Deep transformer-based personalized dosimetry from SPECT/CT images: a hybrid approach for [
Deep learning
Monte Carlo simulation
Radiation dosimetry
Radionuclide therapy
[177Lu]Lu-DOTATATE
Journal
European journal of nuclear medicine and molecular imaging
ISSN: 1619-7089
Titre abrégé: Eur J Nucl Med Mol Imaging
Pays: Germany
ID NLM: 101140988
Informations de publication
Date de publication:
25 Jan 2024
25 Jan 2024
Historique:
received:
13
11
2023
accepted:
15
01
2024
medline:
25
1
2024
pubmed:
25
1
2024
entrez:
24
1
2024
Statut:
aheadofprint
Résumé
Accurate dosimetry is critical for ensuring the safety and efficacy of radiopharmaceutical therapies. In current clinical dosimetry practice, MIRD formalisms are widely employed. However, with the rapid advancement of deep learning (DL) algorithms, there has been an increasing interest in leveraging the calculation speed and automation capabilities for different tasks. We aimed to develop a hybrid transformer-based deep learning (DL) model that incorporates a multiple voxel S-value (MSV) approach for voxel-level dosimetry in [ We used a dataset consisting of 22 patients undergoing up to 4 cycles of [ The DL approach showed slightly better performance (voxel relative absolute error (RAE) = 5.28 ± 1.32) compared to MSV (voxel RAE = 5.54 ± 1.4) and outperformed SSV (voxel RAE = 7.8 ± 3.02). Gamma analysis pass rates were 99.0 ± 1.2%, 98.8 ± 1.3%, and 98.7 ± 1.52% for DL, MSV, and SSV approaches, respectively. The computational time for MC was the highest (~2 days for a single-bed SPECT study) compared to MSV, SSV, and DL, whereas the DL-based approach outperformed the other approaches in terms of time efficiency (3 s for a single-bed SPECT). Organ-wise analysis showed absolute percent errors of 1.44 ± 3.05%, 1.18 ± 2.65%, and 1.15 ± 2.5% for SSV, MSV, and DL approaches, respectively, in lesion-absorbed doses. A hybrid transformer-based deep learning model was developed for fast and accurate dose map generation, outperforming the MIRD approaches, specifically in heterogenous regions. The model achieved accuracy close to MC gold standard and has potential for clinical implementation for use on large-scale datasets.
Identifiants
pubmed: 38267686
doi: 10.1007/s00259-024-06618-9
pii: 10.1007/s00259-024-06618-9
doi:
Types de publication
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Subventions
Organisme : H2020 Euratom
ID : Sinfonia project under grant agreement No 945196
Informations de copyright
© 2024. The Author(s).
Références
Sgouros G, Bodei L, McDevitt MR, Nedrow JR. Radiopharmaceutical therapy in cancer: clinical advances and challenges. Nat Rev Drug Discov. 2020;19:589–608. https://doi.org/10.1038/s41573-020-0073-9 .
doi: 10.1038/s41573-020-0073-9
pubmed: 32728208
pmcid: 7390460
Ersahin D, Doddamane I, Cheng D. Targeted radionuclide therapy. Cancers. 2011;3:3838–55. https://doi.org/10.3390/cancers3043838 .
doi: 10.3390/cancers3043838
pubmed: 24213114
pmcid: 3763399
Kwekkeboom DJ, de Herder WW, Kam BL, van Eijck CH, van Essen M, Kooij PP, et al. Treatment with the radiolabeled somatostatin analog [177Lu-DOTA0, Tyr3] octreotate: toxicity, efficacy, and survival. J Clin Oncol. 2008;26:2124–30.
doi: 10.1200/JCO.2007.15.2553
pubmed: 18445841
Strosberg J, El-Haddad G, Wolin E, Hendifar A, Yao J, Chasen B, et al. Phase 3 trial of (177)Lu-dotatate for midgut neuroendocrine tumors. N Engl J Med. 2017;376:125–35. https://doi.org/10.1056/NEJMoa1607427 .
