Low-count whole-body PET with deep learning in a multicenter and externally validated study
Open accessAbstract
More widespread use of positron emission tomography (PET) imaging is limited by its high cost and radiation dose. Reductions in PET scan time or radiotracer dosage typically degrade diagnostic image quality (DIQ). Deep-learning-based reconstruction may improve DIQ, but such methods have not been clinically evaluated in a realistic multicenter, multivendor environment. In this study, we evaluated the performance and generalizability of a deep-learning-based image-quality enhancement algorithm applied to fourfold reduced-count whole-body PET in a realistic clinical oncologic imaging environment with multiple blinded readers, institutions, and scanner types. We demonstrate that the low-count-enhanced scans were noninferior to the standard scans in DIQ (p < 0.05) and overall diagnostic confidence (p < 0.001) independent of the underlying PET scanner used. Lesion detection for the low-count-enhanced scans had a high patient-level sensitivity of 0.94 (0.83–0.99) and specificity of 0.98 (0.95–0.99). Interscan kappa agreement of 0.85 was comparable to intrareader (0.88) and pairwise inter-reader agreements (maximum of 0.72). SUV quantification was comparable in the reference regions and lesions (lowest p-value=0.59) and had high correlation (lowest CCC = 0.94). Thus, we demonstrated that deep learning can be used to restore diagnostic image quality and maintain SUV accuracy for fourfold reduced-count PET scans, with interscan variations in lesion depiction, lower than intra- and interreader variations. This method generalized to an external validation set of clinical patients from multiple institutions and scanner types. Overall, this method may enable either dose or exam-duration reduction, increasing safety and lowering the cost of PET imaging.
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The success of deep learning in computer vision analysis has significantly advanced its clinical applications in comprehending biomedical images, including CT scans, PET-CT scans, whole-slide images, and MRI scans.
“The success of deep learning in computer vision (CV) analysis has significantly advanced its clinical applications in comprehending biomedical images, including CT scans [88], positron emission tomography (PET)-CT scans [89], whole-slide images (WSIs) [90], and MRI scans [91].”
From Challenges in AI-driven Biomedical Multimodal Data Fusion and Analysis (2025) · cites this paper as [89]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Cites
U-Net: Convolutional Networks for Biomedical Image Segmentation
Ronneberger, O., Fischer, P., & Brox, T. (2015) · Lecture notes in computer science · doi:10.1007/978-3-319-24574-4_28 · Cited by 99,404 · see citing works
From RECIST to PERCIST: Evolving Considerations for PET Response Criteria in Solid Tumors
Wahl, R. L., Jacene, H., Kasamon, Y., & Lodge, M. A. (2009) · Journal of Nuclear Medicine · doi:10.2967/jnumed.108.057307 · Cited by 3,777 · see citing works
Zech, J. R., Badgeley, M. A., Liu, M., Costa, A. B., Titano, J. J., & Oermann, E. K. (2018) · PLOS Medicine · doi:10.1371/journal.pmed.1002683 · Cited by 1,719 · see citing works
Super‐resolution musculoskeletal <scp>MRI</scp> using deep learning
Chaudhari, A. S., Fang, Z., Kogan, F., Wood, J., Stevens, K. J., Gibbons, E. K., Lee, J. H., Gold, G. E., & Hargreaves, B. A. (2018) · Magnetic Resonance in Medicine · doi:10.1002/mrm.27178 · Cited by 406 · see citing works
Radiation exposure from pediatric CT scans and subsequent cancer risk in The Netherlands
Meulepas, J. M., Ronckers, C. M., Smets, A. M. J. B., Nievelstein, R. A. J., Gradowska, P., Lee, C., Jahnen, A., van Straten, M., de Wit, M.-C. Y., Zonnenberg, B., et al. (2019) · JNCI Journal of the National Cancer Institute · doi:10.1093/jnci/djy104 · Cited by 360 · see citing works
Bluemke, D. A., Moy, L., Bredella, M. A., Ertl-Wagner, B. B., Fowler, K. J., Goh, V. J., Halpern, E. F., Hess, C. P., Schiebler, M. L., & Weiss, C. R. (2020) · Radiology · doi:10.1148/radiol.2019192515 · Cited by 343 · see citing works
Sekine, T., Delso, G., Zeimpekis, K. G., de Galiza Barbosa, F., ter Voert, E. E. G. W., Huellner, M., & Veit-Haibach, P. (2018) · Radiology · doi:10.1148/radiol.2017162305 · Cited by 79 · see citing works
Flagged first, then by citations
Research works printed but not linked: 34
- [1]Gallamini A Zwarthoed C Borra A Positron emission tomography (PET) in oncology Cancers 2014 6 1821 1889 10.3390/cancers6041821 25268160 PMC4276948 · DOI
- [2]Zasadny, K. R. & Wahl, R. L. Standardized uptake values of normal tissues at PET with 2-[fluorine-18]-fluoro-2-deoxy-D-glucose: variations with body weight and a method for correction. Radiology189, 847–950 (1993). 8234714
- [3]Strauss L Conti PS The applications of PET in clinical oncology J. Nucl. Med. 1991 32 623 650 2013803
- [4]Thie JA Understanding the standardized uptake value, its methods, and implications for usage J. Nucl. Med. 2004 45 1431 1434 15347707
- [5]Chen KT MR-assisted PET motion correction in simultaneous PET/MRI studies of dementia subjects J. Magn. Reson. Imaging 2018 48 1288 1296 10.1002/jmri.26000 29517819 PMC6129224 · DOI
- [6]Callahan J Binns D Dunn L Kron T Motion effects on SUV and lesion volume in 3D and 4D PET scanning Australas. Phys. Eng. Sci. Med. 2011 34 489 495 10.1007/s13246-011-0109-x 22081269 · DOI
- [8]Pearce MS Radiation exposure from CT scans in childhood and subsequent risk of leukaemia and brain tumours: a retrospective cohort study Lancet 2012 380 499 505 10.1016/S0140-6736(12)60815-0 22681860 PMC3418594 · DOI · Paper page
- [9]Roberts FO Radiation dose to PET technologists and strategies to lower occupational exposure J. Nucl. Med. Technol. 2005 33 44 47 15731021
