AIM/
Articles/
2605004023
  • Open Access
  • Review

MRI-to-CT Generation with Deep Learning: A Review and Future Directions

  • Ruiming Zhu 1,   
  • Xinliang Liu 2,   
  • Wei Qian 1,   
  • Yueyang Teng 1,3,*

Received: 24 Jan 2026 | Revised: 30 Apr 2026 | Accepted: 25 May 2026 | Published: 15 Jun 2026

Abstract

Magnetic Resonance Imaging (MRI) and Computed Tomography (CT) play a vital role in the diagnosis of various pathologies and radiotherapy planning, in which CT is typically used for dose calculation. But MRI-only radiotherapy planning, with a synthetic CT (sCT) image synthesized from MRI, offers advantages in terms of time efficiency and patient safety. This method avoids the need for CT scans while retaining dose calculation information. Recently, deep learning models for image-to-image translation have shown great potential for MRI-to-CT synthesis, as they can efficiently preserve the common structure of the image data across different domains while changing the distinctive attributes of each domain. In this review, we discuss the four main deep learning methodologies for MRI-to-CT image synthesis: Convolutional Neural Networks (CNNs), Generative Adversarial Networks (GANs), Transformer models, and Diffusion models, discussing the potential of each model and provide insights into how to improve current MRI-to-CT image synthesis approaches.

References 

  • 1.

    Armanious, K.; Jiang, C.; Fischer, M.; et al. MedGAN: Medical image translation using GANs. Comput. Med. Imaging Graph. 2020, 79, 101684.

  • 2.

    Zhan, B.; Li, D.; Wang, Y.; et al. LR-cGAN: Latent representation based conditional generative adversarial network for multi-modality MRI synthesis. Biomed. Signal Process. Control. 2021, 66, 102457.

  • 3.

    Khosravi, B.; Purkayastha, S.; Erickson, B.J.; et al. Exploring the potential of generative artificial intelligence in medical image synthesis: Opportunities, challenges, and future directions. Lancet Digit. Health. 2025, 7, 100890.

  • 4.

    Wang, Y.R.; Baratto, L.; Hawk, K.E.; et al. Artificial intelligence enables whole-body positron emission tomography scans with minimal radiation exposure. Eur. J. Nucl. Med. Mol. Imaging 2021, 48, 2771–2781.

  • 5.

    Merida, I.; Costes, N.; Heckemann, R.; et al. Pseudo-CT generation in brain MR-PET attenuation correction: Comparison of several multi-atlas methods. EJNMMI Phys. 2015, 2(Suppl. 1), A29.

  • 6.

    Burgos, N.; Cardoso, M.J.; Thielemans, K.; et al. Attenuation correction synthesis for hybrid PET-MR scanners: Application to brain studies. IEEE Trans. Med. Imaging 2014, 33, 2332–2341.

  • 7.

    Uh, J.; Merchant, T.E.; Li, Y.; et al. MRI-based treatment planning with pseudo CT generated through atlas registration. Med. Phys. 2014, 41, 051711.

  • 8.

    Torrado-Carvajal, A.; Herraiz, J.L.; Alcain, E.; et al. Fast patch-based pseudo-CT synthesis from T1-weighted MR images for PET/MR attenuation correction in brain studies. J. Nucl. Med. 2016, 57, 136–143.

  • 9.

    Sjölund, J.; Forsberg, D.; Andersson, M.; et al. Generating patient specific pseudo-CT of the head from MR using atlas-based regression. Phys. Med. Biol. 2015, 60, 825–839.

  • 10.

    Torrado-Carvajal, A.; Herraiz, J.L.; Hernandez-Tamames, J.A.; et al. Multi-atlas and label fusion approach for patient-specific MRI based skull estimation. Magn. Reson. Med. 2016, 75, 1797–1807.

  • 11.

    Berker, Y.; Franke, J.; Salomon, A.; et al. MRI-based attenuation correction for hybrid PET/MRI systems: A 4-class tissue segmentation technique using a combined ultrashort-echo-time/Dixon MRI sequence. J. Nucl. Med. 2012, 53, 796–804.

  • 12.

