang Y, Wernick MN, Pretorius PH, King MA. Low-dose cardiac-gated spect studies using a residual convolutional neural network. 2019 IEEE 16th International Symposium on Biomedical Imaging (ISBI 2019); 2019. p. 653-6.

  25. 25.

    Zeraatkar N, Sajedi S, Fard BT, Kaviani S, Akbarzadeh A, Farahani M, et al. Development and calibration of a new gamma camera detector using large square photomultiplier tubes. J Instrum 2017;12:P09008.

    Article  Google Scholar 

  26. 26.

    Li W, Wang G, Fidon L, Ourselin S, Cardoso MJ, Vercauteren T. On the compactness, efficiency, and representation of 3D convolutional networks: Brain parcellation as a pretext task. International Conference on Information Processing in Medical Imaging; 2017. p. 348-60.

  27. 27.

    Gibson E, Li W, Sudre C, Fidon L, Shakir DI, Wang G, et al. NiftyNet: A deep-learning platform for medical imaging. Comput Methods Programs Biomed 2018;158:113-22.

    Article  Google Scholar 

  28. 28.

    Abadi M, Barham P, Chen J, Chen Z, Davis A, Dean J et al. TensorFlow: A system for large-scale machine learning. Proceedings of the 12th USENIX conference on Operating Systems Design and Implementation. Savannah, GA, USA; 2016. p. 265-83.

  29. 29.

    Dorbala S, Blankstein R, Skali H, Park M-A, Fantony J, Mauceri C, et al. Approaches to reducing radiation dose from radionuclide myocardial perfusion imaging. J Nucl Med 2015;56:592-9.

    CAS  Article  Google Scholar 

  30. 30.

    Acampa W, Buechel RR, Gimelli A. Low dose in nuclear cardiology: State of the art in the era of new cadmium--zinc-telluride cameras. Eur Heart J Cardiovasc Imaging 2016;17:591-5.

    Article  Google Scholar 

  31. 31.

    Lecchi M, Martinelli I, Zoccarato O, Maioli C, Lucignani G, Del Sole A. Comparative analysis of full-time, half-time, and quarter-time myocardial ECG-gated SPECT quantification in normal-weight and overweight patients. J Nucl Cardiol 2017;24:876-87.

    CAS  Article  Google Scholar 

  32. 32.

    Caobelli F, Thackeray JT, Soffientini A, Bengel FM, Pizzocaro C, Guerra UP. Feasibility of one-eighth time gated myocardial perfusion SPECT functional imaging using IQ-SPECT. Eur J Nucl Med Mol Imaging 2015;42:1920-8.

    Article  Google Scholar 

  33. 33.

    Kortelainen MJ, Koivumäki TM, Vauhkonen MJ, Hakulinen MA. Dependence of left ventricular functional parameters on image acquisition time in cardiac-gated myocardial perfusion SPECT. J Nucl Cardiol 2015;22:643-51.

    Article  Google Scholar 

  34. 34.

    Asao K, Takaki A, Tominaga M, Sasaki M. The interpolated projection data estimation method improves the image quality of myocardial perfusion SPECT with a short acquisition time. Ann Nucl Med 2012;26:123-30.

    Article  Google Scholar 

  35. 35.

    Lee H, Lee J, Kim H, Cho B, Cho S. Deep-neural-network-based sinogram synthesis for sparse-view CT Image reconstruction. EEE Trans Radiat Plasma Med Sci 2018;3:109-19.

    Article  Google Scholar 

  36. 36.

    Deidda D, Karakatsanis NA, Robson PM, Efthimiou N, Fayad ZA, Aykroyd RG, et al. Effect of PET-MR inconsistency in the kernel image reconstruction method. IEEE Trans Rad Plasma Med Sci 2019;3:400-9.

    Article  Google Scholar 

  37. 37.

    Jonsson BA, Bjornsdottir G, Thorgeirsson TE, Ellingsen LM, Walters GB, Gudbjartsson DF, et al. Brain age prediction using deep learning uncovers associated sequence variants. Nat Commun 2019;10:5409.

    CAS  Article  Google Scholar 

  38. 38.

    Shiri I, Sheikhzadeh P, Ay MR. Deep-Fill: Deep learning based sinogram domain gap filling in positron emission tomography. 2019. arXiv:1906.07168.

Download references

Acknowledgments

This work was supported by the Swiss National Science Foundation under Grant SNRF 320030_176052.

Disclosure

There is no conflict of interest.

Author information

Affiliations

Authors

Corresponding author

Correspondence to Habib Zaidi PhD.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

The authors of this article have provided a PowerPoint file, available for download at SpringerLink, which summarises the contents of the paper and is free for re-use at meetings and presentations. Search for the article DOI on SpringerLink.com.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Supplementary file1 (PPTX 290 kb)

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Shiri, I., AmirMozafari Sabet, K., Arabi, H. et al. Standard SPECT myocardial perfusion estimation from half-time acquisitions using deep convolutional residual neural networks. J. Nucl. Cardiol. (2020). https://doi.org/10.1007/s12350-020-02119-y

Download citation

Keywords

  • SPECT
  • myocardial perfusion imaging
  • deep learning
  • short acquisition