doi: 10.1056/NEJMoa1607427
pubmed: 28076709
pmcid: 5895095
Zaknun JJ, Bodei L, Mueller-Brand J, Pavel M, Baum RP, Hörsch D, et al. The joint IAEA, EANM, and SNMMI practical guidance on peptide receptor radionuclide therapy (PRRNT) in neuroendocrine tumours. Eur J Nucl Med Mol Imag. 2013;40:800–16.
doi: 10.1007/s00259-012-2330-6
Sundlov A, Gleisner KS, Tennvall J, Ljungberg M, Warfvinge CF, Holgersson K, et al. Phase II trial demonstrates the efficacy and safety of individualized, dosimetry-based (177)Lu-DOTATATE treatment of NET patients. Eur J Nucl Med Mol Imaging. 2022;49:3830–40. https://doi.org/10.1007/s00259-022-05786-w .
doi: 10.1007/s00259-022-05786-w
pubmed: 35451612
pmcid: 9399027
Brosch-Lenz J, Yousefirizi F, Zukotynski K, Beauregard JM, Gaudet V, Saboury B, et al. Role of artificial intelligence in theranostics: toward routine personalized radiopharmaceutical therapies. PET Clin. 2021;16:627–41. https://doi.org/10.1016/j.cpet.2021.06.002 .
doi: 10.1016/j.cpet.2021.06.002
pubmed: 34537133
Dickson J, Ross J, Vöö S. Quantitative SPECT: the time is now. EJNMMI Physics. 2019;6:1–7.
doi: 10.1186/s40658-019-0241-3
Brady SL, Shulkin BL. Analysis of quantitative [I-123] mIBG SPECT/CT in a phantom and in patients with neuroblastoma. EJNMMI Physics. 2019;6:1–14.
doi: 10.1186/s40658-019-0267-6
Strigari L, Konijnenberg M, Chiesa C, Bardies M, Du Y, Gleisner KS, et al. The evidence base for the use of internal dosimetry in the clinical practice of molecular radiotherapy. Eur J Nucl Med Mol Imaging. 2014;41:1976–88. https://doi.org/10.1007/s00259-014-2824-5 .
doi: 10.1007/s00259-014-2824-5
pubmed: 24915892
Piwowarska-Bilska H, Kurkowska S, Birkenfeld B. Individualization of radionuclide therapies: challenges and prospects. Cancers. 2022;14:3418.
doi: 10.3390/cancers14143418
pubmed: 35884478
pmcid: 9316481
Chiesa C, Bardiès M, Zaidi H. Voxel-based dosimetry is superior to mean-absorbed dose approach for establishing dose-effect relationship in targeted radionuclide therapy. Med Phys. 2019;46:5403–6.
doi: 10.1002/mp.13851
pubmed: 31584697
Uribe C, Peterson A, Van B, Fedrigo R, Carlson J, Sunderland J, et al. An international study of factors affecting variability of dosimetry calculations, part 1: design and early results of the SNMMI dosimetry challenge. J Nucl Med. 2021;62:36S-47S.
doi: 10.2967/jnumed.121.262748
pubmed: 34857620
Bolch WE, Eckerman KF, Sgouros G, Thomas SR. MIRD pamphlet no. 21: a generalized schema for radiopharmaceutical dosimetry—standardization of nomenclature. J Nucl Med. 2009;50:477–84.
doi: 10.2967/jnumed.108.056036
pubmed: 19258258
Akhavanallaf A, Shiri I, Arabi H, Zaidi H. Whole-body voxel-based internal dosimetry using deep learning. Eur J Nucl Med Mol Imaging. 2021;48:670–82. https://doi.org/10.1007/s00259-020-05013-4 .
doi: 10.1007/s00259-020-05013-4
pubmed: 32875430
Furhang EE, Chui CS, Sgouros G. A Monte Carlo approach to patient-specific dosimetry. Med Phys. 1996;23:1523–9.
