Pith. sign in

REVIEW 4 major objections 4 minor 39 references

Kinetic-Diffusion-Rotation Algorithm for Dose Estimation in Electron Beam Therapy

T0 review · 4 major / 4 minor · reviewed 2026-08-11 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read A hybrid scheme that switches from explicit collisions to a moment-preserving random walk estimates electron dose on a lung CT phantom nearly 33 times faster than analog Monte Carlo, at 7.7% average relative error.

desk verdict A credible proof-of-concept for kinetic-diffusion Monte Carlo in electron therapy, with a genuinely new analytic mean formula; the 7.7% error is a self-consistency check, not an independent validation. read the letter →

arxiv 2412.05063 v3 pith:3R4Y4RRH submitted 2024-12-06 q-bio.QM physics.med-ph

Klaas Willems , Vince Maes , Zhirui Tang , Giovanni Samaey This is my paper · ORCID
classification q-bio.QMphysics.med-ph MSC 35Q2060G5065C0592-10
keywords kinetic-diffusionMonteCarloelectronbeamtherapydoseestimationmoment-preservingrandomwalkmultiplescatteringdistributionvelocity-jumpprocesscontinuousslowing-downapproximationCT-basedsimulation
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper establishes that a kinetic-diffusion hybrid, originally used for neutral-particle transport in fusion, can be adapted to electron beam therapy as a kinetic-diffusion-rotation (KDR) scheme. In tissue where collisions are frequent, KDR stops simulating each collision explicitly and instead completes each step with a random walk whose mean and variance match the true kinetic motion. The paper derives an analytic formula for the mean under correlated successive velocities and supplies the variance from a precomputed lookup table, since no analytic variance is available. On a 2D CT scan of a lung patient with a simplified electron model, KDR with a multiple-scattering distribution reproduces the analog dose map to 7.7% average relative error and runs nearly 33 times faster. If this carries over to realistic energy-dependent models, it would make fast dose estimation practical where collision rates are high.

What carries the argument

The load-bearing object is the KDR step: a fixed-distance step that first performs a kinetic flight $x'_k = x_k + \Delta s_k \Omega_k$ and then, if the flight ends before the full step, completes the remaining distance with a random walk $x_{k+1} = x'_k + A_k d_k + R(\phi_k\hat{z})R(\theta_k\hat{y})\sqrt{V}\xi$. The advection vector $A_k$ is the paper's analytic mean of the kinetic motion, $A_k = \Omega_k \frac{E[\cos\theta]}{1-E[\cos\theta]}\frac{1}{\Sigma_t t_k}\left(1-e^{\Sigma_t t_k(E[\cos\theta]-1)}\right)$, which encodes the rotational correlation between successive scattering events. The diffusion matrix $V$ comes from a lookup table and is rotated into the particle's current direction, and the post-step orientation is sampled from a precomputed multiple-scattering distribution. This construction makes each step cost one collision event regardless of collision rate while preserving the mean and variance of the underlying velocity-jump process.

What would settle it

Run KDR on a lung CT phantom with the step size set equal to or larger than the voxel size and compare dose in low-density lung regions to an analog kinetic simulation; a boundary error clearly exceeding 7.7% would falsify the heterogeneous-media accuracy claim.

Watch

Extended reading notes

Core claim

The central claim is that a KDR step, consisting of one explicit kinetic flight followed by a biased Gaussian random walk, preserves enough of the true electron dynamics to serve as a dose estimator in electron beam therapy. The mean of the kinetic motion is derived exactly for fixed-energy scattering with rotationally dependent velocities, while the variance is obtained from a lookup table generated by kinetic simulations over energy, step size, and density. The orientation after the diffusive step is either read from a multiple-scattering distribution or reconstructed from the displacement; the paper reports that the multiple-scattering version is essential, reducing average relative dose error from 60% to 7.7% on the lung CT test case while achieving a speedup of nearly 33 compared with analog particle tracing.

Load-bearing premise

The weakest load-bearing assumption is that a random walk step crossing a voxel boundary can be executed with the starting voxel's material parameters; the paper states this can cause errors near material boundaries and the reported test uses a step size smaller than the voxel, so this error is not covered by the 7.7% figure.