- [10]Huang B Law MWM Khong PL Whole-body PET/CT scanning: estimation of radiation dose and cancer risk Radiology 2009 251 166 174 10.1148/radiol.2511081300 19251940 · DOI
- [11]Ardila D End-to-end lung cancer screening with three-dimensional deep learning on low-dose chest computed tomography Nat. Med. 2019 25 954 961 10.1038/s41591-019-0447-x 31110349 · DOI · Paper page
- [12]McKinney SM International evaluation of an AI system for breast cancer screening Nature 2020 577 89 94 10.1038/s41586-019-1799-6 31894144 · DOI · Paper page
- [13]Campanella G Clinical-grade computational pathology using weakly supervised deep learning on whole slide images Nat. Med. 2019 25 1301 1309 10.1038/s41591-019-0508-1 31308507 PMC7418463 · DOI · Paper page
- [14]Esteva A Dermatologist-level classification of skin cancer with deep neural networks Nature 2017 542 115 118 10.1038/nature21056 28117445 PMC8382232 · DOI · Paper page
- [15]Chen H Low-Dose CT with a residual encoder-decoder convolutional neural network IEEE Trans. Med. Imaging 2017 36 2524 2535 10.1109/TMI.2017.2715284 28622671 PMC5727581 · DOI
- [17]Kaplan S Zhu Y Full-dose PET image estimation from low-dose PET image using deep learning: a pilot study J. Digit. Imaging 2019 32 773 778 10.1007/s10278-018-0150-3 30402670 PMC6737135 · DOI
- [18]Kang J Prediction of standard-dose brain PET image by using MRI and low-dose brain [18F]FDG PET images Med. Phys. 2015 42 5301 5309 10.1118/1.4928400 26328979 PMC4545052 · DOI
- [19]Chen KT Ultra-low-dose 18F-florbetaben amyloid PET imaging using deep learning with multi-contrast MRI inputs Radiology 2019 290 649 656 10.1148/radiol.2018180940 30526350 PMC6394782 · DOI
- [20]Wang Y 3D conditional generative adversarial networks for high-quality PET image estimation at low dose Neuroimage 2018 174 550 562 10.1016/j.neuroimage.2018.03.045 29571715 PMC6410574 · DOI
- [21]Wang, Y. R. (Joyce) et al. Artificial intelligence enables whole-body positron emission tomography scans with minimalradiation exposure. Eur. J. Nucl. Med. Mol. Imaging Aug; 48:2771-2781. (2021). PMC8266729 33527176
- [22]Recht M. P. et al. Using Deep Learning to Accelerate Knee MRI at 3 T: Results of an Interchangeability Study. AJR Am J Roentgenol 215:1421-1429 (2020). PMC8209682 32755163
- [23]Chaudhari, A. S. et al. Diagnostic accuracy of quantitative multi-contrast 5-minute knee MRI using prospective artificial intelligence image quality enhancement title. Am. J. Roentgenol. 216, 1614–1625 (2021). PMC8862596 32755384
- [24]Akamatsu G Improvement in PET/CT image quality with a combination of point-spread function and time-of-flight in relation to reconstruction parameters J. Nucl. Med. 2012 53 1716 1722 10.2967/jnumed.112.103861 22952340 · DOI · Paper page
- [25]Neyshabur B Bhojanapalli S McAllester D Srebro N Exploring generalization in deep learning Adv. Neural Inf. Process. Syst. 2017 2017 5948 5957
- [27]Paschali, M., Conjeti, S., Navarro, F. & Navab, N. Generalizability vs. robustness: investigating medical imaging networks using adversarial examples. Med. Image Comput. Comput. Assist. Interv. 10.1007/978-3-030-00928-1_56 (2018). · DOI
- [29]Julious SA Sample sizes for clinical trials with Normal data Stat. Med. 2004 23 1921 1986 10.1002/sim.1783 15195324 · DOI · Paper page
- [31]Gatidis S Towards tracer dose reduction in PET studies: simulation of dose reduction by retrospective randomized undersampling of list-mode data Hell. J. Nucl. Med. 2016 19 15 18 26929936
- [32]Seith F Simulation of tracer dose reduction in 18F-FDG PET/MRI: effects on oncologic reading, image quality, and artifacts J. Nucl. Med. 2017 58 1699 1705 10.2967/jnumed.116.184440 28360205 · DOI
- [33]Schaefferkoetter JD Quantitative accuracy and lesion detectability of low-dose 18F-FDG PET for lung cancer screening J. Nucl. Med. 2017 58 399 405 10.2967/jnumed.116.177592 27688481 · DOI
- [34]Chaudhari AS Prospective deployment of deep learning in MRI: a framework for important considerations, challenges, and recommendations for best practices J. Magn. Reson. Imaging 2021 54 357 371 10.1002/jmri.27331 32830874 PMC8639049 · DOI
- [35]Chen, K. T. et al. True ultra-low-dose amyloid PET/MRI enhanced with deep learning for clinical interpretation. Eur. J. Nucl. Med. Mol. Imaging.48, 2416–2425. (2021). PMC8891344 33416955
- [37]Gholami S Salavati A Houshmand S Werner TJ Alavi A Assessment of atherosclerosis in large vessel walls: a comprehensive review of FDG-PET/CT image acquisition protocols and methods for uptake quantification J. Nucl. Cardiol. 2015 22 468 479 10.1007/s12350-015-0069-8 25827619 · DOI
- [39]Ahlman MA Internal tissue references for 18Fluorodeoxyglucose vascular inflammation imaging: Implications for cardiovascular risk stratification and clinical trials PLoS ONE 2017 12 1 12 10.1371/journal.pone.0187995 PMC5683610 29131857 · DOI
- [40]Paldino MJ Comparison of the diagnostic accuracy of PET/MRI to PET/CT-acquired FDG brain exams for seizure focus detection: a prospective study Pediatr. Radiol. 2017 47 1500 1507 10.1007/s00247-017-3888-8 28512714 · DOI
- [41]Nestle U Comparison of different methods for delineation of 18F-FDG PET-positive tissue for target volume definition in radiotherapy of patients with non-Small cell lung cancer J. Nucl. Med. 2005 46 1342 1348 16085592
Cited by
Sorted by citations
Deep learning-based PET image denoising and reconstruction: a review
Hashimoto, F., Onishi, Y., Ote, K., Tashima, H., Reader, A. J., & Yamaya, T. (2024) · Radiological Physics and Technology · doi:10.1007/s12194-024-00780-3 · Cited by 103 · see citing works
“Currently, deep learning-based PET image restoration technology has already been implemented in commercial PET scanners [13, 14, 147] with Food and Drug Administration-cleared commercially available software [148–152], making significant contributions in clinical practice.”