    Hsu, S.H.; Cao, Y.; Huang, K.; et al. Investigation of a method for generating synthetic CT models from MRI scans of the head and neck for radiation therapy. Phys. Med. Biol. 2013, 58, 8419–8435.

  • 13.

    Zheng, W.; Kim, J.P.; Kadbi, M.; et al. Magnetic resonance–based automatic air segmentation for generation of synthetic computed tomography scans in the head region. Int. J. Radiat. Oncol. Biol. Phys. 2015, 93, 497–506.

  • 14.

    Ladefoged, C.N.; Benoit, D.; Law, I.; et al. Region specific optimization of continuous linear attenuation coefficients based on UTE (RESOLUTE): Application to PET/MR brain imaging. Phys. Med. Biol. 2015, 60, 8047–8065.

  • 15.

    Izquierdo-Garcia, D.; Hansen, A.E.; Förster, S.; et al. An SPM8-based approach for attenuation correction combining segmentation and nonrigid template formation: Application to simultaneous PET/MR brain imaging. J. Nucl. Med. 2014, 55, 1825–1830.

  • 16.

    Yu, B.; Wang, Y.; Wang, L.; et al. Medical image synthesis via deep learning. In Deep Learning in Medical Image Analysis; Advances in Experimental Medicine and Biology; Springer: Cham, Switzerland, 2020; Vol. 1213, pp. 23–44.

  • 17.

    Johansson, A.; Garpebring, A.; Karlsson, M.; et al. Improved quality of computed tomography substitute derived from magnetic resonance (MR) data by incorporation of spatial information–potential application for MR-only radiotherapy and attenuation correction in positron emission tomography. Acta Oncol. 2013, 52, 1369–1373.

  • 18.

    Dowling, J.A.; Lambert, J.; Parker, J.; et al. An atlas-based electron density mapping method for magnetic resonance imaging (MRI)-alone treatment planning and adaptive MRI-based prostate radiation therapy. Int. J. Radiat. Oncol. Biol. Phys. 2012, 83, 5–11.

  • 19.

    Bahrami, A.; Karimian, A.; Fatemizadeh, E.; et al. A new deep convolutional neural network design with efficient learning capability: Application to CT image synthesis from MRI. Med. Phys. 2020, 47, 5158–5171.

  • 20.

    Wang, C.; Uh, J.; He, X.; et al. Transfer learning-based synthetic CT generation for MR-only proton therapy planning in children with pelvic sarcomas. In Medical Imaging 2021: Physics of Medical Imaging; SPIE Digital Library: Bellingham, WA, USA, 2021; Vol. 11595, pp. 1112–1118.

  • 21.

    Spadea, M.F.; Pileggi, G.; Zaffino, P.; et al. Deep convolution neural network (DCNN) multiplane approach to synthetic CT generation from MR images—Application in brain proton therapy. Int. J. Radiat. Oncol. Biol. Phys. 2019, 105, 495–503.

  • 22.

    Han, X. MR-based synthetic CT generation using a deep convolutional neural network method. Med. Phys. 2017, 44, 1408–1419.

  • 23.

    Florkow, M.C.; Zijlstra, F.; Willemsen, K.; et al. Deep learning-based MR-to-CT synthesis: The influence of varying gradient echo-based MR images as input channels. Magn. Reson. Med. 2020, 83, 1429–1441.

  • 24.

    Zhao, B.; Cheng, T.; Zhang, X.; et al. CT synthesis from MR in the pelvic area using Residual Transformer Conditional GAN. Comput. Med. Imaging Graph. 2023, 103, 102150.

  • 25.

    Siam, Z.S.; Akon, M.Y.; Munmun, I.J.; et al. A paired CT and MRI dataset for advanced medical imaging applications. Data Brief. 2025, 61, 111768.

  • 26.

    Gu, X.; Wang, L.; Deng, Z.; et al. Adaptive spatial and frequency experts fusion network for medical image fusion. Biomed. Signal Process. Control. 2024, 96, 106478.

  • 27.

    Nyholm, T.; Svensson, S.; Andersson, S.; et al. MR and CT data with multiobserver delineations of organs in the pelvic area—Part of the Gold Atlas project. Med. Phys. 2018, 45, 1295–1300.

  • 28.

