Mitigating inherent noise in Monte Carlo dose distributions using dilated U-Net

DeepSMCP - Deep-learning powered denoising of Monte Carlo dose distributions within the Swiss Monte Carlo Plan

This work demonstrated the development of a fast, deep-learning framework (DeepSMCP) to mitigate noise in Monte Carlo dose distributions (MC-DDs) of photon treatment plans with high statistical uncertainty (SU) and its integration into the Swiss Monte Carlo Plan (SMCP). To this end, a two-channel input (MC-DD and computed tomography (CT) scan) 3D U-net was trained, validated and tested (80%/10%/10%) on high/low-SU MC-DD-pairs of 106 clinically-motivated VMAT arcs for 29 available CTs, augmented to 3074 pairs. The model was integrated into SMCP to enable a "one-click" workflow of calculating and denoising MC-DDs of high SU to obtain MC-DDs of low SU. The model accuracy was evaluated on the test set using Gamma passing rate (2% global, 2 mm, 10% threshold) comparing denoised and low-SU MC-DD. Calculation time for the whole workflow was recorded. Denoised MC-DDs match low-SU MC-DDs with average (standard deviation) Gamma passing rate of 82.9% (4.7%). Additional application of DeepSMCP to 12 unseen clinically-motivated cases of different treatment sites, including treatment sites not present during training, resulted in an average Gamma passing rate of 91.0%. Denoised DDs were obtained on average in 35.1 s, a 340-fold efficiency gain compared to low-SU MC-DD calculation. DeepSMCP presented a first seamlessly integrated promising denoising framework for MC-DDs.

A new GPU-based Monte Carlo code for helium ion therapy

PURPOSE

This work presents an effort to extend the capabilities of the previously introduced GPU-based Monte Carlo code ARCHER for helium ion therapy.

METHODS

ARCHER performs helium ion transport simulations in voxelized geometry, covering kinetic energy levels up to 220 MeV/u. The physical processes are modeled using a class II condensed-history algorithm, considering ionization, energy straggling, multiple scattering, and elastic and inelastic nuclear interactions. A new nuclear-event-repeat algorithm is proposed to generate inelastic nuclear reaction products. Secondary protons, deuterons, tritons, and 3He particles are tracked, while other particles either deposit their energy locally or are ignored. The code is developed under the compute unified device architecture (CUDA) platform to improve computational efficiency. Validations are conducted by benchmarking our code against TOPAS in different phantoms.

RESULTS

Dose distribution comparisons demonstrate strong agreement between our code and TOPAS. The mean point-by-point local relative errors in the region where the dose exceeds 10% of the maximum dose range from 0.25% to 1.31% for all phantoms. In the strict 1%/1 mm criterion, gamma passing rates for a head-neck case, chest case, and prostate case are 99.8%, 96.9%, and 99.6%, respectively. Except for the lung phantom, ARCHER takes less than 10 s to simulate 10 million primary helium ions using a single NVIDIA GeForce RTX 3080 card (NVIDIA Corporation, Santa Clara, USA), while TOPAS requires several minutes on a computational platform with two Intel Xeon Gold 6348 CPUs (Intel Corporation, Santa Clara, USA) with 56 cores.

CONCLUSION

This work presents the development and benchmarking of the first GPU-based dose engine for helium ion therapy. The code has been proven to achieve high levels of accuracy and efficiency.

Deep learning for high-resolution dose prediction in high dose rate brachytherapy for breast cancer treatment