MethodPain, C. D., Egan, G. F., & Chen, Z. (2022) · European Journal of Nuclear Medicine and Molecular Imaging · doi:10.1007/s00259-022-05746-4 · Cited by 99 · see citing works
“The neural network used in this case was trained using data sourced independently from the three evaluation sites to provide an unbiased evaluation (Image from [100])”
BackgroundMethodInfluentialChallenges in AI-driven Biomedical Multimodal Data Fusion and Analysis
Liu, J., Cen, X., Yi, C., Wang, F.-A., Ding, J., Cheng, J., Wu, Q., Gai, B., Zhou, Y., He, R., et al. (2025) · Genomics, Proteomics & Bioinformatics · doi:10.1093/gpbjnl/qzaf011 · Cited by 75 · see citing works
Artificial Intelligence for PET and SPECT Image Enhancement
Balaji, V., Song, T.-A., Malekzadeh, M., Heidari, P., & Dutta, J. (2023) · Journal of Nuclear Medicine · doi:10.2967/jnumed.122.265000 · Cited by 69
“…enhancement is underscored by the availability of U.S. Food and Drug Administration – approved vendor-neutral commercial software such as SubtlePET (Subtle Medical) for AI-based denoising, as exempli fi ed in a study using SubtlePET ’ s CNN to enhance low-count scans to diagnostic quality ( 57 ).”
MethodWeyts, K., Lasnon, C., Ciappuccini, R., Lequesne, J., Corroyer-Dulmont, A., Quak, E., Clarisse, B., Roussel, L., Bardet, S., & Jaudet, C. (2022) · European Journal of Nuclear Medicine and Molecular Imaging · doi:10.1007/s00259-022-05800-1 · Cited by 59 · see citing works
Advances in PET/CT Technology: An Update
Aide, N., Lasnon, C., Desmonts, C., Armstrong, I. S., Walker, M. D., & McGowan, D. R. (2021) · Seminars in Nuclear Medicine · doi:10.1053/j.semnuclmed.2021.10.005 · Cited by 55 · see citing works
Pan, S.-Y., Abouei, E., Peng, J.-B., Qian, J., Wynne, J., Wang, T.-H., Chang, C.-W., Roper, J., Nye, J., Mao, H., et al. (2023) · Medical Physics (Lancaster) · doi:10.1002/mp.17068 · Cited by 51 · see citing works
“We varied the generation timestep for PET-CM using even spaced timesteps ranging from 1 to 150, with step sizes 1, 2 (proposed), 5, and 10.”
BackgroundMethodArtificial intelligence in oncologic imaging
Chen, M., Terzic, A., Becker, A. S., Johnson, J. M., Wu, C. C., Wintermark, M., Wald, C., & Wu, J. (2022) · European Journal of Radiology Open · doi:10.1016/j.ejro.2022.100441 · Cited by 41 · see citing works
“In another study, this group demonstrated that deep learning can enhance noisy PET images acquired with 4-times fewer counts for clinical purposes in a blinded, multicenter study [13].”
BackgroundSanaat, A., Akhavanalaf, A., Shiri, I., Salimi, Y., Arabi, H., & Zaidi, H. (2022) · Human Brain Mapping · doi:10.1002/hbm.26068 · Cited by 28
“Another study reported promising results for fast TOF and non-TOF PET scanning with 75% scanning time reduction (Chaudhari et al., 2021).”
BackgroundNg, C. K. C. (2023) · Children · doi:10.3390/children10081372 · Cited by 30 · see citing works
Artificial intelligence guided enhancement of digital PET: scans as fast as CT?
Hosch, R., Weber, M., Sraieb, M., Flaschel, N., Haubold, J., Kim, M. S., Umutlu, L., Kleesiek, J., Herrmann, K., Nensa, F., et al. (2022) · European Journal of Nuclear Medicine and Molecular Imaging · doi:10.1007/s00259-022-05901-x · Cited by 27 · see citing works
“Additionally, acquisition times were longer than the maximum scan velocity of recent PET/CT scanners by a factor of at least 10 [27, 28].”
BackgroundResultInfluentialNanoradiopharmaceuticals: Design Principles, Radiolabeling Strategies, and Biomedicine Applications
Núñez-Salinas, A., Parra-Garretón, C., Acuña, D., Peñaloza, S., Günther, G., Bollo, S., Arriagada, F., & Morales, J. O. (2025) · Pharmaceutics · doi:10.3390/pharmaceutics17070912 · Cited by 24 · see citing works
“Convolutional neural networks, generative adversarial networks, and diffusion models improve tumor detection and image quality in PET and SPECT scans, restoring diagnostic accuracy in low-dose scans with high sensitivity and specificity [153–155].”