Editorial extensions

If this is right

  • The number of explicit collisions per step drops from an average $\Sigma_t\Delta s$ to $1-e^{-\Sigma_t\Delta s}$, so runtime no longer scales with collision frequency in dense tissue.
  • With a multiple-scattering orientation update, KDR reproduces analog dose maps on the lung CT benchmark with 7.7% average relative error.
  • Without the multiple-scattering update the method over-advects and the dose error grows to 60%, so the orientation update is a necessary component of the algorithm.
  • The variance lookup table depends only on energy, step size, and density, making it reusable across patient geometries.
  • The observed speedup of nearly 33 applies to the simplified fixed-energy electron model; extending KDR to full energy-dependent collision models is left as future work.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • Because the paper chooses a step size smaller than the CT grid cell, the boundary-crossing error it identifies is not exercised; a natural next experiment is to test KDR with step sizes comparable to voxel size in a lung-air phantom.
  • The same moment-matching construction should transfer to photon or proton transport, where high collisionality also bottlenecks analog Monte Carlo; the lookup-table route for variance would apply wherever no analytic variance exists.
  • The reported low-energy mismatch in the mean formula suggests that an energy-dependent correction or piecewise energy steps could reduce bias in realistic beams without sacrificing the speedup.
  • A stronger test would compare KDR against an optimized condensed-history Monte Carlo code on the same phantom, since the 33x speedup is measured against a simple analog tracker.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 4 minor

Summary. The paper adapts the kinetic-diffusion Monte Carlo (KDMC) method, originally developed for neutral transport in fusion plasmas, to electron transport for dose estimation in electron beam therapy. The proposed Kinetic-Diffusion-Rotation (KDR) algorithm splits each step into a kinetic part and a random-walk part, where the random walk is designed to preserve the first two moments of the kinetic motion. The authors derive an analytic formula for the mean kinetic motion for a velocity-jump process with correlated velocities, use a lookup table for the variance, and add an empirically fitted multiple-scattering distribution. The algorithm is tested on a 2D CT lung case with a simplified particle model, reporting a speed-up of nearly 33x over an analog kinetic simulation with an average relative dose error of 7.7%.

Significance. The extension of KDMC to electron transport, where post-collisional velocities are correlated, is a genuine methodological contribution. The analytic mean derivation (Appendix A) is a useful step, and the algorithm is clearly described with pseudocode. The paper explicitly uses a simplified particle model and should be judged as a proof-of-concept. However, the validation is a self-consistency check: both the variance lookup table and the multiple-scattering distribution are generated from analog kinetic simulations of the same model that KDR is benchmarked against. Thus the reported 7.7% error measures only the diffusion-approximation error relative to the simplified kinetic model, not any error relative to physical electron transport. The paper would be strengthened by acknowledging this more explicitly and by addressing the unquantified boundary-crossing error in heterogeneous media.