Objective.Monte Carlo (MC) simulations are the benchmark for accurate radiotherapy dose calculations, notably in patient-specific high dose rate brachytherapy (HDR BT), in cases where considering tissue heterogeneities is critical. However, the lengthy computational time limits the practical application of MC simulations. Prior research used deep learning (DL) for dose prediction as an alternative to MC simulations. While accurate dose predictions akin to MC were attained, graphics processing unit limitations constrained these predictions to large voxels of 3 mm × 3 mm × 3 mm. This study aimed to enable dose predictions as accurate as MC simulations in 1 mm × 1 mm × 1 mm voxels within a clinically acceptable timeframe.Approach.Computed tomography scans of 98 breast cancer patients treated with Iridium-192-based HDR BT were used: 70 for training, 14 for validation, and 14 for testing. A new cropping strategy based on the distance to the seed was devised to reduce the volume size, enabling efficient training of 3D DL models using 1 mm × 1 mm × 1 mm dose grids. Additionally, novel DL architecture with layer-level fusion were proposed to predict MC simulated dose to medium-in-medium (Dm,m). These architectures fuse information from TG-43 dose to water-in-water (Dw,w) with patient tissue composition at the layer-level. Different inputs describing patient body composition were investigated.Main results.The proposed approach demonstrated state-of-the-art performance, on par with the MCDm,mmaps, but 300 times faster. The mean absolute percent error for dosimetric indices between the MC and DL-predicted complete treatment plans was 0.17% ± 0.15% for the planning target volumeV100, 0.30% ± 0.32% for the skinD2cc, 0.82% ± 0.79% for the lungD2cc, 0.34% ± 0.29% for the chest wallD2ccand 1.08% ± 0.98% for the heartD2cc.Significance.Unlike the time-consuming MC simulations, the proposed novel strategy efficiently converts TG-43Dw,wmaps into preciseDm,mmaps at high resolution, enabling clinical integration.

Deep learning-based fast denoising of Monte Carlo dose calculation in carbon ion radiotherapy

BACKGROUND

Plan verification is one of the important steps of quality assurance (QA) in carbon ion radiotherapy. Conventional methods of plan verification are based on phantom measurement, which is labor-intensive and time-consuming. Although the plan verification method based on Monte Carlo (MC) simulation provides a more accurate modeling of the physics, it is also time-consuming when simulating with a large number of particles. Therefore, how to ensure the accuracy of simulation results while reducing simulation time is the current difficulty and focus.

PURPOSE

The purpose of this work was to evaluate the feasibility of using deep learning-based MC denoising method to accelerate carbon-ion radiotherapy plan verification.

METHODS

Three models, including CycleGAN, 3DUNet and GhostUNet with Ghost module, were used to denoise the 1 × 106 carbon ions-based MC dose distribution to the accuracy of 1 × 108 carbon ions-based dose distribution. The CycleGAN's generator, 3DUNet and GhostUNet were all derived from the 3DUNet network. A total of 59 cases including 29 patients with head-and-neck cancers and 30 patients with lung cancers were collected, and 48 cases were randomly selected as the training set of the CycleGAN network and six cases as the test set. For the 3DUNet and GhostUNet models, the numbers of training set, validation set, and test set were 47, 6, and 6, respectively. Finally, the three models were evaluated qualitatively and quantitatively using RMSE and three-dimensional gamma analysis (3 mm, 3%).

RESULTS

The three end-to-end trained models could be used for denoising the 1 × 106 carbon ions-based dose distribution, and their generalization was proved. The GhostUNet obtained the lowest RMSE value of 0.075, indicating the smallest difference between its denoised and 1 × 108 carbon ions-based dose distributions. The average gamma passing rate (GPR) between the GhostUNet denoising-based versus 1 × 108 carbon ions-based dose distributions was 99.1%, higher than that of the CycleGAN at 94.3% and the 3DUNet at 96.2%. Among the three models, the GhostUNet model had the fewest parameters (4.27 million) and the shortest training time (99 s per epoch) but achieved the best denoising results.

CONCLUSION

The end-to-end deep network GhostUNet outperforms the CycleGAN, 3DUNet models in denoising MC dose distributions for carbon ion radiotherapy. The network requires less than 5 s to denoise a sample of MC simulation with few particles to obtain a qualitative and quantitative result comparable to the dose distribution simulated by MC with relatively large number particles, offering a significant reduction in computation time.

