Subscribe to the gold package and get unlimited access to Shamra Academy
Register a new userDeepMCDose: A Deep Learning Method for Efficient Monte Carlo Beamlet Dose Calculation by Predictive Denoising in MR-Guided Radiotherapy
123
0
0.0
(
0
)
Ask ChatGPT about the research
No Arabic abstract
The next great leap toward improving treatment of cancer with radiation will require the combined use of online adaptive and magnetic resonance guided radiation therapy techniques with automatic X-ray beam orientation selection. Unfortunately, by uniting these advancements, we are met with a substantial expansion in the required dose information and consequential increase to the overall computational time imposed during radiation treatment planning, which cannot be handled by existing techniques for accelerating Monte Carlo dose calculation. We propose a deep convolutional neural network approach that unlocks new levels of acceleration and accuracy with regards to post-processed Monte Carlo dose results by relying on data-driven learned representations of low-level beamlet dose distributions instead of more limited filter-based denoising techniques that only utilize the information in a single dose input. Our method uses parallel UNET branches acting on three input channels before mixing latent understanding to produce noise-free dose predictions. Our model achieves a normalized mean absolute error of only 0.106% compared with the ground truth dose contrasting the 25.7% error of the under sampled MC dose fed into the network at prediction time. Our models per-beamlet prediction time is ~220ms, including Monte Carlo simulation and network prediction, with substantial additional acceleration expected from batched processing and combination with existing Monte Carlo acceleration techniques. Our method shows promise toward enabling clinical practice of advanced treatment technologies.
rate research
Read More
A new variant of the pencil-beam (PB) algorithm for dose distribution calculation for radiotherapy with protons and heavier ions, the grid-dose spreading (GDS) algorithm, is proposed. The GDS algorithm is intrinsically faster than conventional PB algorithms due to approximations in convolution integral, where physical calculations are decoupled from simple grid-to-grid energy transfer. It was effortlessly implemented to a carbon-ion radiotherapy treatment planning system to enable realistic beam blurring in the field, which was absent with the broad-beam (BB) algorithm. For a typical prostate treatment, the slowing factor of the GDS algorithm relative to the BB algorithm was 1.4, which is a great improvement over the conventional PB algorithms with a typical slowing factor of several tens. The GDS algorithm is mathematically equivalent to the PB algorithm for horizontal and vertical coplanar beams commonly used in carbon-ion radiotherapy while dose deformation within the size of the pristine spread occurs for angled beams, which was within 3 mm for a single proton pencil beam of $30^circ$ incidence, and needs to be assessed against the clinical requirements and tolerances in practical situations.
We have developed a model for proton depth dose and lateral distributions based on Monte Carlo calculations (GEANT4) and an integration procedure of the Bethe-Bloch equation (BBE). The model accounts for the transport of primary and secondary protons, the creation of recoil protons and heavy recoil nuclei as well as lateral scattering of these contributions. The buildup, which is experimentally observed in higher energy depth dose curves, is modeled by inclusion of two different origins: 1. Secondary reaction protons with a contribution of ca. 65 % of the buildup (for monoenergetic protons). 2. Landau tails as well as Gaussian type of fluctuations for range straggling effects. All parameters of the model for initially monoenergetic proton beams have been obtained from Monte Carlo calculations or checked by them. Furthermore, there are a few parameters, which can be obtained by fitting the model to measured depth dose curves in order to describe individual characteristics of the beamline - the most important being the initial energy spread. We find that the free parameters of the depth dose model can be predicted for any intermediate energy from a couple of measured curves.