major comments (4)
  1. [Equation (21) vs. Equation (14)] Equation (21) defines the advection coefficient as A_k = Ω_k * E[cos(θ)]/(1 - E[cos(θ)]) * 1/(Σ_t t_k) * [1 - exp(Σ_t t_k (E[cos(θ)] - 1))], but the derived mean in Equation (14) has the prefactor 1/Σ_t with no additional t_k in the denominator. As written, Equation (21) gives a different drift than the mean kinetic motion derived in Section 3.2.1 and Appendix A. If this is a typo, it must be corrected; if not, the random walk does not preserve the first moment of the kinetic motion.
  2. [Section 3.2.4 and Section 4.2] The boundary-crossing error in heterogeneous media is acknowledged but never quantified. In the lung CT test, the KDR step size is Δs = 0.0725 cm but the CT voxel spacing is not reported. Even if Δs is smaller than the voxel size, a particle starting within Δs of a voxel boundary will execute a portion of its random walk with the wrong material parameters. The reported 7.7% average relative error could mask significantly larger local errors near lung-tissue interfaces. The authors should report the voxel size, count or estimate the number of boundary-crossing steps, or compare against a split-step approach to demonstrate that this error is controlled.
  3. [Section 4] The numerical test freezes the scattering rate and polar angle distribution at 2.61 MeV for all particles, so the algorithm's handling of energy-dependent scattering is not demonstrated. The mean formula (14) is only validated at high energies (Figure 3) where energy loss is negligible; the algorithm's accuracy at lower energies, where E[cos(θ)] varies rapidly (Figure 2b), is untested. The abstract and conclusion should restrict the claims to the fixed-energy simplified model, and a discussion of how energy dependence would affect the error and speed-up is needed.
  4. [Sections 3.2.1 and 3.2.2] The variance lookup table and the multiple-scattering distribution are both generated by kinetic simulations of the same particle model used as the benchmark. Consequently, the 7.7% error is an approximation error of the KDR algorithm relative to the simplified kinetic model, not a modeling error relative to physical dose. The abstract's phrase 'small modeling error' is misleading unless this distinction is made explicit. The authors should rephrase to 'small approximation error relative to the analog simulation of the simplified particle model' or provide a comparison against a more accurate transport model.
minor comments (4)
  1. [Appendix A, Equation (38)] In Equation (38), the expectation E[Ω_j | Ω_{-1}] is written as Ω_0 E[cos(θ)]^{j+1}, but Ω_0 is a random variable; it should be Ω_{-1} E[cos(θ)]^{j+1}. The typo appears in the derivation and should be corrected to avoid confusion.
  2. [Figure 5 and Section 3.2.2] The log-normal fit to the multiple-scattering angle is demonstrated for a single parameter combination (E=3.117 MeV, Δs=0.1 cm, ρ=1.0 g/cm³). Since the lookup table spans a wide parameter range, the authors should state whether the log-normal fit quality (R²=0.972) is representative or add a brief assessment for other values.
  3. [Section 4.1 and Figure 6] The speed-up plot in Figure 6 shows a non-monotonic behavior at intermediate collisionality, but the empirical speed-up is based on a single measurement without error bars. Adding confidence intervals or repeated runs would improve the reliability of the speed-up claim.
  4. [Reference [38]] Reference [38] is listed as a repository of implementations but lacks a persistent identifier or URL. For reproducibility, a DOI or a stable link should be provided.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the KDR mean is derived analytically and the variance/MS lookup tables are generic precomputed coefficients, not predictions fitted to the benchmark dose.

full rationale

The derivation chain is self-contained. The analytic mean formula (14) is derived in Appendix A from the velocity-jump process with correlated velocities and verified against kinetic simulation; it is not imported from a fit. The variance lookup table and multiple-scattering distribution are generated by kinetic tracing of the same simplified particle model, but they are precomputed local transport coefficients over a generic parameter grid (Table 1), and the lung test in Section 4.2 is a separate heterogeneous geometry not used in their construction. Thus the 7.7% error is an out-of-sample test of how the moment-preserving random walk combined with energy loss, rotation, and boundary handling accumulates, even though the reference simulation is the same particle model. The boundary-crossing caveat in Section 3.2.4 is a genuine numerical accuracy limitation, not a circularity. Self-citations to [24,25,39] provide background theory and thesis context rather than an unverified uniqueness premise, so they do not make the argument circular.

Assumptions & free parameters 3 free parameters · 6 assumptions · 0 invented entities

The central approximation pulls several inputs from prior literature (CSD equation, screened Rutherford DCS, PENELOPE stopping power) and calibrates the random walk coefficients from kinetic simulations of the same model. The mean is analytic; the variance and multiple-scattering distribution are empirical, and the numerical test freezes electron energy at 2.61 MeV, so the ledger is dominated by calibration assumptions rather than invented entities.

