This study introduces a novel deep learning approach to enhance the accuracy and efficiency of diagnosis in single-photon emission computed tomography myocardial perfusion imaging (SPECT MPI). To address key limitations of current convolutional neural network (CNN)-based methods—such as insufficient information capture, difficulty in removing redundant features, and limited capacity for modeling long-range dependencies—we reconstruct the three-dimensional structure of myocardial perfusion images in a stacked format and propose a multi-branch medical transformer network. This architecture extracts comprehensive features from different anatomical views while integrating critical information, leveraging the Transformer's strength in capturing long-range dependencies to overcome traditional CNN shortcomings. Experimental results demonstrate that the proposed method consistently outperforms conventional CNN-based models across multiple evaluation metrics, achieving improved feature extraction and higher diagnostic accuracy. Comparative experiments and ablation studies further validate the effectiveness of the multi-branch Transformer architecture. The proposed multi-branch vision transformer provides a powerful tool for automated SPECT MPI diagnosis, enhancing diagnostic performance and offering potential support for clinical decision-making.

Fuling Zhao, Xuande Zhang, Long Xu, Xin Huang. Multi-Branch Medical Transformer for SPECT Myocardial Perfusion Imaging: A Novel Approach to Diagnosis. Academic Journal of Computing & Information Science (2025), Vol. 8, Issue 11: 38-51. https://doi.org/10.25236/AJCIS.2025.081105.

[1] Stark B, Johnson C, Roth G. Global prevalence of coronary artery disease: an update from the Global Burden of Disease Study. J Am Coll Cardiol 2024;83(13 Suppl):2320. doi:10.1016/S0735-1097(24)04310-9.

[2] Liu X, Wu Y, Li F, Qi X, Niu L, Wu Y, et al. Global burden of early-onset ischemic heart disease, 1990 to 2019. JACC Adv 2025;4(1):101466. doi:10.1016/j.jacadv.2024.101466.

[3] Notghi A, Low CS. Myocardial perfusion scintigraphy: past, present and future. Br J Radiol 2011;84(Spec Iss 3):S229–S236. doi:10.1259/bjr/14625142.

[4] International Atomic Energy Agency. Nuclear cardiology: guidance on the implementation of SPECT myocardial perfusion imaging. IAEA Human Health Series No. 23 (Rev. 1). Vienna: IAEA; 2016.

[5] Zhang H, Qie Y. Applying deep learning to medical imaging: a review. Appl Sci 2023; 13:10521. doi:10.3390/app131810521.

[6] Yu H, Yang LT, Zhang Q, Armstrong D, Deen MJ. Convolutional neural networks for medical image analysis: state-of-the-art, comparisons, improvement and perspectives. Neurocomputing 2021; 444:92–110. doi:10.1016/j.neucom.2020.04.157.

[7] He K, Zhang X, Ren S, Sun J. Deep residual learning for image recognition. In: Proc IEEE Conf Comput Vis Pattern Recognit; 2016. p. 770–778. doi:10.1109/CVPR.2016.90.

[8] Simonyan K, Zisserman A. Very deep convolutional networks for large-scale image recognition. arXiv preprint arXiv:1409.1556, 2014.

[9] Szegedy C, Liu W, Jia Y, Sermanet P, Reed S, Anguelov D, et al. Going deeper with convolutions. Proc IEEE Conf Comput Vis Pattern Recognit 2015;1–9. doi:10.1109/CVPR.2015.7298594

[10] Zhao Z, Alzubaidi L, Zhang J, Duan Y, Gu Y. A comparison review of transfer learning and self-supervised learning: Definitions, applications, advantages and limitations. Expert Syst Appl 2024; 242:122807. doi:10.1016/j.eswa.2023.122807

[11] Ellis RJ, Sander RM, Limon A. Twelve key challenges in medical machine learning and solutions. Intell Based Med 2022; 6:100068. doi:10.1016/j.ibmed.2022.100068

[12] Pan SJ, Yang Q. A survey on transfer learning. IEEE Trans Knowl Data Eng 2010; 22:1345–1359. doi:10.1109/TKDE.2009.191