Multislice input for 2D and 3D residual convolutional neural network noise reduction in CT

Purpose

Deep convolutional neural network (CNN)-based methods are increasingly used for reducing image noise in computed tomography (CT). Current attempts at CNN denoising are based on 2D or 3D CNN models with a single- or multiple-slice input. Our work aims to investigate if the multiple-slice input improves the denoising performance compared with the single-slice input and if a 3D network architecture is better than a 2D version at utilizing the multislice input.

Approach

Two categories of network architectures can be used for the multislice input. First, multislice images can be stacked channel-wise as the multichannel input to a 2D CNN model. Second, multislice images can be employed as the 3D volumetric input to a 3D CNN model, in which the 3D convolution layers are adopted. We make performance comparisons among 2D CNN models with one, three, and seven input slices and two versions of 3D CNN models with seven input slices and one or three output slices. Evaluation was performed on liver CT images using three quantitative metrics with full-dose images as reference. Visual assessment was made by an experienced radiologist.

Results

When the input channels of the 2D CNN model increases from one to three to seven, a trend of improved performance was observed. Comparing the three models with the seven-slice input, the 3D CNN model with a one-slice output outperforms the other models in terms of noise texture and homogeneity in liver parenchyma as well as subjective visualization of vessels.

Conclusions

We conclude the that multislice input is an effective strategy for improving performance for 2D deep CNN denoising models. The pure 3D CNN model tends to have a better performance than the other models in terms of continuity across axial slices, but the difference was not significant compared with the 2D CNN model with the same number of slices as the input.

A novel multichannel deep learning model for fast denoising of Monte Carlo dose calculations: preclinical applications

Objective. In preclinical radiotherapy with kilovolt (kV) x-ray beams, accurate treatment planning is needed to improve the translation potential to clinical trials. Monte Carlo based radiation transport simulations are the gold standard to calculate the absorbed dose distribution in external beam radiotherapy. However, these simulations are notorious for their long computation time, causing a bottleneck in the workflow. Previous studies have used deep learning models to speed up these simulations for clinical megavolt (MV) beams. For kV beams, dose distributions are more affected by tissue type than for MV beams, leading to steep dose gradients. This study aims to speed up preclinical kV dose simulations by proposing a novel deep learning pipeline. Approach. A deep learning model is proposed that denoises low precision (∼106 simulated particles) dose distributions to produce high precision (109 simulated particles) dose distributions. To effectively denoise the steep dose gradients in preclinical kV dose distributions, the model uses the novel approach to use the low precision Monte Carlo dose calculation as well as the Monte Carlo uncertainty (MCU) map and the mass density map as additional input channels. The model was trained on a large synthetic dataset and tested on a real dataset with a different data distribution. To keep model inference time to a minimum, a novel method for inference optimization was developed as well. Main results. The proposed model provides dose distributions which achieve a median gamma pass rate (3%/0.3 mm) of 98% with a lower bound of 95% when compared to the high precision Monte Carlo dose distributions from the test set, which represents a different dataset distribution than the training set. Using the proposed model together with the novel inference optimization method, the total computation time was reduced from approximately 45 min to less than six seconds on average. Significance. This study presents the first model that can denoise preclinical kV instead of clinical MV Monte Carlo dose distributions. This was achieved by using the MCU and mass density maps as additional model inputs. Additionally, this study shows that training such a model on a synthetic dataset is not only a viable option, but even increases the generalization of the model compared to training on real data due to the sheer size and variety of the synthetic dataset. The application of this model will enable speeding up treatment plan optimization in the preclinical workflow.

Denoising proton therapy Monte Carlo dose distributions in multiple tumor sites: A comparative neural networks architecture study

INTRODUCTION

Monte Carlo (MC) algorithms provide accurate modeling of dose calculation by simulating the delivery and interaction of many particles through patient geometry. Fast MC simulations using large number of particles are desirable as they can lead to reliable clinical decisions. In this work, we assume that faster simulations with fewer particles can approximate slower ones by denoising them with deep learning.

MATERIALS AND METHODS

We use mean squared error (MSE) as loss function to train networks (sNet and dUNet), with 2.5D and 3D setups considering volumes of 7 and 24 slices. Our models are trained on proton therapy MC dose distributions of six different tumor sites acquired from 50 patients. We provide networks with input MC dose distributions simulated using 1 × 10⁶ particles while keeping 1 × 10⁹ particles as reference.