free parameters (3)
  • Variance lookup table entries = Not reported numerically
    Variance of kinetic motion in x, y, z is generated by 10,000-particle kinetic simulations for a grid of energies, densities, and step sizes (Table 1) and interpolated during KDR steps. These empirical values act as the diffusion coefficient of the random walk.
  • Log-normal fit parameters for the multiple scattering angle = Not reported
    The polar multiple scattering angle distribution is fitted with a log-normal distribution (Figure 5); KDR samples velocities from this fitted distribution, so the fit parameters are calibrated to kinetic simulation data.
  • KDR step size Delta s = 0.0725 cm in the lung test
    Chosen by the user; determines the balance between accuracy and speed. No convergence study is reported, so the 33x speedup depends on this choice.
assumptions (6)
  • domain assumption The CSD Boltzmann equation (Eq. 2) with screened Rutherford elastic scattering and continuous slowing-down approximation adequately models electron dose for this study.
    Used throughout as the governing equation; bremsstrahlung, hard inelastic collisions, and inelastic angular deflection are neglected (Section 2.1).
  • ad hoc to paper Within a random walk step, the scattering rate and polar angle distribution are constant and energy loss is neglected for the mean and variance.
    Mean derivation in Appendix A assumes fixed Sigma_t and E[cos(theta)]; the paper acknowledges overestimation at low energy (Section 3.2.1, Figure 3).
  • domain assumption A normal approximation for the positional increment is valid for the correlated velocity-jump process when the collision count is high.
    KDR adopts the central limit theorem / diffusion limit argument from KDMC [25]; no formal proof is given for the correlated case.
  • ad hoc to paper The multiple scattering angle theta_MS is adequately approximated by a log-normal distribution.
    Figure 5 shows a Q-Q plot with R^2=0.972; the log-normal fit is used to sample post-random-walk velocities (Section 3.2.2).
  • domain assumption Tissue can be characterized by density alone, with chemical composition fixed as water, and density derived linearly from CT pixel values.
    Lookup tables and the lung test use this; realistic tissue composition variations are not modeled (Sections 3.2.1 and 4.2).
  • ad hoc to paper In the numerical test, all particles scatter as if their energy is 2.61 MeV.
    Section 4 states this unphysical simplification; it avoids generating energy-dependent lookup tables. The claimed speedup and error are measured only for this fixed-energy model.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Kinetic-Diffusion-Rotation Algorithm for Dose Estimation in Electron Beam Therapy." pith.science (2026). https://pith.science/paper/3R4Y4RRH

@misc{pith2026241205063,
  author       = {Pith},
  title        = {Pith review of: Kinetic-Diffusion-Rotation Algorithm for Dose Estimation in Electron Beam Therapy},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/3R4Y4RRH}},
  note         = {Machine review of arXiv:2412.05063}
}
read the original abstract

Monte Carlo methods are state-of-the-art when it comes to dosimetric computations in radiotherapy. However, the execution time of these methods suffers in high-collisional regimes. We address this problem by introducing a kinetic-diffusion particle tracing scheme. This algorithm, first proposed in the context of neutral transport in fusion energy, relies on explicit simulation of the kinetic motion in low-collisional regimes and dynamically switches to motion based on a random walk in high-collisional regimes. The random walk motion maintains the first two moments (mean and variance) of the kinetic motion. We derive an analytic formula for the mean kinetic motion and discuss the addition of a multiple scattering distribution to the algorithm. In contrast to neutral transport, the radiation transfer setting does not readily admit to an analytical expression for the variance of the kinetic motion, and we therefore resort to the use of a lookup table. We test the algorithm for dosimetric computations in radiation therapy on a 2D CT scan of a lung patient. Using a simple particle model, our Python implementation of the algorithm is nearly 33 times faster than an equivalent kinetic simulation at the cost of a small modeling error.

Figures

Figures reproduced from arXiv: 2412.05063 by the authors.

Figure 1
Figure 1. Angular deflections at scattering events [ [PITH_FULL_IMAGE:figures/full_fig_p005_1.png] view at source ↗
Figure 2
Figure 2. Mean kinetic motion (left) and the dependence of the mean polar [PITH_FULL_IMAGE:figures/full_fig_p009_2.png] view at source ↗
Figure 3
Figure 3. Comparison of experimentally obtained mean kinetic motion (10.000 [PITH_FULL_IMAGE:figures/full_fig_p010_3.png] view at source ↗
Figures from the paper (5 more)
Figure 4
Figure 4. Figure 4: Lookup table for the variance in the x, y and z direction generated [PITH_FULL_IMAGE:figures/full_fig_p011_4.png]
Figure 5
Figure 5. Figure 5: Polar multiple scattering distribution θMS for a particle with energy Emax = 3.117 MeV, that has travelled ∆s = 0.1 centimetres in a medium with density ρ = 1.0 g/cm3 . The distribution is fitted with a log-normal distribution. 3.2.3 Kinetic-Diffusion-Rotation Monte Ca…
Figure 6
Figure 6. Figure 6: Speed-up of KDR compared to purely kinetic simulation for a particle [PITH_FULL_IMAGE:figures/full_fig_p015_6.png]
Figure 7
Figure 7. Figure 7: 2D CT scan of a lung and the computed dose distribution using [PITH_FULL_IMAGE:figures/full_fig_p016_7.png]
Figure 8
Figure 8. Figure 8: Pointwise relative error of the dose distributions. [PITH_FULL_IMAGE:figures/full_fig_p017_8.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