Wang, Y., Wang, P., Adams, L. C., Sheybani, N., Qu, L.-Q., Sarrami, A., Theruvath, A., Gatidis, S., Ho, T., Zhou, Q., et al. (2023) · European Journal of Nuclear Medicine and Molecular Imaging · doi:10.1007/s00259-022-06097-w · Cited by 24 · see citing works
Positron Range Corrections and Denoising Techniques for Gallium-68 PET Imaging: A Literature Review
Gavriilidis, P., Koole, M., Annunziata, S., Mottaghy, F., & Wierts, R. (2022) · Diagnostics · doi:10.3390/diagnostics12102335 · Cited by 24 · see citing works
MRSaiFE: An AI-based Approach Towards the Real-Time Prediction of Specific Absorption Rate
Gokyar, S., Robb, F., Kainz, W., Chaudhari, A., & Winkler, S. (2021) · IEEE Access · doi:10.1109/access.2021.3118290 · Cited by 23 · see citing works
Self-citationRoya, M., Mostafapour, S., Mohr, P., Providência, L., Li, Z., van Snick, J. H., Brouwers, A., Noordzij, W., Willemsen, A., Dierckx, R., et al. (2023) · Cancers · doi:10.3390/cancers15215173 · Cited by 21 · see citing works
DRMC: A Generalist Model with Dynamic Routing for Multi-Center PET Image Synthesis
Yang, Z., Zhou, Y., Zhang, H., Wei, B., Fan, Y., & Xu, Y. (2023) · International Conference on Medical Image Computing and Computer-Assisted Intervention · doi:10.48550/arxiv.2307.05249 · Cited by 21
“Positron emission tomography (PET) image synthesis [1,2,3,4,5,6,7,8,9,10] aims at recovering high-quality full-dose PET images from low-dose ones.”
BackgroundInfluentialCosta, P. F., Jentzen, W., Brahmer, A., Mavroeidi, I. A., Zarrad, F., Umutlu, L., Fendler, W. P., Rischpler, C., Herrmann, K., Conti, M., et al. (2022) · BMC Cancer · doi:10.1186/s12885-022-09993-4 · Cited by 22 · see citing works
“Moreover, motion artifacts can be reduced [10].”
Evaluation of a High-Sensitivity Organ-Targeted PET Camera
Stiles, J., Baldassi, B., Bubon, O., Poladyan, H., Freitas, V., Scaranelo, A., Mulligan, A., Waterston, M., & Reznik, A. (2022) · Italian National Conference on Sensors · doi:10.3390/s22134678 · Cited by 19
Theruvath, A., Siedek, F., Yerneni, K., Muehe, A., Spunt, S., Pribnow, A., Moseley, M., Lu, Y., Zhao, Q., Gulaka, P., et al. (2021) · Radiology: Artificial Intelligence · doi:10.1148/ryai.2021200232 · Cited by 19 · see citing works
“This DL CNN had been previously validated for augmentation of brain scans in adult patients (24,25) and was cross-trained for augmentation of whole-body scans.”
MethodSelf-citation18F-FDG PET/CT for Response Assessment in Lung Cancer.
Marcus, C., Tajmir, S. H., Rowe, S. P., Sheikhbahaei, S., & Solnes, L. (2022) · Seminars in nuclear medicine · doi:10.1053/j.semnuclmed.2022.04.001 · Cited by 18 · see citing works
Jabbarpour, A., Ghassel, S., Lang, J., Leung, E., Gal, G. L., Klein, R., & Moulton, E. J. (2023) · Seminars in Nuclear Medicine · doi:10.1053/j.semnuclmed.2023.03.002 · Cited by 22 · see citing works
Seyyedi, N., Ghafari, A., Seyyedi, N., & Sheikhzadeh, P. (2024) · BMC Medical Imaging · doi:10.1186/s12880-024-01417-y · Cited by 16 · see citing works
“…review for medical imaging reconstruction and synthesis studies, this paper totally included thirteen CNN-based models ( [12–14, 21, 27, 31, 36, 42, 43, 51, 56, 58, 67]), fifteen UNET-based models ( [17, 19, 22, 26, 28, 37, 39, 40, 44, 45, 48, 55, 61, 62, 66]), twenty one GAN-based models ( [15,…”
MethodPET image enhancement using artificial intelligence for better characterization of epilepsy lesions
Flaus, A., Deddah, T., Reilhac, A., Leiris, N. D., Janier, M. F., Mérida, I., Grenier, T., McGinnity, C. J., Hammers, A., Lartizien, C., et al. (2022) · Frontiers in Medicine · doi:10.3389/fmed.2022.1042706 · Cited by 18 · see citing works
“scored the diagnostic image quality on a 5-point Likert scale (1 = non-diagnostic, 2 = poor, 3 = standard, 4 = good, 5 = excellent image quality) (52) and as previously, indicated if a hypometabolic lesion was present and scored ODC.”
BackgroundMethodThe Role of AI in Lymphoma: An Update.
Cairns, J., Frood, R., Patel, C., & Scarsbrook, A. (2025) · Seminars in nuclear medicine · doi:10.1053/j.semnuclmed.2025.02.007 · Cited by 15 · see citing works
Frood, R., Willaime, J. M. Y., Miles, B., Chambers, G., Al-Chalabi, H., Ali, T., Hougham, N., Brooks, N., Petrides, G., Naylor, M. J., et al. (2024) · Frontiers in Nuclear Medicine · doi:10.3389/fnume.2023.1327186 · Cited by 14
Yang, Z., Zhou, Y., Chen, H.-W., Zhang, H., Zhao, D., Wei, B., & Xu, Y. (2026) · Medical Image Anal. · doi:10.1016/j.media.2026.104059 · Cited by 13
“…images (Xiang et al., 2017; Xu et al., 2017; Chan et al., 2018; Wang et al., 2018, 2019; Schae ff erkoetter et al., 2020; Zhou et al., 2020b; Chaudhari et al., 2021; Luo et al., 2021; Zhou et al., 2022; Zeng et al., 2022; Luo et al., 2022; Yang et al., 2023; Fu et al., 2023; Jang et al.,…”
InfluentialApplications of Artificial Intelligence for Pediatric Cancer Imaging
Singh, S., Sarrami, A., Gatidis, S., Varniab, Z. S., Chaudhari, A. S., & Daldrup-Link, H. E. (2024) · AJR. American journal of roentgenology · doi:10.2214/ajr.24.31076 · Cited by 13 · see citing works
Self-citationIntelligent ultrafast total-body PET for sedation-free pediatric [18F]FDG imaging
Zhou, X., Fu, Y., Dong, S., Li, L., Xue, S., Chen, R.-H., Huang, G., Liu, J., & Shi, K.-Y. (2024) · European Journal of Nuclear Medicine and Molecular Imaging · doi:10.1007/s00259-024-06649-2 · Cited by 13
The Role of AI in the Evaluation of Neuroendocrine Tumors: Current State of the Art.
