Contrast agent dose reduction in computed tomography with deep learning using a conditional generative adversarial network.
Contrast media
Image processing, computer-assisted
Tomography, spiral computed
Journal
European radiology
ISSN: 1432-1084
Titre abrégé: Eur Radiol
Pays: Germany
ID NLM: 9114774
Informations de publication
Date de publication:
Aug 2021
Aug 2021
Historique:
received:
24
09
2020
accepted:
21
01
2021
revised:
13
12
2020
pubmed:
26
2
2021
medline:
14
7
2021
entrez:
25
2
2021
Statut:
ppublish
Résumé
To reduce the dose of intravenous iodine-based contrast media (ICM) in CT through virtual contrast-enhanced images using generative adversarial networks. Dual-energy CTs in the arterial phase of 85 patients were randomly split into an 80/20 train/test collective. Four different generative adversarial networks (GANs) based on image pairs, which comprised one image with virtually reduced ICM and the original full ICM CT slice, were trained, testing two input formats (2D and 2.5D) and two reduced ICM dose levels (-50% and -80%). The amount of intravenous ICM was reduced by creating virtual non-contrast series using dual-energy and adding the corresponding percentage of the iodine map. The evaluation was based on different scores (L1 loss, SSIM, PSNR, FID), which evaluate the image quality and similarity. Additionally, a visual Turing test (VTT) with three radiologists was used to assess the similarity and pathological consistency. The -80% models reach an SSIM of > 98%, PSNR of > 48, L1 of between 7.5 and 8, and an FID of between 1.6 and 1.7. In comparison, the -50% models reach a SSIM of > 99%, PSNR of > 51, L1 of between 6.0 and 6.1, and an FID between 0.8 and 0.95. For the crucial question of pathological consistency, only the 50% ICM reduction networks achieved 100% consistency, which is required for clinical use. The required amount of ICM for CT can be reduced by 50% while maintaining image quality and diagnostic accuracy using GANs. Further phantom studies and animal experiments are required to confirm these initial results. • The amount of contrast media required for CT can be reduced by 50% using generative adversarial networks. • Not only the image quality but especially the pathological consistency must be evaluated to assess safety. • A too pronounced contrast media reduction could influence the pathological consistency in our collective at 80%.
Identifiants
pubmed: 33630160
doi: 10.1007/s00330-021-07714-2
pii: 10.1007/s00330-021-07714-2
pmc: PMC8270814
doi:
Substances chimiques
Contrast Media
0
Types de publication
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Pagination
6087-6095Subventions
Organisme : Deutsche Forschungsgemeinschaft
ID : FU 356/12-1
Informations de copyright
© 2021. The Author(s).
Références
Tonelli M, Riella M (2014) Chronic kidney disease and the aging population. Am J Physiol Renal Physiol 306:F469–F472. https://doi.org/10.1152/ajprenal.00063.2014
doi: 10.1152/ajprenal.00063.2014
pubmed: 24500692
Davenport MS, Perazella MA, Yee J et al (2020) Use of intravenous iodinated contrast media in patients with kidney disease: consensus statements from the American College of Radiology and the National Kidney Foundation. Radiology 294: 660–668. https://doi.org/10.1148/radiol.2019192094
Higashigaito K, Schmid T, Puippe G et al (2016) CT Angiography of the aorta: prospective evaluation of individualized low-volume contrast media protocols. Radiology 280:960–968. https://doi.org/10.1148/radiol.2016151982
doi: 10.1148/radiol.2016151982
pubmed: 26937711
Johnson TRC, Krauss B, Sedlmair M et al (2007) Material differentiation by dual energy CT: initial experience. Eur Radiol 17:1510–1517. https://doi.org/10.1007/s00330-006-0517-6
doi: 10.1007/s00330-006-0517-6
pubmed: 17151859
Wang T-C, Liu M-Y, Zhu J-Y, Tao A, Kautz J, Catanzaro B (2018) High-resolution image synthesis and semantic manipulation with conditional GANs. In: The IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