39 extracted references · 38 canonical work pages

  1. [1]

    Designing, benchmarking, and applying a Monte Carlo electron transport code

    Al-Beteri, A., and Raeside, D. Designing, benchmarking, and applying a Monte Carlo electron transport code. Computer Methods and Programs in Biomedicine 39, 3-4 (Apr. 1993), 147–167

    work page 1993

  2. [2]

    K., and Hwang, J

    Blitzstein, J. K., and Hwang, J. Introduction to probability, second edition ed. CRC Press, Boca Raton, 2019

    work page 2019

  3. [3]

    Hybrid Particle-Continuum Numerical Methods for Aerospace Applications, May 2012

    Boyd, I., and Deschenes, T. Hybrid Particle-Continuum Numerical Methods for Aerospace Applications, May 2012

    work page 2012

  4. [4]

    The Boltzmann equation and its applications

    Cercignani, C. The Boltzmann equation and its applications . No. 67 in Applied mathematical sciences. Springer, New York, NY Heidelberg, 1988

    work page 1988

  5. [5]

    X., Duggan, D

    Ding, G. X., Duggan, D. M., Coffey, C. W., Shokrani, P., and Cygler, J. E. First macro Monte Carlo based commercial dose calculation module for electron beam treatment planning—new issues for clinical consideration. Physics in Medicine and Biology 51 , 11 (June 2006), 2781–2799

    work page 2006

  6. [6]

    du Plessis, F. C. P., Willemse, C. A., Lotter, M. G., and Goedhals, L. The indirect use of CT numbers to establish material properties needed for Monte Carlo calculation of dose distributions in patients. MEDICAL PHYSICS 25, 7 (July 1998), 1195–1201. Num Pages: 7 Patent Number: 1 Place: Hoboken Publisher: Wiley Web of Science ID: WOS:000074805900014

    work page 1998

  7. [7]

    M., and Jones, G

    Flegal, J. M., and Jones, G. L. Batch means and spectral variance estimators in Markov chain Monte Carlo. The Annals of Statistics 38 , 2 (Apr. 2010)

    work page 2010

  8. [8]

    K., and Hauck, C

    Garrett, C. K., and Hauck, C. D. A Comparison of Moment Closures for Linear Kinetic Transport Equations: The Line Source Benchmark. Transport Theory and Statistical Physics 42 , 6-7 (Sept. 2013), 203–235

    work page 2013

Show all 39 references
  1. [9]

    GPUMCD: A new GPU-oriented Monte Carlo dose calculation platform

    Hissoiny, S., Ozell, B., Bouchard, H., and Despr ´es, P. GPUMCD: A new GPU-oriented Monte Carlo dose calculation platform. Medical Physics 38, 2 (Jan. 2011), 754–764

    2011

  2. [10]

    Review of fast Monte Carlo codes for dose calculation in radiation therapy treatment planning

    Jabbari, K. Review of fast Monte Carlo codes for dose calculation in radiation therapy treatment planning. Journal of Medical Signals & Sensors 1 , 1 (2011), 73

    2011

  3. [11]

    J., and Tanuma, S.Monte Carlo strategies for sim- ulations of electron backscattering from surfaces

    Jablonski, A., Powell, C. J., and Tanuma, S.Monte Carlo strategies for sim- ulations of electron backscattering from surfaces. Surface and Interface Analysis 37, 11 (Nov. 2005), 861–874

    2005

  4. [12]

    SU-D-218-06: Acceleration of Optical Photon Monte Carlo Simu- lations Using the Macro Monte Carlo Method

    Jacqmin, D. SU-D-218-06: Acceleration of Optical Photon Monte Carlo Simu- lations Using the Macro Monte Carlo Method. Medical Physics 39, 6Part3 (June 2012), 3623–3623