Lopez-Ramirez, F., Yasrab, M., Tixier, F., Kawamoto, S., Fishman, E. K., & Chu, L. C. (2025) · Seminars in nuclear medicine · doi:10.1053/j.semnuclmed.2025.02.003 · Cited by 11 · see citing works
Margail, C., Merlin, C., Billoux, T., Wallaert, M., Otman, H., van Sas, N. C. F., Molnar, I., Guillemin, F., Boyer, L., Guy, L., et al. (2023) · EJNMMI Research · doi:10.1186/s13550-023-00999-y · Cited by 12 · see citing works
“It has not been trained in our center [3] but has been trained on millions of images through dataset augmentation [4].”
MethodMolecular Imaging with PET-CT and PET-MRI in Pediatric Musculoskeletal Diseases.
Guja, K., Behr, G., Bedmutha, A., Kuhn, M., Nadel, H. R., & Pandit-Taskar, N. (2024) · Seminars in nuclear medicine · doi:10.1053/j.semnuclmed.2024.03.003 · Cited by 10
Quak, E., Weyts, K., Jaudet, C., Prigent, A., Foucras, G., & Lasnon, C. (2023) · Frontiers in Medicine · doi:10.3389/fmed.2023.1137514 · Cited by 10 · see citing works
“Furthermore, it is always preferable for the patient’s sake to decrease rather than increase the injected dose (8, 9).”
BackgroundStaging of prostate Cancer with ultra-fast PSMA-PET scans enhanced by AI
Kersting, D., Borys, K., Küper, A. T., Kim, M., Haubold, J., Goerttler, T., Umutlu, L., Costa, P. F., Kleesiek, J., Rischpler, C., et al. (2025) · European Journal of Nuclear Medicine and Molecular Imaging · doi:10.1007/s00259-024-07060-7 · Cited by 11 · see citing works
Tang, C., Eisenmenger, L., Rivera-Rivera, L., Huo, E., Junn, J., Kuner, A., Oechtering, T., Peret, A., Starekova, J., & Johnson, K. M. (2024) · Journal of Magnetic Resonance Imaging · doi:10.1002/jmri.29672 · Cited by 8 · see citing works
Dutta, K., Laforest, R., Luo, J.-Q., Jha, A. K., & Shoghi, K. (2024) · Medical Physics (Lancaster) · doi:10.1002/mp.17105 · Cited by 7 · see citing works
“10,61 Although the total image is important for diagnosis, segmentation greatly augments clinical decisions by obtaining clinically relevant boundaries and volumetric assessment for treatment planning.”
Summa, M. D., Ruggiero, M. R., Spinosa, S., Iachetti, G., Esposito, S., Annunziata, S., & Pizzuto, D. A. (2024) · Clinical and Translational Imaging · doi:10.1007/s40336-024-00625-4 · Cited by 8 · see citing works
“The relative differences between full dose and low-count scans are consistent across all scanners, demonstrating that AI can always improve low-count images regardless of scanner hardware and software [26].”
Chen, K.-T., Tesfay, R., Koran, M. E., Ouyang, J., Shams, S., Young, C. B., Davidzon, G., Liang, T., Khalighi, M., Mormino, E., et al. (2023) · American Journal of Neuroradiology · doi:10.3174/ajnr.a7961 · Cited by 7 · see citing works
Self-citationHu, Y.-Y., Lv, D., Jian, S., Lang, L., Cui, C.-Y., Liang, M., Song, L., Li, S.-J., & Wu, Z.-F. (2023) · Quantitative Imaging in Medicine and Surgery · doi:10.21037/qims-22-1181 · Cited by 7
“There are also studies (13,33-35) using fewer doses of PET images, and CNN also shows an advantage in tumor quantification.”
BackgroundThe requirements for performing artificial-intelligence-related research and model development
Pareek, A., Lungren, M. P., & Halabi, S. S. (2022) · Pediatric Radiology · doi:10.1007/s00247-022-05483-8 · Cited by 6 · see citing works
Pan, S.-Y., Abouei, E., Peng, J.-B., Qian, J., Wynne, J. F., Wang, T.-H., Chang, C.-W., Roper, J., Nye, J. A., Mao, H., et al. (2023) · arXiv.org · doi:10.48550/arxiv.2308.13072 · Cited by 7
“We varied the generation timestep for PET-CM using even spaced timesteps ranging from 1 to 150, with step sizes 1, 2 (proposed), 5, and 10.”
BackgroundMethodMaes, J., Carron, C., DeKeyser, S., Brants, T., Ridder, V. D., Chaouki, A., Steenhout, C., Vandenberghe, S., D’Asseler, Y., Raes, L., et al. (2025) · European Journal of Nuclear Medicine and Molecular Imaging · doi:10.1007/s00259-025-07672-7 · Cited by 6 · see citing works
“In their study, Chaudhari et al., claimed a 75% reduction in measured counts, but this was in Ameri - can centers using clinical protocols that result in a signifi - cantly higher number of counts for the clinical scans [28].”
InfluentialNuclear medicine technologists practice impacted by AI denoising applications in PET/CT images
Champendal, M., Ribeiro, R. T., Müller, H., Prior, J. O., & dos Reis, C. S. (2024) · Radiography · doi:10.1016/j.radi.2024.06.010 · Cited by 9 · see citing works
Deep denoiser prior driven relaxed iterated Tikhonov method for low-count PET image restoration
Chang, W., D'Ascenzo, N., Antonecchia, E., Li, B.-X., Yang, J., Mu, D., Li, A., & Xie, Q.-G. (2024) · Physics in Medicine and Biology · doi:10.1088/1361-6560/ad67a3 · Cited by 5 · see citing works
“However, more widespread use of PET is limited by its high radiation exposure and low patient throughput (Chaudhari et al 2021).”
