Rapid estimation of patient-specific organ doses using a deep learning network.
CT
deep learning
dosimetry
Journal
Medical physics
ISSN: 2473-4209
Titre abrégé: Med Phys
Pays: United States
ID NLM: 0425746
Informations de publication
Date de publication:
Nov 2023
Nov 2023
Historique:
revised:
23
01
2023
received:
03
08
2022
accepted:
26
02
2023
medline:
6
11
2023
pubmed:
15
3
2023
entrez:
14
3
2023
Statut:
ppublish
Résumé
Patient-specific organ-dose estimation in diagnostic CT examinations can provide useful insights on individualized secondary cancer risks, protocol optimization, and patient management. Current dose estimation techniques mainly rely on time-consuming Monte Carlo methods or/and generalized anthropomorphic phantoms. We proposed a proof-of-concept rapid workflow based on deep learning networks to estimate organ doses for individuals following thorax Computed Tomography (CT) examinations. CT scan data from 95 individuals undergoing thorax CT examinations were used. Monte Carlo simulations were performed and three-dimensional (3D) dose distributions for each patient were obtained. A fully connected sequential deep learning network model was constructed and trained for each organ considered in this study. Water-equivalent diameter (WED), scan length, and tube current were the independent variables. Organ doses for heart, lungs, esophagus, and bones were calculated from the Monte Carlo 3D distribution and used to train the deep learning networks. Organ dose predictions from each network were evaluated using an independent data set of 19 patients. The trained networks provided organ dose predictions within a second. There was very good agreement between the deep learning network predictions and reference organ dose values calculated from Monte Carlo simulations. The average difference was -1.5% for heart, -1.6% for esophagus, -1.0% for lungs, and -0.4% for bones in the 95 patients dataset, and -5.1%, 4.3%, 0.9%, and 1.4% respectively in the 19 patients test dataset. The proposed workflow demonstrated that patient-specific organ-doses can be estimated in nearly real-time using deep learning networks. The workflow can be readily implemented and requires a small set of representative data for training.
Sections du résumé
BACKGROUND
BACKGROUND
Patient-specific organ-dose estimation in diagnostic CT examinations can provide useful insights on individualized secondary cancer risks, protocol optimization, and patient management. Current dose estimation techniques mainly rely on time-consuming Monte Carlo methods or/and generalized anthropomorphic phantoms.
PURPOSE
OBJECTIVE
We proposed a proof-of-concept rapid workflow based on deep learning networks to estimate organ doses for individuals following thorax Computed Tomography (CT) examinations.
METHODS
METHODS
CT scan data from 95 individuals undergoing thorax CT examinations were used. Monte Carlo simulations were performed and three-dimensional (3D) dose distributions for each patient were obtained. A fully connected sequential deep learning network model was constructed and trained for each organ considered in this study. Water-equivalent diameter (WED), scan length, and tube current were the independent variables. Organ doses for heart, lungs, esophagus, and bones were calculated from the Monte Carlo 3D distribution and used to train the deep learning networks. Organ dose predictions from each network were evaluated using an independent data set of 19 patients.
RESULTS
RESULTS
The trained networks provided organ dose predictions within a second. There was very good agreement between the deep learning network predictions and reference organ dose values calculated from Monte Carlo simulations. The average difference was -1.5% for heart, -1.6% for esophagus, -1.0% for lungs, and -0.4% for bones in the 95 patients dataset, and -5.1%, 4.3%, 0.9%, and 1.4% respectively in the 19 patients test dataset.
CONCLUSIONS
CONCLUSIONS
The proposed workflow demonstrated that patient-specific organ-doses can be estimated in nearly real-time using deep learning networks. The workflow can be readily implemented and requires a small set of representative data for training.
Types de publication
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Pagination
7236-7244Subventions
Organisme : Euratom research and training programme 2019-2020
ID : 945196
Informations de copyright
© 2023 The Authors. Medical Physics published by Wiley Periodicals LLC on behalf of American Association of Physicists in Medicine.
Références
Damilakis J, Dosimetry CT. Invest Radiol. 2021;56(1):62-68.
Kalender WA. Dose in x-ray computed tomography. Phys Med Biol. 2014;59(3):R129-R150.
McCollough CH, Bruesewitz MR, Kofler JM. CT dose reduction and dose management tools: overview of available options. RadioGraphics. 2006;26(2):503-512.
Lawson M, Berk K, Badawy M, Qi Y, Kuganesan A, Metcalfe P. Comparison of organ and effective dose estimations from different Monte Carlo simulation-based software methods in infant CT and comparison with direct phantom measurements. J Appl Clin Med Phys. 2022;23(6).
Chen W, Kolditz D, Beister M, Bohle R, Kalender WA. Fast on-site Monte Carlo tool for dose calculations in CT applications. Med Phys. 2012;39(6Part1):2985-2996.
Perisinakis K, Tzedakis A, Damilakis J. On the use of Monte Carlo-derived dosimetric data in the estimation of patient dose from CT examinations. Med Phys. 2008;35(5):2018-2028.
DeMarco JJ, Cagnon CH, Cody DD, et al. A Monte Carlo based method to estimate radiation dose from multidetector CT (MDCT): cylindrical and anthropomorphic phantoms. Phys Med Biol. 2005;50(17):3989-4004.
Maier J, Klein L, Eulig E, Sawall S, Kachelrieß M. Real-time estimation of patient-specific dose distributions for medical CT using the deep dose estimation. Med Phys. 2022;49(4):2259-2269.
Tzanis E, Damilakis J. A novel methodology to train and deploy a machine learning model for personalized dose assessment in head CT. Eur Radiol. 2022;32(9):6418-6426.
Götz 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(3):035007.
Schindelin J, Arganda-Carreras I, Frise E, et al. Fiji: an open-source platform for biological-image analysis. Nat Methods. 2012;9(7):676-682.
Picture thresholding using an iterative selection method. IEEE Trans. Syst. Man. Cybern. 1978;8(8):630-632.
McCollough C, Bakalyar DM, Bostani M, et al. Use of water equivalent diameter for calculating patient size and size-specific dose estimates (SSDE) in CT: the report of AAPM task group 220. AAPM Rep. 2014(220):6-23.
Deak P, van Straten M, Shrimpton PC, Zankl M, Kalender WA. Validation of a Monte Carlo tool for patient-specific dose simulations in multi-slice computed tomography. Eur Radiol. 2008;18(4):759-772.
Myronakis M, Perisinakis K, Tzedakis A, Gourtsoyianni S, Damilakis J. Evaluation of a patient-specific Monte Carlo software for CT dosimetry. Radiat Prot Dosimetry. 2009;133(4):248-255.
The Mathworks I. Deep Learning Toolbox: Natick, Massachusetts, United States. 2011. https://www.mathworks.com/help/deeplearning/
Sahbaee P, Abadi E, Segars WP, Marin D, Nelson RC, Samei E. The effect of contrast material on radiation dose at CT: part II. A systematic evaluation across 58 patient models. Radiology. 2017;283(3):749-757.