    2012

  5. [13]

    3D electron dose calculation using a Voxel based Monte Carlo algorithm (VMC)

    Kawrakow, I., Fippel, M., and Friedrich, K. 3D electron dose calculation using a Voxel based Monte Carlo algorithm (VMC). Medical Physics 23, 4 (Apr. 1996), 445–457

    1996

  6. [14]

    Kawrakow, I., Mainegra-Hing, E., Rogers, D. W. O., Tessier, F., and W alters, B. R. B. The EGSnrc Code System: Monte Carlo Simulation of Electron and Photon Transport

  7. [15]

    J., and Hoban, P

    Keall, P. J., and Hoban, P. W. Super-Monte Carlo: A 3-D electron beam dose calculation algorithm. Medical Physics 23 , 12 (Dec. 1996), 2023–2034. 23

    1996

  8. [16]

    J., and Hoban, P

    Keall, P. J., and Hoban, P. W. Superposition dose calculation incorporating Monte Carlo generated electron track kernels. Medical Physics 23, 4 (Apr. 1996), 479–485

    1996

  9. [17]

    M., Jhingran, A., and Court, L

    Kisling, K., Zhang, L., Simonds, H., F akie, N., Yang, J., McCarroll, R., Balter, P., Burger, H., Bogler, O., Howell, R., Schmeler, K., Mejia, M., Beadle, B. M., Jhingran, A., and Court, L. Fully Automatic Treatment Planning for External-Beam Radiation Therapy of Locally Advan...

    2019

  10. [18]

    A robust collision source method for rank adaptive dynamical low-rank approximation in radiation therapy, Nov

    Kusch, J., and Stammer, P. A robust collision source method for rank adaptive dynamical low-rank approximation in radiation therapy, Nov. 2021. arXiv:2111.07160 [cs, math]

    2021 arXiv

  11. [19]

    Models, Numerical Methods and Uncertainty Quantification for Ra- diation Therapy

    K¨upper, K. Models, Numerical Methods and Uncertainty Quantification for Ra- diation Therapy. PhD thesis, Aachen University, Nov. 2016

    2016

  12. [20]

    W., and Keller, J

    Larsen, E. W., and Keller, J. B. Asymptotic solution of neutron transport problems for small mean free paths. Journal of Mathematical Physics 15 , 1 (Jan. 1974), 75–81

    1974

  13. [21]

    Multilevel and Adjoint Monte Carlo Methods for Plasma Edge Neutral Particle Models

    Loevbak, E. Multilevel and Adjoint Monte Carlo Methods for Plasma Edge Neutral Particle Models . PhD thesis, KU Leuven, Jan. 2023

    2023

  14. [22]

    Accelerated simulation of Boltzmann-BGK equa- tions near the diffusive limit with asymptotic-preserving multilevel Monte Carlo

    Løvbak, E., and Samaey, G. Accelerated simulation of Boltzmann-BGK equa- tions near the diffusive limit with asymptotic-preserving multilevel Monte Carlo. SIAM Journal on Scientific Computing 45 , 4 (2023), A1862–A1889. Publisher: SIAM

    2023

  15. [23]

    Hilbert expansion based fluid models for kinetic equations describing neutral particles in the plasma edge of a fusion device

    Maes, V., Dekeyser, W., Koellermeier, J., Baelmans, M., and Samaey, G. Hilbert expansion based fluid models for kinetic equations describing neutral particles in the plasma edge of a fusion device. Physics of Plasmas 30 , 6 (June 2023), 063907

    2023

  16. [24]

    Advanced Monte Carlo simulation and estimation for kinetic neutral particles in the plasma edge of fusion reactors

    Mortier, B. Advanced Monte Carlo simulation and estimation for kinetic neutral particles in the plasma edge of fusion reactors . PhD thesis, KU Leuven, 2020

    2020

  17. [25]

    A Kinetic-Diffusion Asymptotic-Preserving Monte Carlo Algorithm for the Boltzmann-BGK Model in the Diffusive Scaling

    Mortier, B., Baelmans, M., and Samaey, G. A Kinetic-Diffusion Asymptotic-Preserving Monte Carlo Algorithm for the Boltzmann-BGK Model in the Diffusive Scaling. SIAM Journal on Scientific Computing 44 , 2 (Apr. 2022), A720–A744