doi: 10.1118/1.597882
pubmed: 8892249
Gosewisch A, Ilhan H, Tattenberg S, Mairani A, Parodi K, Brosch J, et al. 3D Monte Carlo bone marrow dosimetry for Lu-177-PSMA therapy with guidance of non-invasive 3D localization of active bone marrow via Tc-99m-anti-granulocyte antibody SPECT/CT. EJNMMI Research. 2019;9:1–14.
doi: 10.1186/s13550-019-0548-z
Goetz TI, Lang E, Prante O, Maier A, Cordes M, Kuwert T, et al. Three-dimensional Monte Carlo-based voxel-wise tumor dosimetry in patients with neuroendocrine tumors who underwent 177 Lu-DOTATOC therapy. Annals Nucl Med. 2020;34:244–53.
doi: 10.1007/s12149-020-01440-3
Huizing DMV, de Wit-van der Veen B, Verheij M, Stokkel MPM. Dosimetry methods and clinical applications in peptide receptor radionuclide therapy for neuroendocrine tumours: a literature review. EJNMMI Research. 2018;8:1-11.
Berger MJ. Distribution of absorbed dose around point sources of electrons and beta particles in water and other media. Washington, DC: National Bureau of Standards; 1971.
Bolch WE, Bouchet LG, Robertson JS, Wessels BW, Siegel JA, Howell RW, et al. MIRD pamphlet no. 17: the dosimetry of nonuniform activity distributions—radionuclide S values at the voxel level. J Nucl Med. 1999;40:11S-36S.
pubmed: 9935083
Lee MS, Kim JH, Paeng JC, Kang KW, Jeong JM, Lee DS, et al. Whole-body voxel-based personalized dosimetry: the multiple voxel S-value approach for heterogeneous media with nonuniform activity distributions. J Nucl Med. 2018;59:1133–9.
doi: 10.2967/jnumed.117.201095
pubmed: 29242397
Arabi H, AkhavanAllaf A, Sanaat A, Shiri I, Zaidi H. The promise of artificial intelligence and deep learning in PET and SPECT imaging. Phys Med. 2021;83:122–37. https://doi.org/10.1016/j.ejmp.2021.03.008 .
doi: 10.1016/j.ejmp.2021.03.008
pubmed: 33765602
Visvikis D, Lambin P, BeuschauMauridsen K, Hustinx R, Lassmann M, Rischpler C, et al. Application of artificial intelligence in nuclear medicine and molecular imaging: a review of current status and future perspectives for clinical translation. Eur J Nucl Med Mol Imaging. 2022;49:4452–63. https://doi.org/10.1007/s00259-022-05891-w .
doi: 10.1007/s00259-022-05891-w
pubmed: 35809090
pmcid: 9606092
Hadjiiski L, Cha K, Chan HP, Drukker K, Morra L, Nappi JJ, et al. AAPM task group report 273: recommendations on best practices for AI and machine learning for computer-aided diagnosis in medical imaging. Med Phys. 2023;50:e1–24. https://doi.org/10.1002/mp.16188 .
doi: 10.1002/mp.16188
pubmed: 36565447
van der Velden BHM, Kuijf HJ, Gilhuijs KGA, Viergever MA. Explainable artificial intelligence (XAI) in deep learning-based medical image analysis. Med Image Anal. 2022;79:102470. https://doi.org/10.1016/j.media.2022.102470 .
doi: 10.1016/j.media.2022.102470
pubmed: 35576821
Matsubara K, Ibaraki M, Nemoto M, Watabe H, Kimura Y. A review on AI in PET imaging. Ann Nucl Med. 2022;36:133–43. https://doi.org/10.1007/s12149-021-01710-8 .
doi: 10.1007/s12149-021-01710-8
pubmed: 35029818
Zaidi H, El Naqa I. Quantitative molecular positron emission tomography imaging using advanced deep learning techniques. Annu Rev Biomed Eng. 2021;23:249–76. https://doi.org/10.1146/annurev-bioeng-082420-020343 .
