Real-time, acquisition parameter-free voxel-wise patient-specific Monte Carlo dose reconstruction in whole-body CT scanning using deep neural networks.
Computed tomography
Deep learning
Monte Carlo method
Neural networks
Journal
European radiology
ISSN: 1432-1084
Titre abrégé: Eur Radiol
Pays: Germany
ID NLM: 9114774
Informations de publication
Date de publication:
Dec 2023
Dec 2023
Historique:
received:
07
12
2022
accepted:
14
04
2023
revised:
28
03
2023
medline:
27
11
2023
pubmed:
27
6
2023
entrez:
27
6
2023
Statut:
ppublish
Résumé
We propose a deep learning-guided approach to generate voxel-based absorbed dose maps from whole-body CT acquisitions. The voxel-wise dose maps corresponding to each source position/angle were calculated using Monte Carlo (MC) simulations considering patient- and scanner-specific characteristics (SP_MC). The dose distribution in a uniform cylinder was computed through MC calculations (SP_uniform). The density map and SP_uniform dose maps were fed into a residual deep neural network (DNN) to predict SP_MC through an image regression task. The whole-body dose maps reconstructed by the DNN and MC were compared in the 11 test cases scanned with two tube voltages through transfer learning with/without tube current modulation (TCM). The voxel-wise and organ-wise dose evaluations, such as mean error (ME, mGy), mean absolute error (MAE, mGy), relative error (RE, %), and relative absolute error (RAE, %), were performed. The model performance for the 120 kVp and TCM test set in terms of ME, MAE, RE, and RAE voxel-wise parameters was - 0.0302 ± 0.0244 mGy, 0.0854 ± 0.0279 mGy, - 1.13 ± 1.41%, and 7.17 ± 0.44%, respectively. The organ-wise errors for 120 kVp and TCM scenario averaged over all segmented organs in terms of ME, MAE, RE, and RAE were - 0.144 ± 0.342 mGy, and 0.23 ± 0.28 mGy, - 1.11 ± 2.90%, 2.34 ± 2.03%, respectively. Our proposed deep learning model is able to generate voxel-level dose maps from a whole-body CT scan with reasonable accuracy suitable for organ-level absorbed dose estimation. We proposed a novel method for voxel dose map calculation using deep neural networks. This work is clinically relevant since accurate dose calculation for patients can be carried out within acceptable computational time compared to lengthy Monte Carlo calculations. • We proposed a deep neural network approach as an alternative to Monte Carlo dose calculation. • Our proposed deep learning model is able to generate voxel-level dose maps from a whole-body CT scan with reasonable accuracy, suitable for organ-level dose estimation. • By generating a dose distribution from a single source position, our model can generate accurate and personalized dose maps for a wide range of acquisition parameters.
Identifiants
pubmed: 37368113
doi: 10.1007/s00330-023-09839-y
pii: 10.1007/s00330-023-09839-y
pmc: PMC10667156
doi:
Types de publication
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Pagination
9411-9424Subventions
Organisme : European Commission
ID : Euratom research
Organisme : European Commission
ID : training programme 2019-2020 Sinfonia project under grant agreement No 945196
Informations de copyright
© 2023. The Author(s).
Références
Alkadhi H, Euler A (2020) The future of computed tomography: personalized, functional, and precise. Invest Radiol 55(9):545–555
doi: 10.1097/RLI.0000000000000668
pubmed: 32209817
Ritt P (2022) Recent developments in SPECT/CT. Sem Nucl Med 52(3):276–285
doi: 10.1053/j.semnuclmed.2022.01.004
Aide N, Lasnon C, Desmonts C, Armstrong IS, Walker MD, McGowan DR (2022) Advances in PET/CT technology: an update. Sem Nucl Med 52(3):286-301
Kalender WA (2014) Dose in X-ray computed tomography. Phys Med Biol 59(3):R129–R150
doi: 10.1088/0031-9155/59/3/R129
pubmed: 24434792
Salimi Y, Deevband MR, Ghafarian P, Ay MR (2018) Uncertainties in effective dose estimation for CT transmission scan in total body PET-CT imaging with Auto mA3D tube current modulation. Int J Radiat Res 16(4):465-472
Menzel HG, Clement C, DeLuca P (2009) ICRP Publication 110. Realistic reference phantoms: an ICRP/ICRU joint effort. A report of adult reference computational phantoms. Ann ICRP 39(2):1–164
Tsalafoutas IA, Kharita MS, Al-Naemi H, Kalra MK (2020) Radiation dose monitoring in computed tomography: status, options and limitations. Phys Med 79:1-15
Li X, Steigerwalt D, Rehani MM (2022) T-shirt size as a classification for body habitus in computed tomography (CT) and development of size-based dose reference levels for different indications. Eur J Radiol 151:110289
doi: 10.1016/j.ejrad.2022.110289
pubmed: 35397408
Ria F, Fu W, Hoye J, Segars WP, Kapadia AJ, Samei E (2021) Comparison of 12 surrogates to characterize CT radiation risk across a clinical population. Eur Radiol 31(9):7022-7030
