Artificial neural network modelling of megavoltage photon dose distributions.

An artificial neural network (NN) has been used to model the two-dimensional dose distributions from a Varian 2100C linac. The network was trained using depth dose data for 6 and 10 MV x-rays, collected during the linac commissioning phase. During training, the number of iterations and hidden nodes was adjusted manually until acceptable agreement between measured and predicted data was obtained. In order to validate the network a subset of the data was set aside and not used for training. This enabled the performance of the network to be investigated in terms of generalization and accuracy, together with its ability to interpolate between different field sizes and positions in the beam. Finally, the network was used to generate data points over a 2D grid so that isodose distributions could be visualized. Good agreement was found between measured data and that produced by the trained neural network.

[1]  Mattias Ohlsson,et al.  WeAidU - a decision support system for myocardial perfusion images using artificial neural networks , 2004, Artif. Intell. Medicine.

[2]  B. Faddegon,et al.  Description and dosimetric verification of the PEREGRINE Monte Carlo dose calculation system for photon beams incident on a water phantom. , 2001, Medical physics.

[3]  Jin Park,et al.  A sequential neural network model for diabetes prediction , 2001, Artif. Intell. Medicine.

[4]  M Thelen,et al.  Improved artificial neural networks in prediction of malignancy of lesions in contrast-enhanced MR-mammography. , 2003, Medical physics.

[5]  A. Ahnesjö.  Collapsed cone convolution of radiant energy for photon dose calculation in heterogeneous media. , 1989, Medical physics.

[6]  Robert J. Shalek,et al.  Determination of Absorbed Dose in a Patient Irradiated by Beams of X or Gamma Rays in Radiotherapy Procedures , 1977 .

[7]  E B Podgorsak,et al.  An analytical expression for electron beam central axis depth doses. , 1994, Medical physics.

[8]  A Lamb,et al.  Investigation and modelling of the surface dose from linear accelerator produced 6 and 10 MV photon beams. , 1998, Physics in medicine and biology.

[9]  A. Ahnesjö,et al.  A pencil beam model for photon dose calculation. , 1992, Medical physics.

[10]  S. S. Dlay,et al.  Prediction of nodal spread of breast cancer by using artificial neural network-based analyses of S100A4, nm23 and steroid receptor expression , 2004, Clinical & Experimental Metastasis.

[11]  J Milan,et al.  The storage and manipulation of radiation dose data in a small digital computer. , 1974, The British journal of radiology.

[12]  Lutz Prechelt,et al.  Automatic early stopping using cross validation: quantifying the criteria , 1998, Neural Networks.

[13]  M. Cenci,et al.  PAPNET-assisted primary screening of conventional cervical smears. , 2000, Anticancer research.

[14]  Kevin N. Gurney,et al.  An introduction to neural networks , 2018 .

[15]  C G Rowbottom,et al.  Beam-orientation customization using an artificial neural network. , 1999, Physics in medicine and biology.

[16]  X. Wu,et al.  A neural network regression model for relative dose computation. , 2000, Physics in medicine and biology.

[17]  S Jayaraman,et al.  Computer generation of cobalt-60 single beam dose distribution using an analytical beam model. , 1981, Strahlentherapie.