    Thummerer, A.; van der Bijl, E.; Galapon, A.J.; et al. SynthRAD2025 Grand Challenge dataset: Generating synthetic CTs for radiotherapy from head to abdomen. Med. Phys. 2025, 52, e17981.

  • 29.

    Clark, K.; Vendt, B.; Smith, K.; et al. The Cancer Imaging Archive (TCIA): Maintaining and operating a public information repository. J. Digit. Imaging 2013, 26, 1045–1057.

  • 30.

    Al-Kadi, O.S.; Almallahi, I.; Abu-Srhan, A.; et al. Unpaired MR-CT brain dataset for unsupervised image translation. Data Brief. 2022, 42, 108109.

  • 31.

    He, K.; Zhang, X.; Ren, S.; et al. Deep residual learning for image recognition. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Las Vegas, NV, USA, 27–30 June 2016; pp. 770–778.

  • 32.

    Ronneberger, O.; Fischer, P.; Brox, T. U-Net: Convolutional networks for biomedical image segmentation. In International Conference on Medical Image Computing and Computer-Assisted Intervention, Munich, Germany, 5–9 October 2015; pp. 234–241.

  • 33.

    Nie, D.; Cao, X.; Gao, Y.; et al. Estimating CT image from MRI data using 3D fully convolutional networks. In Proceedings of the International Workshop on Deep Learning in Medical Image Analysis, Athens, Greece, 21 October 2016; pp. 170–178.

  • 34.

    Xiang, L.; Wang, Q.; Nie, D.; et al. Deep embedding convolutional neural network for synthesizing CT image from T1-Weighted MR image. Med. Image Anal. 2018, 47, 31–44.

  • 35.

    Zhao, Y.; Liao, S.; Guo, Y.; et al. Towards MR-only radiotherapy treatment planning: Synthetic CT generation using multi-view deep convolutional neural networks. In Proceedings of the International Conference on Medical Image Computing and Computer-Assisted Intervention, Granada, Spain, 16–20 September 2018; pp. 286–294.

  • 36.

    Li, Y.; Xu, S.; Chen, H.; et al. CT synthesis from multi-sequence MRI using adaptive fusion network. Comput. Biol. Med. 2023, 157, 106738.

  • 37.

    Dalmaz, O.; Yurt, M.; Çukur, T. ResViT: Residual vision transformers for multimodal medical image synthesis. IEEE Trans. Med. Imaging 2022, 41, 2598–2614.

  • 38.

    Li, Y.; Xu, S.; Lu, Y.; Qi, Z. CT synthesis from MRI with an improved multi-scale learning network. Front. Phys. 2023, 11, 1088899.

  • 39.

    Zhong, L.; Chen, Z.; Shu, H.; et al. Multi-scale tokens-aware transformer network for multi-region and multi-sequence MR-to-CT synthesis in a single model. IEEE Trans. Med. Imaging 2023, 43, 794–806.

  • 40.

    Mekki, L.; Ladra, M.; Acharya, S.; et al. Generative evidential synthesis with integrated segmentation framework for MR-only radiation therapy treatment planning. Med. Phys. 2025, 52, e17828.

  • 41.

    You, L.; Wang, H.; Matta, E.J.; et al. SeRL: Style-embedding representation learning for unsupervised CT images synthesis from unpaired MR images. Biomed. Signal Process. Control. 2024, 94, 106280.

  • 42.

    Mendes, J.; Pereira, T.; Silva, F.; et al. Lung CT image synthesis using GANs. Expert Syst. Appl. 2023, 215, 119350.

  • 43.

    Emami, H.; Dong, M.; Nejad-Davarani, S.P.; et al. Generating synthetic CTs from magnetic resonance images using generative adversarial networks. Med. Phys. 2018, 45, 3627–3636.

  • 44.

    Arabi, H.; Zeng, G.; Zheng, G.; et al. Novel adversarial semantic structure deep learning for MRI-guided attenuation correction in brain PET/MRI. Eur. J. Nucl. Med. Mol. Imaging 2019, 46, 2746–2759.

  • 45.

    Longuefosse, A.; Denis de Senneville, B.; Dournes, G.; et al. Anatomical feature-prioritized loss for enhanced MR to CT translation. Phys. Med. Biol. 2025, 70, 145012.