Klein L, Liu C, Steidel J et al (2022) Patient-specific radiation risk-based tube current modulation for diagnostic CT. Med Phys 49(7):4391-4403
Martin CJ, Barnard M (2022) How much should we be concerned about cumulative effective doses in medical imaging? J Radiol Prot 42(1):011514. https://doi.org/10.1088/1361-6498/ac31c1
Zewde N, Ria F, Rehani MM (2022) Organ doses and cancer risk assessment in patients exposed to high doses from recurrent CT exams. Eur J Radiol 149:110224
doi: 10.1016/j.ejrad.2022.110224
pubmed: 35217250
Arellano RS, Yang K, Rehani MM (2021) Analysis of patients receiving >/= 100 mSv during a computed tomography intervention. Eur Radiol 31(5):3065–3070
doi: 10.1007/s00330-020-07458-5
pubmed: 33180164
Bayer HealthCare. Radimetrics enterprise platform: dose management solution. [cited 2016; Available from: http://www.radiologysolutions.bayer.com/products/ct/dosemanagement/rep/
Akhavanallaf A, Fayad H, Salimi Y et al (2022) An update on computational anthropomorphic anatomical models. Digit Health 8:20552076221111941
Xie T, Akhavanallaf A, Zaidi H (2019) Construction of patient-specific computational models for organ dose estimation in radiological imaging. Med Phys 46(5):2403–2411
doi: 10.1002/mp.13471
pubmed: 30854654
Guberina N, Suntharalingam S, Naßenstein K et al (2018) Verification of organ doses calculated by a dose monitoring software tool based on Monte Carlo simulation in thoracic CT protocols. Acta Radiol 59(3):322-326
Al-Senan R, Brown K, Erdman M, King S et al (2020) The uncertainty of thyroid dose estimate in chest CT. Biomed Phys Eng Express 6(6)
Tahiri M, Mkimel M, Benameur Y, El Baydaoui R, Mesradi MR, El Rhazouani O (2021) Organ dose estimation for adult chest CT examination using GATE Monte Carlo simulation. Phys Part Nucl Lett 18(4):502-509
Sharma S, Kapadia A, Fu W, Abadi E, Segars WP, Samei E (2019) A real-time Monte Carlo tool for individualized dose estimations in clinical CT. Phys Med Biol 64(21):215020
De Man B, Wu M, FitzGerald P, Kalra M, Yin Z (2015) Dose reconstruction for real-time patient-specific dose estimation in CT. Med Phys 42(5):2740-51
van der Velden BHM, Kuijf HJ, Gilhuijs KGA, Viergever MA (2022) Explainable artificial intelligence (XAI) in deep learning-based medical image analysis. Med Image Anal 79:102470
Fu Y, Lei Y, Wang T, Curran WJ, Liu T, Yang X (2020) Deep learning in medical image registration: a review. Phys Med Biol 65(20):20TR01
Fu Y, Lei Y, Wang T, Curran WJ, Liu T, Yang X (2021) A review of deep learning based methods for medical image multi-organ segmentation. Phys Med 85:107-122
Maier J, Klein L, Eulig E, Sawall S, Kachelrieß M (2022) Real-time estimation of patient-specific dose distributions for medical CT using the deep dose estimation. Med Phys 49(4):2259-2269
Tzanis E, Damilakis J (2022) A novel methodology to train and deploy a machine learning model for personalized dose assessment in head CT. Eur Radiol 32(9):6418–6426
doi: 10.1007/s00330-022-08756-w
pubmed: 35384458
Kulathilake KASH, Abdullah NA, Sabri AQM, Lai KW (2023) A review on deep learning approaches for low-dose computed tomography restoration. Complex Intell Systems 9(3):2713-2745. https://doi.org/10.1007/s40747-021-00405-x
Salimi Y, Shiri I, Akhavanallaf A et al (2021) Deep learning-based fully automated Z-axis coverage range definition from scout scans to eliminate overscanning in chest CT imaging. Insights Imaging 12(1):162
Schneider W, Bortfeld T, Schlegel W (2000) Correlation between CT numbers and tissue parameters needed for Monte Carlo simulations of clinical dose distributions. Phys Med Biol 45(2):459–478
doi: 10.1088/0031-9155/45/2/314
pubmed: 10701515
Akhavanallaf A, Xie T, Zaidi H (2020) Assessment of uncertainties associated with Monte Carlo-based personalized dosimetry in clinical CT examinations. Phys Med Biol 65(4):045008
doi: 10.1088/1361-6560/ab6b45
pubmed: 31935713
Hendricks JS, McKinney GW, Fensin ML et al (2008) MCNPX 2.6.0 Extensions. Los Alamos National Laboratory Tech. Rep. LA-UR-08-02216. Los Alamos, NM, USA
McCollough CH, McCollough SL, Schneider JJ et al (2022) Dependence of Water-equivalent diameter and size-specific dose estimates on CT tube potential. Radiology 303(2):404-411
Low DA, Harms WB, Mutic S, Purdy JA (1998) A technique for the quantitative evaluation of dose distributions. Med Phys 25(5):656-61
AAPM (2014) Use of water equivalent diameter for calculating patient size and size-specific dose estimates (SSDE) in CT. AAPM
Wang A, Maslowski A, Wareing T, Star-Lack J, Gilat Schmidt T (2019) A fast, linear Boltzmann transport equation solver for computed tomography dose calculation (Acuros CTD). Med Phys 46(2):925-933