Odena A, Dumoulin V, Olah C (2016) Deconvolution and checkerboard artifacts. Distill. https://doi.org/10.23915/distill.00003
Wojna Z, Ferrari V, Guadarrama S et al (2019) The devil is in the decoder: classification, regression and GANs. Int J Comput Vis 127:1694–1706. https://doi.org/10.1007/s11263-019-01170-8
doi: 10.1007/s11263-019-01170-8
Heusel M, Ramsauer H, Unterthiner T et al (2017) GANs trained by a two time-scale update rule converge to a local Nash equilibrium. In: Proceedings of the 31st International Conference on Neural Information Processing Systems. Curran Associates Inc., Red Hook, NY, USA, pp 6629–6640
Wang Z, Bovik AC, Sheikh HR, Simoncelli EP (2004) Image quality assessment: from error visibility to structural similarity. IEEE Trans Image Process 13:600–612
Horé A, Ziou D (2010) Image quality metrics: PSNR vs. SSIM. In: 2010 20th International Conference on Pattern Recognition. pp 2366–2369
Salimans T, Goodfellow I, Zaremba W, Cheung V, Radford A, Chen X (2016) Improved techniques for training gans. In: Advances in neural information processing systems, pp 2234–2242
Chuquicusma MJ, Hussein S, Burt J, Bagci U (2018) How to fool radiologists with generative adversarial networks? A visual Turing test for lung cancer diagnosis. In: 2018 IEEE 15th international symposium on biomedical imaging (ISBI 2018). IEEE, pp 240–244
Cohen JP, Luck M, Honari S (2018) Distribution matching losses can hallucinate features in medical image translation. In: Frangi AF, Schnabel JA, Davatzikos C et al (eds) Medical Image Computing and Computer Assisted Intervention – MICCAI 2018. Springer International Publishing, Cham, pp 529–536
doi: 10.1007/978-3-030-00928-1_60
Kleesiek J, Morshuis JN, Isensee F et al (2019) Can virtual contrast enhancement in brain MRI replace gadolinium?: a feasibility study. Invest Radiol 54:653–660
doi: 10.1097/RLI.0000000000000583
Wolterink JM, Leiner T, Viergever MA, Išgum I (2017) Generative adversarial networks for noise reduction in low-dose CT. IEEE Trans Med Imaging 36:2536–2545. https://doi.org/10.1109/TMI.2017.2708987
doi: 10.1109/TMI.2017.2708987
pubmed: 28574346
Yang Q, Yan P, Zhang Y et al (2018) Low-dose CT image denoising using a generative adversarial network with Wasserstein distance and perceptual loss. IEEE Trans Med Imaging 37:1348–1357. https://doi.org/10.1109/TMI.2018.2827462
doi: 10.1109/TMI.2018.2827462
pubmed: 29870364
pmcid: 6021013
Suntharalingam S, Mikat C, Stenzel E et al (2017) Submillisievert standard-pitch CT pulmonary angiography with ultra-low dose contrast media administration: a comparison to standard CT imaging. PLoS One 12:e0186694. https://doi.org/10.1371/journal.pone.0186694
doi: 10.1371/journal.pone.0186694
pubmed: 29045463
pmcid: 5646863
Zhang W, Ba Z, Wang Z et al (2018) Diagnostic performance of low-radiation-dose and low-contrast-dose (double low-dose) coronary CT angiography for coronary artery stenosis. Medicine (Baltimore) 97:e11798. https://doi.org/10.1097/MD.0000000000011798
doi: 10.1097/MD.0000000000011798
Lira D, Padole A, Kalra MK, Singh S (2014) Tube potential and CT radiation dose optimization. AJR Am J Roentgenol 204:W4–W10. https://doi.org/10.2214/AJR.14.13281
doi: 10.2214/AJR.14.13281
Fursevich DM, LiMarzi GM, O’Dell MC, Hernandez MA, Sensakovic WF et al (2016) Bariatric CT imaging: challenges and solutions. Radiographics 36:1076–1086. https://doi.org/10.1148/rg.2016150198
Flegal KM, Carroll MD, Ogden CL, Curtin LR (2010) Prevalence and trends in obesity among US adults, 1999-2008. JAMA 303:235–241. https://doi.org/10.1001/jama.2009.2014
doi: 10.1001/jama.2009.2014
pubmed: 20071471
Finkelstein EA, Khavjou OA, Thompson H et al (2012) Obesity and severe obesity forecasts through 2030. Am J Prev Med 42:563–570. https://doi.org/10.1016/j.amepre.2011.10.026
doi: 10.1016/j.amepre.2011.10.026
pubmed: 22608371
Han C, Hayashi H, Rundo L, et al (2018) GAN-based synthetic brain MR image generation. In: 2018 IEEE 15th International Symposium on Biomedical Imaging (ISBI 2018). pp 734–738