Qualitative analysis of PD-L1 expression in non-small-cell lung cancer based on chest CT radiomics
Fu, Y., Zhang, H., Xue, P., Ren, M.-R., Xiao, T.-H., Zhang, Z., Huang, Y., & Dong, E.-Q. (2023) · Biomedical Signal Processing and Control · doi:10.1016/j.bspc.2023.104815 · Cited by 5 · see citing works
Möglichkeiten der künstlichen Intelligenz im Strahlenschutz
Pashazadeh, A., & Hoeschen, C. (2023) · Die Radiologie · doi:10.1007/s00117-023-01167-y · Cited by 5
Weyts, K., Quak, E., Licaj, I., Ciappuccini, R., Lasnon, C., Corroyer-Dulmont, A., Foucras, G., Bardet, S., & Jaudet, C. (2023) · Diagnostics · doi:10.3390/diagnostics13091626 · Cited by 5 · see citing works
Efficient Parameter Adaptation for Multi-Modal Medical Image Segmentation and Prognosis
Saeed, N., Hardan, S., Ridzuan, M., Saadi, N., Nandakumar, K., & Yaqub, M. (2025) · arXiv.org · doi:10.48550/arxiv.2504.13645 · Cited by 4
“Another reason is that PET scans are not always accessible in every healthcare facility, leading to missing PET images in resource-limited settings [10].”
Braune, A., Hosch, R., Kersting, D., Müller, J., Hofheinz, F., Herrmann, K., Nensa, F., Kotzerke, J., & Seifert, R. (2025) · EJNMMI Physics · doi:10.1186/s40658-025-00745-4 · Cited by 4 · see citing works
“The integration of deep-learning neural network algorithms can help to enhance image quality and especially reduce image noise [2, 3], such that image quality similar to PET scans acquired according to standards in clinical routine (2–3 min per bed position) can be restored in PET scans acquired…”
Pang, L.-F., Zhang, Z., Liu, G.-B., Hu, P.-C., Chen, S.-G., Gu, Y., Huang, Y.-K., Zhang, J., Shi, Y., Cao, T., et al. (2024) · Molecular Imaging and Biology · doi:10.1007/s11307-024-01943-9 · Cited by 4
“In contrast, Chaudhari's CNN network exhibits the ability to generalize towards an external validation set comprised of clinical patients from various institutions and scanner types [12].”
Can AI Make Paediatric PET/CT Lower Dose? Opportunities, Evidence and Cautions.
Currie, G., & Rohren, E. (2026) · Seminars in nuclear medicine · doi:10.1053/j.semnuclmed.2026.07.002 · Cited by 3
Artificial Intelligence and Digital Technology in Cardiovascular Imaging: A Narrative Review
Papadopoulos, C. H., Karelas, D., Floropoulou, C., Tzavida, K., Oikonomidis, D. L., Tasoulis, A., Tatsis, E., Kouloulias, I., & Kadoglou, N. P. E. (2026) · BioTech · doi:10.3390/biotech15010022 · Cited by 1 · see citing works
“Nuclear cardiology: DL-based attenuation correction restores diagnostic performance of myocardial perfusion SPECT without CT, and DL denoising allows PET and SPECT imaging at lower counts or doses while preserving lesion detectability and quantitative accuracy [19–21].”
User-centric eXplainable AI criteria for implementing AI-based denoising in PET/CT
Champendal, M., Ribeiro, R. T., Müller, H., Prior, J. O., & dos Reis, C. S. (2025) · Radiography · doi:10.1016/j.radi.2025.103194 · Cited by 3 · see citing works
Clinical validation of a deep learning model for low-count PET image enhancement
Long, Q., Tian, Y., Pan, B.-Y., Xu, Z., Zhang, W., Xu, L.-B., Fan, W., Pan, T., & Gong, N. (2025) · European Journal of Nuclear Medicine and Molecular Imaging · doi:10.1007/s00259-025-07370-4 · Cited by 3 · see citing works
Yan, L., Wang, Z., Li, D.-C., Wang, Y.-Y., Yang, G.-J., Zhao, Y.-J., Kong, Y., Wang, R., Wu, R., & Wang, Z. (2024) · Quantitative Imaging in Medicine and Surgery · doi:10.21037/qims-23-817 · Cited by 3
“Therefore, the performance is less optimized to a specific PET model but more generalized to multiple centers and/or scanners, as has been demonstrated in a multicenter and multivendor study (26).”
Jo, B., & Lee, S.-J. (2023) · Nuclear Engineering and Technology · doi:10.1016/j.net.2023.12.028 · Cited by 3
Tsougos, I., Valotassiou, V. J., Tsivaka, D., Σάτρα, Μ., Angelidis, G., Papatriantafyllou, J. D., Panagiotidis, E., Dardiotis, E., Hadjigeorgiou, G. M., & Georgoulias, P. (2026) · Annals of Nuclear Medicine · doi:10.1007/s12149-026-02194-0 · Cited by 2 · see citing works
Accelerating imaging: deep learning for enhanced 123I-ioflupane SPECT efficiency
Ishiwata, Y., Horie, K., Aritome, K., Aoki, R., Iizuka, H., Aso, S., Takeuchi, Y., Misumi, Y., Haga, A., Kato, S., et al. (2025) · Japanese Journal of Radiology · doi:10.1007/s11604-025-01933-z · Cited by 2
“[ 20] achieved a four-fold count reduction using a deep learning enhancement model without compromising diagnostic quality [ 20].”