  • 46.

    Han, S.; Hémon, C.; Texier, B.; et al. Balancing data consistency and diversity: Preprocessing and online data augmentation for multi-center deep learning-based MR-to-CT synthesis. Pattern Recognit. Lett. 2025, 189, 56–63.

  • 47.

    Getzmann, J.M.; Deininger-Czermak, E.; Melissanidis, S.; et al. Deep learning-based pseudo-CT synthesis from zero echo time MR sequences of the pelvis. Insights Imaging 2024, 15, 202.

  • 48.

    Yang, H.; Sun, J.; Carass, A.; et al. Unsupervised MR-to-CT synthesis using structure-constrained CycleGAN. IEEE Trans. Med. Imaging 2020, 39, 4249–4261.

  • 49.

    Ge, Y.; Wei, D.; Xue, Z.; et al. Unpaired MR to CT synthesis with explicit structural constrained adversarial learning. In Proceedings of the 2019 IEEE 16th International Symposium on Biomedical Imaging (ISBI 2019), Venice, Italy, 8–11 April 2019; pp. 1096–1099.

  • 50.

    Kang, S.K.; An, H.J.; Jin, H.; et al. Synthetic CT generation from weakly paired MR images using cycle-consistent GAN for MR-guided radiotherapy. Biomed. Eng. Lett. 2021, 11, 263–271.

  • 51.

    Wang, J.; Wu, X.; Jiang, X.; et al. MRI to CT synthesis using contrastive learning. In Proceedings of the 2021 IEEE International Conference on Medical Imaging Physics and Engineering (ICMIPE), Hefei, China, 13–14 November 2021; pp. 1–5.

  • 52.

    Wang, J.; Yan, B.; Wu, X.; et al. Development of an unsupervised cycle contrastive unpaired translation network for MRI-to-CT synthesis. J. Appl. Clin. Med. Phys. 2022, 23, e13775.

  • 53.

    Zhu, R.; Liu, X.; Li, M.; et al. A Novel Perceptual Constrained cycleGAN with Attention Mechanisms for Unsupervised MR-to-CT Synthesis. Int. J. Imaging Syst. Technol. 2024, 34, e23169.

  • 54.

    Wang, R.; Heimann, A.F.; Tannast, M.; et al. CycleSGAN: A cycle-consistent and semantics-preserving generative adversarial network for unpaired MR-to-CT image synthesis. Comput. Med. Imaging Graph. 2024, 117, 102431.

  • 55.

    Tang, M.; Jiang, J.; Zhang, X.; et al. Dynamic Multi-scale Feature Integration Network for unsupervised MR-CT synthesis. Neural Netw. 2025, 189, 107584.

  • 56.

    Zhu, R.; Ruan, Y.; Li, M.; et al. Quasi-supervised MR-CT image conversion based on unpaired data. Phys. Med. Biol. 2025, 70, 125010.

  • 57.

    Lyu, Q.; Wang, G. Conversion between CT and MRI images using diffusion and score-matching models. arXiv 2022, preprint, arXiv:2209.12104.

  • 58.

    Özbey, M.; Dalmaz, O.; Dar, S.U.; et al. Unsupervised medical image translation with adversarial diffusion models. IEEE Trans. Med. Imaging 2023, 42, 3524–3539.

  • 59.

    Graf, R.; Schmitt, J.; Schlaeger, S.; et al. Denoising diffusion-based MRI to CT image translation enables automated spinal segmentation. Eur. Radiol. Exp. 2023, 7, 70.

  • 60.

    Pan, S.; Wang, T.; Qiu, R.L.; et al. 2D medical image synthesis using transformer-based denoising diffusion probabilistic model. Phys. Med. Biol. 2023, 68, 105004.

  • 61.

    Li, X.; Bellotti, R.; Bachtiary, B.; et al. A unified generation-registration framework for improved MR-based CT synthesis in proton therapy. Med. Phys. 2024, 51, 8302–8316.

  • 62.

    Zheng, J.; Shen, Z.; Zhang, L.; et al. Structure-guided MR-to-CT synthesis with spatial and semantic alignments for attenuation correction of whole-body PET/MR imaging. Med. Image Anal. 2025, 103, 103622.