[13] Amini MR, Feofanov V, Pauletto L, Hadjadj L, Devijver É, Maximov Y. Self-training: a survey. Neurocomputing 2025; 616:128904. doi:10.1016/j.neucom.2024.128904

[14] Jiao J, Droste R, Drukker L, Papageorghiou AT, Noble JA. Self-supervised representation learning for ultrasound video. IEEE Int Symp Biomed Imaging 2020:1847–1850. doi:10.1109/ISBI45749.2020.9098666

[15] Lopes RR, Bleijendaal H, Ramos LA, Verstraelen TE, Amin AS, Wilde AA, et al. Improving electrocardiogram-based detection of rare genetic heart disease using transfer learning: An application to phospholamban p. Arg14del mutation carriers. Comput Biol Med 2021; 131:104262. doi:10.1016/j.compbiomed.2021.104262

[16] Kathamuthu ND, Subramaniam S, Le QH, Muthusamy S, Panchal H, Sundararajan SCM, et al. A deep transfer learning-based convolution neural network model for COVID-19 detection using computed tomography scan images for medical applications. Adv Eng Softw 2023; 175:103317. doi:10.1016/j.advengsoft.2022.103317

[17] Deng J, Dong W, Socher R, Li L-J, Li K, Fei-Fei L. ImageNet: A large-scale hierarchical image database. Proc IEEE Conf Comput Vis Pattern Recognit 2009; 248–255. doi:10.1109/CVPR.2009.5206848

[18] Vaswani A, Shazeer N, Parmar N, Uszkoreit J, Jones L, Gomez AN, et al. Attention is all you need. In: Advances in Neural Information Processing Systems (NeurIPS), 2017, pp. 5998–6008.

[19] Dosovitskiy A, Beyer L, Kolesnikov A, Weissenborn D, Zhai X, Unterthiner T, et al. An image is worth 16x16 words: Transformers for image recognition at scale. arXiv preprint arXiv:2010.11929, 2020.

[20] Murphy ZR, Venkatesh K, Sulam J, Yi PH. Visual transformers and convolutional neural networks for disease classification on radiographs: a comparison of performance, sample efficiency, and hidden stratification. Radiol Artif Intell 2022; 4(6):e220012. doi:10.1148/ryai.220012.

[21] Pachetti E, Colantonio S, Pascali MA. On the effectiveness of 3D vision transformers for the prediction of prostate cancer aggressiveness. In: Proc Int Conf Image Anal Process; 2022. p. 317–328. doi:10.1007/978-3-031-13324-4_27.

[22] Hara K, Kataoka H, Satoh Y. Can spatiotemporal 3D CNNs retrace the history of 2D CNNs and ImageNet? In: Proc IEEE Conf Comput Vis Pattern Recognit; 2018. p. 6546–6555. doi:10.1109/CVPR.2018.00684

[23] Chen X, Liu C. Deep-learning-based methods of attenuation correction for SPECT and PET. J Nucl Cardiol 2023;30(5):1859-1878.

[24] Nguyen TT, Chi TN, Hoang MD, Thai HN, Duc TN. 3D Unet generative adversarial network for attenuation correction of SPECT images. In: 2020 4th international conference on recent advances in signal processing, telecommunications & computing (SigTelCom). Hanoi, Vietnam: IEEE, 2020:93–97.

[25] Chen X, Hendrik Pretorius P, Zhou B, Liu H, Johnson K, Liu YH, King MA, Liu C. Cross-vender, cross-tracer, and cross-protocol deep transfer learning for attenuation map generation of cardiac SPECT. J Nucl Cardiol 2022;29(6):3379-3391. 

[26] Shanbhag AD, Miller RJH, Pieszko K, Lemley M, Kavanagh P, Feher A, et al. Deep learning-based attenuation correction improves diagnostic accuracy of cardiac SPECT. J Nucl Med 2023;64(3):472-478.

[27] Apostolopoulos ID, Papandrianos NI, Feleki A, Moustakidis S, Papageorgiou EI. Deep learning-enhanced nuclear medicine SPECT imaging applied to cardiac studies. EJNMMI Phys 2023;10(1):6.

