Pith. sign in

REVIEW 5 major objections 5 minor 37 references

Muographic Image Upsampling with Machine Learning for Built Infrastructure Applications

T0 review · 5 major / 5 minor · reviewed 2026-08-09 · deepseek-v4-flash

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

Pith's one-line read A one-day muography scan of concrete can be upsampled to the image quality of a three-week scan, and a second network can erase the smearing artifacts that hide rebar and tendon ducts.

desk verdict A clean simulation-only proof of concept for GAN-based muography upsampling, with the natural caveat that the headline numbers remain unverified until someone runs it on real detector data. read the letter →

arxiv 2502.02624 v2 pith:MTXENHUH submitted 2025-02-04 eess.IV cs.CV

William O'Donnell , David Mahon , Guangliang Yang , Simon Gardner This is my paper · ORCID
classification eess.IVcs.CV
keywords muographynon-destructivetestingmachinelearningimageprocessingupsamplingsemanticsegmentationsimulationcosmicrays
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 tries to establish that machine learning can make muography—a slow, noisy imaging method that uses cosmic-ray muons to see inside concrete—usable for inspecting bridges and other infrastructure. The authors train a conditional Wasserstein generative adversarial network with gradient penalty to turn severely undersampled scans into clean images, reporting that a simulated one-day scan can be upsampled to match the perceptual quality (SSIM) of a 21-day scan and the noise level (PSNR) of a 31-day scan. A second segmentation network trained on the same simulated data then shows that the cleaned images preserve and sometimes improve the visibility of rebar grids and tendon ducts, with Dice scores of 0.8174 and 0.8663 on those classes. The segmentation network also learns to ignore the z-direction smearing that normally obscures object boundaries in muography. If the results transfer from simulation to real detectors, inspection times could drop from weeks to days.

What carries the argument

The load-bearing mechanism is a paired-image conditional generative adversarial network: a generator with an encoder–decoder structure and skip connections turns a low-sampling-time muography slice into a 100-day image, while a discriminator plus a per-pixel L1 loss drives the output toward the simulated ground truth. A Wasserstein distance with gradient penalty stabilises training, and each input is re-sampled at a random equivalent sampling time each epoch so the model learns to upsample across the full range. A second cWGAN-GP is trained for five-class semantic segmentation on 100-day images only, using a combined cross-entropy and Dice loss, so it can evaluate the upsampler's feature-level effects without being biased by upsampled outputs. The training images come from Monte Carlo simulation of a scintillating-fibre muon tracker, with muon tracks reconstructed by the point-of-closest-approach algorithm and voxelised at 2 mm into 500×500 X–Y slices.

What would settle it

Take a real reinforced-concrete specimen, run a one-day muography acquisition and then continue the same scan to 21 or 31 days, apply the trained upsampler to the one-day data, and compare the upsampled image to the actual long-exposure image using SSIM and PSNR; if the simulated equivalence does not reproduce on real detector data, the central claim collapses.

Watch

Extended reading notes

Core claim

On the paper's own terms, the central discovery is that a two-stage deep-learning pipeline can substantially undo the two main defects of muon scattering tomography of concrete: sparse statistics from short acquisition times and smearing along the detector-normal direction from the inverse imaging problem. Trained on a large Monte Carlo dataset of reinforced concrete blocks with known geometry, the upsampling model maps images corresponding to 1–99 days of muon exposure onto a 100-day ground truth. Averaged over the 6900-image test set, one-day inputs reach SSIM 0.88 and PSNR 37 dB after upsampling, matching a 21-day raw image perceptually and a 31-day raw image in noise; improvement shrinks as sampling time grows and converges near 50–85 days. The segmentation model, trained only on 100-day images, gives Dice coefficients of 0.8174 for rebar grids and 0.8663 for tendon ducts, and it learns to classify z-smear shadows as background, removing some of the worst artifacts. The authors frame the work as a step toward making muography practical for reinforced-concrete non-destructive evaluation.

Load-bearing premise

The load-bearing premise is that Monte Carlo simulations of a muon tracker looking at concrete blocks are faithful enough to real cosmic-ray muography that a model trained on simulated 100-day targets will behave the same way when applied to real scans of reinforced concrete; the authors themselves note in Section 4.3 that validation on real data is future work.

Editorial extensions

If this is right

  • On the 6900-image test set, upsampled one-day inputs reach an average SSIM of 0.88 and PSNR of 37 dB, matching the perceptual quality of raw 21-day images and the noise level of raw 31-day images.
  • Upsampling improves segmentation most for tendon ducts and rebar at low sampling times, with Dice differences shrinking to near zero as equivalent sampling time reaches 50–70 days.
  • The segmentation network learns to ignore z-direction smearing shadows in X–Y slices, so it can report rebar at its true location even when the muography image shows a shadowed grid.
  • The benefit of upsampling is feature-dependent: thin 8–10 mm rebar can be partially washed out between 20 and 85 days, and air voids remain poorly detected, with Dice 0.1265 on 100-day inputs.
  • For PSNR, the upsampler converges with raw inputs only around 80–85 days, because it smooths away the exact pixel-level noise fluctuations present in the 100-day ground truth.