Search for a new material for a medical Cherenkov radiation detector
Savchenko, A., & Tishchenko, A. (2023) · Nuclear Instruments and Methods in Physics Research Section A : Accelerators, Spectrometers, Detectors and Associated Equipment · doi:10.1016/j.nima.2023.169021 · Cited by 2
Immuno-imaging in solid organ transplantation
Maas, R. J. F., Versteeg, I., de Jong, L. C., Fayad, Z., Aarntzen, E., Mulder, W. J. M., Teunissen, A., & Duivenvoorden, R. (2026) · American Journal of Transplantation · doi:10.1016/j.ajt.2026.02.037 · Cited by 1
VQPET: Leveraging Vector-Quantized Codebook Prior for PET Image Synthesis
Chen, H.-W., Yang, Z., Zhou, Y., Zhang, X.-Q., Zhang, H., Zhao, D., Wei, B., Zhou, G., & Xu, Y. (2026) · IEEE Transactions on Medical Imaging · doi:10.1109/tmi.2026.3657118 · Cited by 1 · see citing works
“Recent studies [2], [3], [4], [5], [6], [7], [8], [9], [10], [11], [12] primarily rely on deep learning techniques to learn a mapping function that transforms LQ images into HQ images.”
Image SNR Enhancement for a Short Axial FOV Brain PET System Using Generative Deep Learning
Nazari-Farsani, S., Jafaritadi, M., Fisher, J., Chin, M., Chinn, G., Khalighi, M., Zaharchuk, G., & Levin, C. S. (2026) · IEEE Transactions on Radiation and Plasma Medical Sciences · doi:10.1109/trpms.2025.3560667 · Cited by 1 · see citing works
“In recent years, DL-based image denoising and deblurring have become increasingly popular for the post-reconstruction enhancement of low-dose/short-acquisition brain PET imaging [14], [15], [16], [17].”
Self-citationKim, S., Park, J. H., Kim, C.-H., You, S., Choi, J.-S., Chang, J. W., Jo, I. Y., Lee, B.-J., Park, I.-S., Kim, H. S., et al. (2026) · Cancers · doi:10.3390/cancers18020210 · Cited by 1
“Therefore, evaluating the generalizability of models across diverse datasets is essential [29].”
Chen, H.-Z., Ding, H.-X., Jiang, Y.-F., Lan, J., Li, K.-C., Cheng, G. W. Y., Ng, N.-C., Pu, Y., Cai, J., Lin, L.-T., et al. (2025) · IEEE International Conference on Bioinformatics and Biomedicine · doi:10.1109/bibm66473.2025.11357187 · Cited by 1 · see citing works
“PET itself is also subject to variations in count statistics, which further complicate fusion [6].”
Zhao, L., Li, L.-E., Zhang, T.-T., He, Q., Yan, J., Yang, X., Wang, K., Yang, M., Xu, L., Hu, K., et al. (2026) · The Innovation Drug Discovery · doi:10.59717/j.xinn-drugdisc.2026.100030 · Cited by 1 · see citing works
Self-citationADVANCING RADIATION PROTECTION IN PET/CT.
Kurth, J. (2026) · Annals of the ICRP · doi:10.1177/01466453251385014
Muneer, A., Saad, M. B., Xu, H., & Wu, J. (2026) · Frontiers in Nuclear Medicine · doi:10.3389/fnume.2026.1956567
“Prospective, multivendor reader evidence remains much less common than retrospective technical validation (11, 18).”
FDG-PET in Lymphoma: The Deauville Era and Beyond.
Thakur, P. B., & Basu, S. (2026) · PET Clinics · doi:10.1016/j.cpet.2026.06.006
Artificial Intelligence Across the PET Reconstruction Pipeline: An Update
Farhadi, F., Rajagopal, J. R., Bolet, S. I., Dutta, J., Catalano, O., Iravani, A., Esfahani, S., Jadvar, H., & Heidari, P. (2026) · AJR. American journal of roentgenology · doi:10.2214/ajr.26.34681
Self-citationThe Role of Artificial Intelligence in PET Imaging.
Veziroglu, E. M., Dayma, K., Hasani, N., Kulkarni, R., Farhadi, F., & Klyuzhin, I. (2026) · The Radiologic clinics of North America · doi:10.1016/j.rcl.2026.07.010
Wang, Y., Cheng, C., Huang, B., & Zuo, C. (2026) · Japanese Journal of Radiology · doi:10.1007/s11604-026-02066-7
Yar, G. N. A. H., Peiris, H., Jamadar, S., Fornito, A., & Chen, Z.-L. (2026) · doi:10.48550/arxiv.2608.11514
“Inconsistent performance becomes more pronounced under real-world clinical variability [6].”
Recent advances in nuclear medicine for cancer diagnosis and targeted therapy
Oisakede, E. O., Atitebi, O. F., Bamidele, D. B., Matope, E. M., Alabi, J. O., & Olawade, D. B. (2026) · Cancer Bioscience · doi:10.1016/j.cbs.2026.100006
Hewitt, P., Kim, H., & Vida, T. A. (2026) · Frontiers in Aging Neuroscience · doi:10.3389/fnagi.2026.1748671
“PET offers high molecular specificity with millimeter-level resolution but is limited by radiation exposure, high cost, and tracer variability (Chaudhari et al., 2021; Duclos et al., 2021; Akamatsu et al., 2023).”
Alpizar-Rojas, M., Jaromirski-Segura, H., Herrera-Jiménez, L., Castro-Gamboa, J., Mora-Jiménez, J., Arguedas-Chácon, S., & Zavaleta-Monestel, E. (2026) · Cureus · doi:10.7759/cureus.113457
“Quantitative outputs may vary according to scanner type, acquisition settings, reconstruction parameters, and model training data [2,3,6,11,12,23].”
InfluentialDzramado, V. L., Lasim, O. U., & Owusu-Aboagye, D. (2026) · Frontiers in Digital Health · doi:10.3389/fdgth.2026.1834800
“…the primary search strategy, driven by subscription access constraints, may reduce retrieval sensitivity for grey literature and non-African-indexed studies; this is partially mitigated by the extensive African-specific database coverage and grey literature searches described in Table 1 (37, 39).”