RESULTS

On average over 10 new patients with different tumor sites, in 2.5D and 3D, our models recover relative residual error on target volume, ΔD₉₅^TV of 0.67 ± 0.43% and 1.32 ± 0.87% for sNet vs. 0.83 ± 0.53% and 1.66 ± 0.98% for dUNet, compared to the noisy input at 12.40 ± 4.06%. Moreover, the denoising time for a dose distribution is: < 9s and  < 1s for sNet vs. < 16s and  < 1.5s for dUNet in 2.5D and 3D, in comparison to about 100 min (MC simulation using 1 × 10⁹ particles).

CONCLUSION

We propose a fast framework that can successfully denoise MC dose distributions. Starting from MC doses with 1 × 10⁶ particles only, the networks provide comparable results as MC doses with1 × 10⁹ particles, reducing simulation time significantly.

Advanced Monte Carlo simulations of emission tomography imaging systems with GATE

Built on top of the Geant4 toolkit, GATE is collaboratively developed for more than 15 years to design Monte Carlo simulations of nuclear-based imaging systems. It is, in particular, used by researchers and industrials to design, optimize, understand and create innovative emission tomography systems. In this paper, we reviewed the recent developments that have been proposed to simulate modern detectors and provide a comprehensive report on imaging systems that have been simulated and evaluated in GATE. Additionally, some methodological developments that are not specific for imaging but that can improve detector modeling and provide computation time gains, such as Variance Reduction Techniques and Artificial Intelligence integration, are described and discussed.

Deep dose plugin: towards real-time Monte Carlo dose calculation through a deep learning-based denoising algorithm

Monte Carlo (MC) simulation is considered the gold standard method for radiotherapy dose calculation. However, achieving high precision requires a large number of simulation histories, which is time-consuming. The use of computer graphics processing units (GPUs) has greatly accelerated MC simulation and allows dose calculation within a few minutes for a typical radiotherapy treatment plan. However, some clinical applications demand real-time efficiency for MC dose calculation. To tackle this problem, we have developed a real-time, deep learning (DL)-based dose denoiser that can be plugged into a current GPU-based MC dose engine to enable real-time MC dose calculation. We used two different acceleration strategies to achieve this goal: (1) we applied voxel unshuffle and voxel shuffle operators to decrease the input and output sizes without any information loss, and (2) we decoupled the 3D volumetric convolution into a 2D axial convolution and a 1D slice convolution. In addition, we used a weakly supervised learning framework to train the network, which greatly reduces the size of the required training dataset and thus enables fast fine-tuning-based adaptation of the trained model to different radiation beams. Experimental results show that the proposed denoiser can run in as little as 39 ms, which is 11.6 times faster than the baseline model. As a result, the whole MC dose calculation pipeline can be finished within ∼ 0.15 s, including both GPU MC dose calculation and DL-based denoising, achieving the real-time efficiency needed for some radiotherapy applications, such as online adaptive radiotherapy.

Artificial intelligence and machine learning for medical imaging: A technology review

Artificial intelligence (AI) has recently become a very popular buzzword, as a consequence of disruptive technical advances and impressive experimental results, notably in the field of image analysis and processing. In medicine, specialties where images are central, like radiology, pathology or oncology, have seized the opportunity and considerable efforts in research and development have been deployed to transfer the potential of AI to clinical applications. With AI becoming a more mainstream tool for typical medical imaging analysis tasks, such as diagnosis, segmentation, or classification, the key for a safe and efficient use of clinical AI applications relies, in part, on informed practitioners. The aim of this review is to present the basic technological pillars of AI, together with the state-of-the-art machine learning methods and their application to medical imaging. In addition, we discuss the new trends and future research directions. This will help the reader to understand how AI methods are now becoming an ubiquitous tool in any medical image analysis workflow and pave the way for the clinical implementation of AI-based solutions.