Reading between the lines

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

  • If the simulation-to-real transfer holds, the same upsampler could shorten routine muography inspections of concrete from weeks to a single day, and the segmentation output could be used directly for locating rebar and tendon ducts.
  • The reported 'one-day equals 21/31 days' equivalence is measured against a simulated 100-day target; on real data the equivalence could shift, so the fairest check is to compare the upsampled one-day reconstruction with a genuinely long real acquisition of the same block.
  • Because the segmentation model already removes z-smear shadows, a single network trained directly on geometry ground truths might do both denoising and feature location, avoiding the small mismatch the authors observe between upsampled outputs and their training distribution.
  • A class-weighted loss would likely improve the very low air-void Dice score (0.1265), which currently limits the method as a defect-detection tool despite its success on rebar and ducts.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

5 major / 5 minor

Summary. This manuscript applies a conditional Wasserstein GAN with gradient penalty (cWGAN-GP) to upsample undersampled cosmic-ray muography images of reinforced concrete, and a second cWGAN-GP for semantic segmentation of features such as rebar grids and tendon ducts. Training and evaluation use a large Geant4/EcoMug simulation dataset of 700 concrete block designs with equivalent sampling times from 1 to 100 days. The authors report that 1-day upsampled images reach SSIM and PSNR levels equivalent to 21-day and 31-day unaltered images, and that segmentation Dice coefficients improve for several structural classes, with the segmentation model also learning to ignore z-plane smearing artifacts. The paper explicitly acknowledges in Section 4.3 that testing on real-world data is future work, but the abstract and conclusions present the headline results as demonstrated capabilities.

Significance. If the simulation-based results transfer to real detectors, the work offers a practical route to reducing muography acquisition times and automating feature detection in reinforced concrete. Strengths of the paper include a large and varied paired simulation dataset, the use of semantic segmentation as an interpretable evaluation tool, and candid discussion of limitations, with the authors identifying the need for real-data validation. The main weakness is that all headline quantities—the 21/31-day equivalence, Dice improvements, and smearing removal—are computed entirely within the Geant4/EcoMug simulation, with no real scans, no non-learning baselines, and no error bars; the claims are therefore currently simulation-based predictions rather than established capabilities.

major comments (5)
  1. [Abstract and Section 4.3] The paper overstates simulation-based results as demonstrated capabilities. The abstract states that the results 'demonstrate significant improvements in both acquisition speed and image quality,' and the conclusions similarly report the 21-day SSIM and 31-day PSNR equivalences as findings. However, Section 4.3 concedes that 'it is important to verify these models on real-world data' and that testing on real data is future work. All metrics are computed against simulated 100-day targets generated by the same Geant4/EcoMug model. The headline claims should be explicitly framed as simulation-based predictions, with the external-validity caveat stated in the abstract and conclusions, not only in a later future-work paragraph.
  2. [Section 3.1, Figure 1] No non-learning baseline is provided, so the cWGAN-GP is not shown to be responsible for the reported improvements. The equivalence claims (1-day upsampled SSIM matching 21-day unaltered input, and PSNR matching 31-day input) are relative only to the unaltered simulated inputs and the 100-day ground truth. A simple classical denoiser or smoother (e.g., Gaussian or median filtering, non-local means) applied to the 1-day inputs would provide a fair comparison and would test whether the gains are specific to the learned upsampling. Without such a baseline, the contribution of the GAN architecture is not isolated.
  3. [Section 3.2, Figures 3 and 4] The segmentation model is described as an 'independent evaluation metric,' but it is not independent of the simulation pipeline. It is trained on 100-day simulated muography images with geometric ground truths derived from the same Monte Carlo dataset that generated the upsampling training data. Dice improvements on upsampled images may therefore reflect that upsampled outputs are closer to the 100-day segmentation training distribution, rather than that true feature information has been recovered. Relatedly, the claim that segmentation 'mitigate[s] or entirely remove[s] z-plane smearing artifacts' is too strong: the segmentation model learns to ignore certain artifacts because its training labels contain no smearing, but the underlying muographic images are not deconvolved. Please rephrase the smearing-removal claim as a learned artifact-classification effect rather than an image-reconstruction result.
  4. [Section 2.2, Eq. (6)] Equation (6) is technically incorrect as written: it places the gradient penalty term in the generator loss, but in WGAN-GP the gradient penalty regularizes the discriminator (critic) and is part of the critic's loss, not the generator's. The text preceding Eq. (6) states that 'the generator's loss function, LG, is updated to include the gradient penalty,' which would not correspond to the standard WGAN-GP training procedure. Please correct the equation and the associated text, or clarify if a non-standard objective was used and explain why.
  5. [Sections 3.1 and 3.2, Figures 1 and 3] The paper reports averages over the 6900-image test set without error bars, confidence intervals, or statistical tests. Point estimates such as '1-day upsampled images exhibited SSIM and PSNR scores of 0.88 and 37 dB' and the Dice differences in Figure 3 cannot be assessed for significance. Reporting standard deviations or confidence intervals across the test images, and where relevant a paired significance test for Dice improvements, would substantially strengthen the claims and allow comparison with future work.
minor comments (5)
  1. [Section 2, first paragraph after the sample list] The dataset size description is internally inconsistent: the text says '70,000 500 × 500 images, each with 100 different versions,' which would total 7,000,000 images, but 700 unique blocks times 100 days yields 70,000 images. Please clarify the number of unique geometries, the number of cumulative-day images per geometry, and the total dataset size.
  2. [Section 2.2] The values of the hyperparameters λpixel and λGP are not reported, and the learning-rate schedule is described only qualitatively ('reduced ... by an order of magnitude every 25 epochs'). Providing these values would improve reproducibility.
  3. [Section 2.1.2] There is a typo, 'sampled from from that of,' and other minor grammatical issues throughout. A thorough proofreading pass is recommended.
  4. [Section 4.2, air void discussion] The discussion attributes the low air-void Dice (0.1265) to class imbalance, but no class-frequency statistics are given. A table of pixel percentages per class in the training/test sets would make the imbalance argument quantitative.
  5. [Section 3, Figures 1 and 3] The figure captions and axis labels are occasionally redundant (e.g., repeating 'Dice-Sørensen Coefficient' on the left of all four subplots in Figure 3) and the Dice-difference subplots use inconsistent y-axis scales. Harmonizing the axes would ease cross-class comparison.

