Toward IMRT 2D dose modeling using artificial neural networks: A feasibility study

被引:13
|
作者
Kalantzis, Georgios [1 ,2 ]
Vasquez-Quino, Luis A. [1 ]
Zalman, Travis [1 ]
Pratx, Guillem [2 ]
Lei, Yu [1 ]
机构
[1] Univ Texas Hlth Sci Ctr San Antonio, Dept Radiat Oncol, San Antonio, TX 78229 USA
[2] Stanford Univ, Dept Radiat Oncol, Sch Med, Stanford, CA 94305 USA
关键词
artificial neural network; IMRT; dose modeling; back-propagation; EPID; QUALITY-ASSURANCE; DOSIMETRIC VERIFICATION; QUANTITATIVE-EVALUATION; FILM;
D O I
10.1118/1.3639998
中图分类号
R8 [特种医学]; R445 [影像诊断学];
学科分类号
1002 ; 100207 ; 1009 ;
摘要
Purpose: To investigate the feasibility of artificial neural networks (ANN) to reconstruct dose maps for intensity modulated radiation treatment (IMRT) fields compared with those of the treatment planning system (TPS). Methods: An artificial feed forward neural network and the back-propagation learning algorithm have been used to replicate dose calculations of IMRT fields obtained from PINNACLE(3) v9.0. The ANN was trained with fluence and dose maps of IMRT fields for 6 MV x-rays, which were obtained from the amorphous silicon (a-Si) electronic portal imaging device of Novalis TX. Those fluence distributions were imported to the TPS and the dose maps were calculated on the horizontal midpoint plane of a water equivalent homogeneous cylindrical virtual phantom. Each exported 2D dose distribution from the TPS was classified into two clusters of high and low dose regions, respectively, based on the K-means algorithm and the Euclidian metric in the fluence-dose domain. The data of each cluster were divided into two sets for the training and validation phase of the ANN, respectively. After the completion of the ANN training phase, 2D dose maps were reconstructed by the ANN and isodose distributions were created. The dose maps reconstructed by ANN were evaluated and compared with the TPS, where the mean absolute deviation of the dose and the gamma-index were used. Results: A good agreement between the doses calculated from the TPS and the trained ANN was achieved. In particular, an average relative dosimetric difference of 4.6% and an average gamma-index passing rate of 93% were obtained for low dose regions, and a dosimetric difference of 2.3% and an average gamma-index passing rate of 97% for high dose region. Conclusions: An artificial neural network has been developed to convert fluence maps to corresponding dose maps. The feasibility and potential of an artificial neural network to replicate complex convolution kernels in the TPS for IMRT dose calculations have been demonstrated. (C) 2011 American Association of Physicists in Medicine. [DOI: 10.1118/1.3639998]
引用
收藏
页码:5807 / 5817
页数:11
相关论文
共 50 条
  • [1] 2D Dose Reconstruction by Artificial Neural Network for Pretreatment Verification of IMRT Fields
    Mahdavi, Seied Rabie
    Bakhshandeh, Mohsen
    Rostami, Aram
    Arfaee, Ali Jabbary
    JOURNAL OF MEDICAL IMAGING AND RADIATION SCIENCES, 2018, 49 (03) : 286 - 292
  • [2] Dose calculations using artificial neural networks:: A feasibility study for photon beams
    Vasseur, Aurelien
    Makovicka, Libor
    Martin, Eric
    Sauget, Marc
    Contassot-Vivier, Sylvain
    Bahi, Jacques
    NUCLEAR INSTRUMENTS & METHODS IN PHYSICS RESEARCH SECTION B-BEAM INTERACTIONS WITH MATERIALS AND ATOMS, 2008, 266 (07): : 1085 - 1093
  • [3] 2D COORDINATE TRANSFORMATION USING ARTIFICIAL NEURAL NETWORKS
    Konakoglu, B.
    Cakir, L.
    Gokalp, E.
    3RD INTERNATIONAL GEOADVANCES WORKSHOP, 2016, 42-2 (W1): : 183 - 186
  • [4] 2D Modeling of Turbulent Flow around a Cylindrical Storage Tank by Artificial Neural Networks
    Lauret, Pierre
    Heymes, Frederic
    Aprin, Laurent
    Johannet, Anne
    Slangen, Pierre
    ICHEAP12: 12TH INTERNATIONAL CONFERENCE ON CHEMICAL & PROCESS ENGINEERING, 2015, 43 : 1621 - 1626
  • [5] Effect of training artificial neural networks on 2D image: An example study on mammography
    Zhang, Xuejun
    Fujita, Hiroshi
    Chen, Jing
    Zhang, Zuojun
    2009 INTERNATIONAL CONFERENCE ON ARTIFICIAL INTELLIGENCE AND COMPUTATIONAL INTELLIGENCE, VOL IV, PROCEEDINGS, 2009, : 214 - +
  • [6] Region Specific Dose Prediction Using Deep Neural Networks: A Feasibility Study On the Planning Target Volume of Prostate IMRT Patients
    Nguyen, D.
    Jiang, S.
    MEDICAL PHYSICS, 2020, 47 (06) : E656 - E656
  • [7] Image Analysis using Convolutional Neural Networks for Modeling 2D Fracture Propagation
    Miller, Robyn L.
    Moore, Bryan
    Viswanathan, Hari
    Srinivasan, Gowri
    2017 17TH IEEE INTERNATIONAL CONFERENCE ON DATA MINING WORKSHOPS (ICDMW 2017), 2017, : 979 - 982
  • [8] Steel crack depth estimation based on 2D images using artificial neural networks
    Mohamed, Yasser S.
    Shehata, Hesham M.
    Abdellatif, Mohamed
    Awad, Taher H.
    ALEXANDRIA ENGINEERING JOURNAL, 2019, 58 (04) : 1167 - 1174
  • [9] Using Different Types of Artificial Neural Networks to Classify 2D Matrix Codes and Their Rotations-A Comparative Study
    Karrach, Ladislav
    Pivarciova, Elena
    JOURNAL OF IMAGING, 2023, 9 (09)
  • [10] A Gated IMRT Study Based On Comparative 2D and 3D Dose Measurements
    Dvorak, P.
    Stevens, S.
    McQuaid, M.
    Dawes, S.
    Richmond, A.
    MEDICAL PHYSICS, 2012, 39 (06) : 3879 - 3879