    2022

  18. [26]

    Estimation as a post-processing step for random walk approximations of the Boltzmann-BGK model

    Mortier, B., Maes, V., and Samaey, G. Estimation as a post-processing step for random walk approximations of the Boltzmann-BGK model. Contributions to Plasma Physics 62 , 5-6 (June 2022)

    2022

  19. [27]

    Neuenschwander, H., and Born, E. J. A macro Monte Carlo method for electron beam dose calculations. Physics in Medicine and Biology 37 , 1 (Jan. 1992), 107–125

    1992

  20. [28]

    R., and Reckwerdt, P

    Neuenschwander, H., Mackie, T. R., and Reckwerdt, P. J. MMC-a high- performance Monte Carlo code for electron beam treatment planning. Physics in Medicine and Biology 40 , 4 (Apr. 1995), 543–574

    1995

  21. [29]

    PENELOPE 2018: A code system for Monte Carlo simulation of electron and photon transport: Workshop Proceedings, Barcelona, Spain, 28 January – 1 February 2019

    Nuclear Energy Agency. PENELOPE 2018: A code system for Monte Carlo simulation of electron and photon transport: Workshop Proceedings, Barcelona, Spain, 28 January – 1 February 2019 . PENELOPE: A code system for Monte Carlo simulation of electron and photon transport. OECD, Se...

    2018

  22. [30]

    Models and Numerical Methods for Time- and Energy-Dependent Particle Transport

    Olbrant, E. Models and Numerical Methods for Time- and Energy-Dependent Particle Transport. PhD thesis, R WTH Aachen University, Apr. 2012

    2012

  23. [31]

    G., and Hillen, T

    Othmer, H. G., and Hillen, T. The Diffusion Limit of Transport Equations Derived from Velocity-Jump Processes. SIAM Journal on Applied Mathematics 61, 3 (Jan. 2000), 751–775

    2000

  24. [32]

    Monte Carlo treatment planning for photon and electron beams

    Reynaert, N., V an Der Marck, S., Schaart, D., V an Der Zee, W., V an Vliet-Vroegindeweij, C., Tomsej, M., Jansen, J., Heijmen, B., Coghe, M., and De W agter, C. Monte Carlo treatment planning for photon and electron beams. Radiation Physics and Chemistry 76 , 4 (Apr. 2007), 643–686

    2007

  25. [33]

    J., Jhingran, A., Kisling, K., Cardenas, C., Simonds, H., and Court, L

    Rhee, D. J., Jhingran, A., Kisling, K., Cardenas, C., Simonds, H., and Court, L. Automated Radiation Treatment Planning for Cervical Cancer. Sem- inars in Radiation Oncology 30 , 4 (Oct. 2020), 340–347

    2020

  26. [34]

    Seltzer, S. M. Electron-photon Monte Carlo calculations: The ETRAN code. International Journal of Radiation Applications and Instrumentation. Part A. Applied Radiation and Isotopes 42 , 10 (Jan. 1991), 917–941

    1991

  27. [35]

    V assiliev, O. N. Monte Carlo Methods for Radiation Transport . Biological and Medical Physics, Biomedical Engineering. Springer International Publishing, Cham, 2017

    2017

  28. [36]

    R., N., and Yoshihito, N

    W. R., N., and Yoshihito, N. The EGS4 Code System: Solution of gamma-ray and electron transport problems

  29. [37]

    B., and Gnanadesikan, R

    Wilk, M. B., and Gnanadesikan, R. Probability Plotting Methods for the Analysis of Data. Biometrika 55 , 1 (Mar. 1968), 1

    1968

  30. [38]

    Willems, K. electronTransportCode: Repository containing implementations of kinetic, kinetic-diffusion and kinetic-diffusion-rotation particle tracing algorithms for radiation therapy., May 2023

    2023

  31. [39]

    Particle, Fluid and Hybrid Numerical Methods for Radiation Ther- apy

    Willems, K. Particle, Fluid and Hybrid Numerical Methods for Radiation Ther- apy. Master’s thesis, KU Leuven, Leuven, Belgium, 2023. 25

    2023

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.