[28] Shiri I, AmirMozafari Sabet K, Arabi H, Pourkeshavarz M, Teimourian B, Ay MR, et al. Standard SPECT myocardial perfusion estimation from half-time acquisitions using deep convolutional residual neural networks. J Nucl Cardiol 2021;28(6):2761-2779.

[29] Pan Z, Qi N, Meng Q, Pan B, Feng T, Zhao J, Gong NJ. Fast SPECT/CT planar bone imaging enabled by deep learning enhancement. Med Phys 2024;51(8):5414–5426.

[30] Song C, Yang Y, Wernick MN, Pretorius PH, King MA. Low-dose cardiac-gated SPECT studies using a residual convolutional neural network. In: 2019 IEEE 16th international symposium on biomedical imaging (ISBI 2019). Venice, Italy: IEEE, 2019:653–656.

[31] Ramon AJ, Yang Y, Pretorius PH, Johnson KL, King MA, Wernick MN. Improving diagnostic accuracy in low-dose SPECT myocardial perfusion imaging with convolutional denoising networks. IEEE Trans Med Imaging 2020; 39:2893–2903.

[32] Wu R, Liu H, Lai P, Yuan W, Li H, Jiang Y. Sinogram-characteristic-informed network for efficient restoration of low-dose SPECT projection data. Med Phys 2025;52(1):414–432.

[33] Berkaya SK, Sivrikoz IA, Gunal S. Classification models for SPECT myocardial perfusion imaging. Comput Biol Med 2020; 123:103893. doi:10.1016/j.compbiomed.2020.103893

[34] Magboo VPC, Magboo MSA. Diagnosis of coronary artery disease from myocardial perfusion imaging using convolutional neural networks. Procedia Comput Sci 2023; 218:810-817. doi:10.1016/j.procs.2023.01.061

[35] Magboo VPC, Magboo MSA. SPECT-MPI for coronary artery disease: a deep learning approach. Acta Med Philipp 2024 May 15;58(8):67-75. doi: 10.47895/amp.vi0.7582. PMID: 38812768; PMCID: PMC11132284.

[36] Kusumoto D, Akiyama T, Hashimoto M, et al. A deep learning-based automated diagnosis system for SPECT myocardial perfusion imaging. Sci Rep 2024; 14:13583. doi: 10.1038/s41598-024-64445-2.

[37] Xie Y, Zhang J, Xia Y, Wu Q. UniMiSS: Universal medical self-supervised learning via breaking dimensionality barrier. In: Avidan S, Brostow G, Cissé M, Farinella GM, Hassner T, eds. Computer Vision – ECCV 2022. Lecture Notes in Computer Science, vol 13681. Cham: Springer; 2022. p. 555–572. doi:10.1007/978-3-031-19803-8_33.

[38] Wang W, Xie E, Li X, et al. Pyramid Vision Transformer: A versatile backbone for dense prediction without convolutions. In: Proc IEEE/CVF Int Conf Comput Vis; 2021. p. 568–578. doi:10.1109/ICCV48922.2021.00062.

[39] Caron M, Touvron H, Misra I, et al. Emerging properties in self-supervised vision transformers. In: Proc IEEE/CVF Int Conf Comput Vis; 2021. p. 9650–9660. doi:10.1109/ICCV48922.2021.00951.

[40] Chen X, Xie S, He K. An empirical study of training self-supervised vision transformers. In: Proc IEEE/CVF Int Conf Comput Vis; 2021. p. 9640–9649. doi:10.1109/ICCV48922.2021.00950.

[41] Loschilov I, Hutter F. Fixing weight decay regularization in Adam. In: Proc Int Conf Learn Representations; 2018.

[42] Huang G, Liu Z, van der Maaten L, Weinberger KQ. Densely connected convolutional networks. In: Proc IEEE Conf Comput Vis Pattern Recognit; 2017. p. 2261–2269. doi:10.1109/CVPR.2017.243.

[43] Szegedy C, Vanhoucke V, Ioffe S, Shlens J, Wojna Z. Rethinking the Inception architecture for computer vision. Proc IEEE Conf Comput Vis Pattern Recognit 2016; 2818–2826.