doi: 10.1146/annurev-bioeng-082420-020343
pubmed: 33797938
Brosch-Lenz JF, Delker A, Schmidt F, Tran-Gia J. On the use of artificial intelligence for dosimetry of radiopharmaceutical therapies. Nuklearmedizin. 2023;62:379–88. https://doi.org/10.1055/a-2179-6872 .
doi: 10.1055/a-2179-6872
pubmed: 37827503
Lee MS, Hwang D, Kim JH, Lee JS. Deep-dose: a voxel dose estimation method using deep convolutional neural network for personalized internal dosimetry. Sci Rep. 2019;9:10308. https://doi.org/10.1038/s41598-019-46620-y .
doi: 10.1038/s41598-019-46620-y
pubmed: 31311963
pmcid: 6635490
Gotz TI, Schmidkonz C, Chen S, Al-Baddai S, Kuwert T, Lang EW. A deep learning approach to radiation dose estimation. Phys Med Biol. 2020;65:035007. https://doi.org/10.1088/1361-6560/ab65dc .
doi: 10.1088/1361-6560/ab65dc
pubmed: 31881547
Li Z, Fessler JA, Mikell JK, Wilderman SJ, Dewaraja YK. DblurDoseNet: a deep residual learning network for voxel radionuclide dosimetry compensating for single-photon emission computerized tomography imaging resolution. Med Phys. 2022;49:1216–30. https://doi.org/10.1002/mp.15397 .
doi: 10.1002/mp.15397
pubmed: 34882821
Kim KM, Lee MS, Suh MS, Cheon GJ, Lee JS. Voxel-based internal dosimetry for 177Lu-labeled radiopharmaceutical therapy using deep residual learning. Nucl Med Mol Imaging. 2022. https://doi.org/10.1007/s13139-022-00769-z .
doi: 10.1007/s13139-022-00769-z
pubmed: 36998593
pmcid: 9679044
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:201011929. 2020.
Vaswani A, Shazeer N, Parmar N, Uszkoreit J, Jones L, Gomez AN, et al. Attention is all you need. Advances in neural information processing systems. 2017;30.
Hutchins D, Schlag I, Wu Y, Dyer E, Neyshabur B. Block-recurrent transformers. Adv Neural Inf Process t. 2022;35:33248–61.
Wu H, Xu J, Wang J, Long M. Autoformer: decomposition transformers with auto-correlation for long-term series forecasting. Adv Neural Inf Process. 2021;34:22419–30.
Delgado-Santos P, Tolosana R, Guest R, Deravi F, Vera-Rodriguez R. Exploring transformers for behavioural biometrics: a case study in gait recognition. Pattern Recognit. 2023;143:109798.
doi: 10.1016/j.patcog.2023.109798
Del Prete M, Arsenault F, Saighi N, Zhao W, Buteau FA, Celler A, et al. Accuracy and reproducibility of simplified QSPECT dosimetry for personalized (177)Lu-octreotate PRRT. EJNMMI Phys. 2018;5:25. https://doi.org/10.1186/s40658-018-0224-9 .
doi: 10.1186/s40658-018-0224-9
pubmed: 30318563
pmcid: 6186532
Vija AH, Hawman EG, Engdahl JC. Analysis of a SPECT OSEM reconstruction method with 3D beam modeling and optional attenuation correction: phantom studies. 2003 IEEE Nuclear Science Symposium Conference Record (IEEE Cat No 03CH37515): IEEE; 2003. p 2662-6.
Zhao W, Esquinas PL, Frezza A, Hou X, Beauregard JM, Celler A. Accuracy of kidney dosimetry performed using simplified time activity curve modelling methods: a (177)Lu-DOTATATE patient study. Phys Med Biol. 2019;64:175006. https://doi.org/10.1088/1361-6560/ab3039 .
doi: 10.1088/1361-6560/ab3039
pubmed: 31287093
Schneider W. Correlation between CT numbers and tissue parameters needed for Monte Carlo simulations of clinical dose distributions. Phys Med Biol. 2000;45:459–78.
doi: 10.1088/0031-9155/45/2/314
pubmed: 10701515
Waters LS. MCNPX user’s manual. Los Alamos National Laboratory.2002;124.