  • 63.

    Deng, L.; Sun, H.; Huang, S.; et al. Research on multi-sequence MR image synthesis CT algorithm based on unsupervised method. Biomed. Signal Process. Control. 2025, 99, 106819.

  • 64.

    Zhuang, Y.; Mathai, T.S.; Mukherjee, P.; et al. Segmentation of pelvic structures in T2 MRI via MR-to-CT synthesis. Comput. Med. Imaging Graph. 2024, 112, 102335.

  • 65.

    Hsu, S.H.; DuPre, P.; Peng, Q.; et al. A technique to generate synthetic CT from MRI for abdominal radiotherapy. J. Appl. Clin. Med. Phys. 2020, 21, 136–143. https://doi.org/10.1002/acm2.12816.

  • 66.

    McKenzie, E.M.; Santhanam, A.; Ruan, D.; et al. Multimodality image registration in the head-and-neck using a deep learning-derived synthetic CT as a bridge. Med. Phys. 2020, 47, 1094–1104. https://doi.org/10.1002/mp.13976.

  • 67.

    Heo, J.U.; Zhou, F.; Jones, R.; et al. Abdominopelvic MR to CT registration using a synthetic CT intermediate. J. Appl. Clin. Med. Phys. 2022, 23, e13731. https://doi.org/10.1002/acm2.13731.

  • 68.

    Low, D.A.; Harms, W.B.; Mutic, S.; et al. A technique for the quantitative evaluation of dose distributions. Med. Phys. 1998, 25, 656–661. https://doi.org/10.1118/1.598248.

  • 69.

    Low, D.A.; Dempsey, J.F. Evaluation of the gamma dose distribution comparison method. Med. Phys. 2003, 30, 2455–2464. https://doi.org/10.1118/1.1598711.

  • 70.

    Wang, H.; Chandarana, H.; Block, K.T.; et al. Dosimetric evaluation of synthetic CT for magnetic resonance-only based radiotherapy planning of lung cancer. Radiat. Oncol. 2017, 12, 108.

  • 71.

    Akbar, M.U.; Wang, W.; Eklund, A. Beware of diffusion models for synthesizing medical images—A comparison with GANs in terms of memorizing brain MRI and chest X-ray images. Mach. Learn. Sci. Technol. 2025, 6, 015022.

  • 72.

    Tang, R.; Liu, L.; Pandey, A.; et al. What the daam: Interpreting stable diffusion using cross attention. In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), Toronto, Canada, 9–14 July 2023; pp. 5644–5659.

  • 73.

    Horwitz, E.; Hoshen, Y. Conffusion: Confidence intervals for diffusion models. arXiv 2022, preprint, arXiv:2211.09795.

  • 74.

    Luccioni, A.S.; Akiki, C.; Mitchell, M.; et al. Stable bias: Analyzing societal representations in diffusion models. arXiv 2023, preprint, arXiv:2303.11408.

  • 75.

    Seyyed-Kalantari, L.; Zhang, H.; McDermott, M.B.; et al. Underdiagnosis bias of artificial intelligence algorithms applied to chest radiographs in under-served patient populations. Nat. Med. 2021, 27, 2176–2182.

  • 76.

    Earwong, P.; Puttanawarut, C.; Suphaphong, S.; et al. Clinical implementation of deep learning-based synthetic CT for MRI-only volumetric modulated arc therapy in head and neck and pelvic cancer patients. Radiat. Oncol. 2025, 20, 166.

  • 77.

    Aljaafari, L.; Speight, R.; Buckley, D.L.; et al. Clinical validation of using a commercial synthetic-computed tomography solution for brain MRI-only radiotherapy treatment planning. Tech. Innov. Patient Support Radiat. Oncol. 2025, 35, 100328.

Share this article:
Download PDF
DOI
10.53941/aim.2026.100004
Issue
Volume 3, Issue 1
How to Cite
Zhu, R.; Liu, X.; Qian, W.; Teng, Y. MRI-to-CT Generation with Deep Learning: A Review and Future Directions. AI Medicine 2026, 3 (1), 4. https://doi.org/10.53941/aim.2026.100004.
RIS
BibTex
Copyright & License
article copyright Image
Copyright (c) 2026 by the authors.