Monte Carlo (MC) method has been recognized the most accurate dose calculation method for radiotherapy. However, its extremely long computation time impedes clinical applications. Recently, a lot of efforts have been made to realize fast MC dose calculation on GPUs. Nonetheless, most of the GPU-based MC dose engines were developed in NVidia CUDA environment. This limits the code portability to other platforms, hindering the introduction of GPU-based MC simulations to clinical practice. The objective of this paper is to develop a fast cross-platform MC dose engine oclMC using OpenCL environment for external beam photon and electron radiotherapy in MeV energy range. Coupled photon-electron MC simulation was implemented with analogue simulations for photon transports and a Class II condensed history scheme for electron transports. To test the accuracy and efficiency of our dose engine oclMC, we compared dose calculation results of oclMC and gDPM, our previously developed GPU-based MC code, for a 15 MeV electron beam and a 6 MV photon beam on a homogenous water phantom, one slab phantom and one half-slab phantom. Satisfactory agreement was observed in all the cases. The average dose differences within 10% isodose line of the maximum dose were 0.48-0.53% for the electron beam cases and 0.15-0.17% for the photon beam cases. In terms of efficiency, our dose engine oclMC was 6-17% slower than gDPM when running both codes on the same NVidia TITAN card due to both different physics particle transport models and different computational environments between CUDA and OpenCL. The cross-platform portability was also validated by successfully running our new dose engine on a set of different compute devices including an Nvidia GPU card, two AMD GPU cards and an Intel CPU card using one or four cores. Computational efficiency among these platforms was compared.
The MR-Linac is a combination of an MR-scanner and radiotherapy linear accelerator (Linac) which holds the promise to increase the precision of radiotherapy treatments with MR-guided radiotherapy by monitoring motion during radiotherapy with MRI, and adjusting the radiotherapy plan accordingly. Optimal MR-guidance for respiratory motion during radiotherapy requires MR-based 3D motion estimation with a latency of 200-500 ms. Currently this is still challenging since typical methods rely on MR-images, and are therefore limited by the 3D MR-imaging latency. In this work, we present a method to perform non-rigid 3D respiratory motion estimation with 170 ms latency, including both acquisition and reconstruction. The proposed method called real-time low-rank MR-MOTUS reconstructs motion-fields directly from k-space data, and leverages an explicit low-rank decomposition of motion-fields to split the large scale 3D+t motion-field reconstruction problem posed in our previous work into two parts: (I) a medium-scale offline preparation phase and (II) a small-scale online inference phase which exploits the results of the offline phase for real-time computations. The method was validated on free-breathing data of five volunteers, acquired with a 1.5T Elekta Unity MR-Linac. Results show that the reconstructed 3D motion-field are anatomically plausible, highly correlated with a self-navigation motion surrogate (R = 0.975 +/- 0.0110), and can be reconstructed with a total latency of 170 ms that is sufficient for real-time MR-guided abdominal radiotherapy.
Purpose: This paper describes a new method to apply deep-learning algorithms for automatic segmentation of radiosensitive organs from 3D tomographic CT images before computing organ doses using a GPU-based Monte Carlo code. Methods: A deep convolutional neural network (CNN) for organ segmentation is trained to automatically delineate radiosensitive organs from CT. With a GPU-based Monte Carlo dose engine (ARCHER) to derive CT dose of a phantom made from a subjects CT scan, we are then able to compute the patient-specific CT dose for each of the segmented organs. The developed tool is validated by using Relative Dose Error (RDE) against the organ doses calculated by ARCHER with manual segmentation performed by radiologists. The dose computation results are also compared against organ doses from population-average phantoms to demonstrate the improvement achieved by using the developed tool. In this study, two datasets were used: The Lung CT Segmentation Challenge 2017 (LCTSC) dataset, which contains 60 thoracic CT scan patients each with 5 segmented organs, and the Pancreas-CT (PCT) dataset, which contains 43 abdominal CT scan patients each with 8 segmented organs. Five-fold cross-validation of the new method is performed on both datasets. Results: Comparing with the traditional organ dose evaluation method that based on population-average phantom, our proposed method achieved the smaller RDE range on all organs with -4.3%~1.5% vs -31.5%~33.9% (lung), -7.0%~2.3% vs -15.2%~125.1% (heart), -18.8%~40.2% vs -10.3%~124.1% (esophagus) in the LCTSC dataset and -5.6%~1.6% vs -20.3%~57.4% (spleen), -4.5%~4.6% vs -19.5%~61.0% (pancreas), -2.3%~4.4% vs -37.8%~75.8% (left kidney), -14.9%~5.4% vs -39.9% ~14.6% (gall bladder), -0.9%~1.6% vs -30.1%~72.5% (liver), and -23.0%~11.1% vs -52.5%~-1.3% (stomach) in the PCT dataset.
Log in to be able to interact and post comments
comments
Fetching comments
Sign in to be able to follow your search criteria