Hadjitheodorou, P., Sotiriou, M., Kyrou, K., Adamou, G., Eleftheriou, M., Konstantinidou, K., & Vrachimis, A. (2026) · Zeitschrift für Medizinische Physik · doi:10.1016/j.zemedi.2026.06.002
Pair production tomography enables imaging of MeV-scale gamma-emitting theranostic radionuclides
Das, B., Peter, R., & Seo, Y. (2026) · Research Square · doi:10.21203/rs.3.rs-9349692/v1
Long, Q., Tian, Y., Hu, Y., Xu, Z., Zhang, W., Xu, S.-S., Liu, W., Jin, J., & Peng, Y.-S. (2026) · EJNMMI Physics · doi:10.1186/s40658-026-00841-z
“The NI margin for overall image quality (Δ= −0.5 on the 5-point Likert scale) was prespecified to represent no more than a half-category (half-grade) decrease, which is commonly adopted in blinded-reader NI evaluations of low-count PET deep-learning approaches [25, 53].”
InfluentialMaintaining SUV accuracy in low-count PET with PETfectior: a deep learning denoising solution
Habarnau, Y. R., Bustos, N., Corona, P., Gonz'alez, C., Traverso, S., Matorra, F., Funes, F., Giraut, J. M., Pelegrina, L., Bruno, G., et al. (2025) · Biomedical engineering and physics express · doi:10.1088/2057-1976/ae826d
“It was conducted in accordance with the Declaration of Helsinki and informed consent was obtained from all patients or their legal guardians.”
Zhang, Y., Meng, H.-P., Yu, T.-E., Xiang, L., Guo, M.-Z., Li, B., Zhou, T., & Zhang, M. (2025) · Medical Physics (Lancaster) · doi:10.1002/mp.70198
“22,34,35 The advantages of this approach are multifaceted.”
Silva, L. C., Constantino, C., Teixeira, R., Castanheira, J., Oliveira, F. P. M., & Costa, D. (2025) · Journal of Imaging Informatics in Medicine · doi:10.1007/s10278-025-01638-9
Hardware-Compatible U-Net for Low-Dose PET Reconstruction
Dao, E.-K., Zukotynski, K., Black, S. E., & Gaudet, V. (2025) · IEEE International Symposium on Multiple-Valued Logic (ISMVL) · doi:10.1109/ismvl64713.2025.00019
“Efforts to mitigate increased noise associated with dose reduction have historically focused on algorithmic approaches, such as iterative reconstruction [3], advanced filtering techniques [4], and deep-learning models [5].”
Ultra-low dose immunoPET using 64Cu-rituximab tracer for a human CD20 mouse model
Habte, F., & Natarajan, A. (2025) · Frontiers in Medicine · doi:10.3389/fmed.2025.1548132
Ring-array Ultrasound CT Reflection Imaging Applied to Breast Tumor Detection
Shi, L., Huang, X., Yuan, X., & Zhang, W.-D. (2024) · Journal of Physics, Conference Series · doi:10.1088/1742-6596/2822/1/012018
Algorithm for Reducing Overall Biological Detriment Caused by PET/CT: an Age-Based Study
Spadafora, M., Sannino, P., Mansi, L., Mainolfi, C., Capasso, R., Giorgio, E. D., Fiordoro, S., Imbimbo, S., Masone, F., & Evangelista, L. (2023) · Nuclear medicine and molecular imaging · doi:10.1007/s13139-023-00788-4
“To overcome these issues, an artificial intelligence algorithm may be able to significantly reduce PET imaging time and radiation doses while maintaining image quality and SUV accuracy [30].”
BackgroundDigitale Positronenemissionstomografie – Entwicklung, Detektortechnologie und Perspektiven
Kazek, S. L., Jentzen, W., Seifert, R., Herrmann, K., & Kersting, D. (2022) · Angewandte Nuklearmedizin · doi:10.1055/a-1715-5184
Wang, Y., Qu, L.-Q., Sheybani, N., Luo, X., Wang, J., Hawk, K., Theruvath, A., Gatidis, S., Xiao, X., Pribnow, A., et al. (2022) · arXiv.org · doi:10.48550/arxiv.2205.04044
“Both actions, however, lower detection of PET annihilation events, which degrade the diagnostic image quality - since image quality is proportional to the number of coincidence events in the PET detector due to radiopharmaceutical positron annihilation 12 .”
BackgroundMethodInfluential
- Authors
- Chaudhari, A. S., Mittra, E., Davidzon, G. A., Gulaka, P., Gandhi, H., Brown, A., Zhang, T., Srinivas, S., Gong, E., Zaharchuk, G., & Jadvar, H.
- Affiliations
- Stanford University (US)
- Oregon Health & Science University (US)
- University of Pittsburgh Medical Center (US)
- University of Southern California (US)
- Corresponding author
- Chaudhari, Akshay S.
- ORCID
- Venue
- npj Digital Medicine
- Published
- 2021-08-23
- Volume
- 4
- Issue
- 1
- Pages
- 127-127
- ISSN
- 2398-6352 (ISSN-L)
- PMC
- PMC8382711
- OpenAlex
- W3194623075
- Cited by
- 83
- References
- 44
- Citations by year
- 2026: 15 · 2025: 17 · 2024: 16 · 2023: 20 · 2022: 12 · 2021: 3
- FWCI
- 6.74
- Type
- journal article
- Language
- en
- Update notices
- correction (2021-09-14)
- Open access
- gold
- doi.org
- Locations
- npj Digital Medicine (publishedVersion, cc-by)
- Europe PMC (PubMed Central) (publishedVersion, other-oa)
- PubMed Central (publishedVersion, cc-by)
- PubMed (publishedVersion)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- Topics
- Health Sciences › Medicine › Radiology, Nuclear Medicine and Imaging › Medical Imaging Techniques and Applications
- Health Sciences › Medicine › Health Informatics › Artificial Intelligence in Healthcare and Education
- Health Sciences › Medicine › Radiology, Nuclear Medicine and Imaging › Radiomics and Machine Learning in Medical Imaging
- Keywords
- low-count PET, whole-body PET, deep learning reconstruction, image enhancement, diagnostic quality, SUV quantification, lesion detection, radiation dose reduction, scan time reduction, interobserver agreement, tumor imaging
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