Dewaraja YK, Mirando DM, Peterson AB, Niedbala J, Millet JD, Mikell JK, et al. A pipeline for automated voxel dosimetry: application in patients with multi-SPECT/CT imaging after (177)Lu-peptide receptor radionuclide therapy. J Nucl Med. 2022;63:1665–72. https://doi.org/10.2967/jnumed.121.263738 .
doi: 10.2967/jnumed.121.263738
pubmed: 35422445
pmcid: 9635679
Salimi Y, Shiri I, Mansouri Z, Zaidi H. Deep learning-assisted multiple organ segmentation from whole-body CT images. medRxiv. 2023:2023.10.20.23297331. https://doi.org/10.1101/2023.10.20.23297331.
Graves SA, Flynn RT, Hyer DE. Dose point kernels for 2,174 radionuclides. Med Phys. 2019;46:5284–93.
doi: 10.1002/mp.13789
pubmed: 31461537
Hatamizadeh A, Tang Y, Nath V, Yang D, Myronenko A, Landman B, et al. Unetr: transformers for 3d medical image segmentation. Proceedings of the IEEE/CVF winter conference on applications of computer vision; 2022. p 574-84.
Isensee F, Jäger PF, Kohl SA, Petersen J, Maier-Hein KH. Automated design of deep learning methods for biomedical image segmentation. arXiv preprint arXiv:190408128. 2019.
Futrega M, Milesi A, Marcinkiewicz M, Ribalta P. Optimized U-Net for brain tumor segmentation. International MICCAI Brainlesion Workshop: Springer; 2021. p. 15–29.
Qin X, Zhang Z, Huang C, Dehghan M, Zaiane OR, Jagersand M. U2-Net: going deeper with nested U-structure for salient object detection. Pattern Recognit. 2020;106:107404.
doi: 10.1016/j.patcog.2020.107404
Ivashchenko OV, Odoherty J, Hardiansyah D, Cremonesi M, Tran-Gia J, Hippeläinen E, et al. Time-activity data fitting in molecular radiotherapy: methodology and pitfalls. Physica Medica. 2023;117:103192.
doi: 10.1016/j.ejmp.2023.103192
pubmed: 38052710
Datta NR, Das KM, Balasubramanium R, Ayyagari S. Spatial information on dose distribution using multisectional dose-volume histograms. Med Dosim. 1996;21:19–22.
doi: 10.1016/0958-3947(95)02057-8
pubmed: 8679066
Li Z, Fessler JA, Mikell JK, Wilderman SJ, Dewaraja YK. DblurDoseNet: a deep residual learning network for voxel radionuclide dosimetry compensating for single-photon emission computerized tomography imaging resolution. Med Phys. 2022;49:1216–30.
doi: 10.1002/mp.15397
pubmed: 34882821
Devlin J, Chang M-W, Lee K, Toutanova K. Bert: Pre-training of deep bidirectional transformers for language understanding. arXiv preprint arXiv:181004805. 2018.
Champendal M, Müller H, Prior JO, Dos Reis CS. A scoping review of interpretability and explainability concerning artificial intelligence methods in medical imaging. Eur J Radiol. 2023;169:111159.
doi: 10.1016/j.ejrad.2023.111159
pubmed: 37976760
Dewaraja YK, Frey EC, Sgouros G, Brill AB, Roberson P, Zanzonico PB, et al. MIRD pamphlet No 23: quantitative SPECT for patient-specific 3-dimensional dosimetry in internal radionuclide therapy. J Nucl Med. 2012;53:1310–25. https://doi.org/10.2967/jnumed.111.100123 .
doi: 10.2967/jnumed.111.100123
pubmed: 22743252