Circularity Check

3 steps flagged · score 6.0 of 10

The segmentation 'smearing removal' is the training objective itself, and the upsampling quality metrics are computed against the same simulated 100-day targets used for training, making the headline equivalence numbers and Dice improvements partially self-confirming.

  1. self definitional [Section 4.2, 'A Solution to Smearing Artefacts' (also Abstract and Section 3.2)]
    "This is due to the segmentation model being trained to reproduce geometric ground truths that were absent of any smearing artefacts, learning the ability to accurately differentiate between shadows and 'true' objects despite a lack of information along the z axis."

    The segmentation ground truths are generated directly from the simulated object geometry, which by construction contains no z-plane smearing. A model trained to output those labels will necessarily produce unsmeared segmentation maps. Presenting this as an 'unexpected capability to mitigate—and in some cases entirely remove—z-plane smearing artifacts' restates the supervised training target rather than reporting an independent discovery. The paper's own sentence identifies the mechanism, confirming that the smearing removal is encoded in the labels before training begins.

  2. fitted input called prediction [Section 2.2 (training objective) and Section 3.1 (evaluation methodology)]
    "The upsampling model generates outputs ˆy that approximate the expected outputs y of 100 days of sampling. ... We averaged the SSIM and PSNR metrics of the 6900 images in the test dataset—calculated with respect to the 100-day ground truth—for both the inputs and outputs of the model at each sampling time."

    The upsampler's loss is the L1 distance between its output and the 100-day simulated image (Equation (7)), and the reported SSIM and PSNR are computed against those same 100-day simulated images. The headline result—'1-day sampled images matched the perceptual qualities of a 21-day image' and 'noise improvement equivalent to 31 days'—is therefore a measure of how well the model optimized its training objective on the held-out test split. It is not a prediction validated against independent real data; the equivalence numbers quantify success at the exact task the model was fitted to perform.

1 more flagged steps
  1. other [Section 3.2, first paragraph]
    "The segmentation model serves as an independent evaluation metric to assess the performance of the upsampling model. To minimise potential bias and maintain objectivity, this model was trained exclusively on high sampling time (100-day) muography images, deliberately excluding upsampler outputs from the training process."

    The claimed independence is only from the upsampler outputs, not from the upsampler's training target. The segmentation model was trained on the same 100-day simulated muography images and the same simulation-derived geometric labels that define the upsampler's ground truth. Consequently, the segmentation model is not an external benchmark: both models are optimized toward the same simulated target distribution, so Dice improvements on upsampled inputs partly reflect that the upsampler was trained to reproduce exactly the images the segmenter was trained to label. The evaluation loop is closed within the simulation.

full rationale

The paper is transparent that all training and evaluation occur within a Geant4/EcoMug simulation, and Section 4.3 explicitly states that 'it is important to verify these models on real-world data' and that real-data testing is future work. That simulation-to-real gap is an external validity risk rather than circularity by itself. The circularity score is elevated because two headline claims reduce to their own inputs: (1) the segmentation model's removal of z-smearing is the direct consequence of training on geometric ground truths that contain no smearing, and (2) the upsampling model's '21-day equivalence' and '31-day noise improvement' are SSIM/PSNR scores measured against the same 100-day simulated images used as the training target. The paper's self-citations (e.g., detector description [30], review [22], prior muography application [1]) are contextual and not load-bearing. The upsampling model does show genuine generalization across held-out simulated block designs, so the circularity is partial rather than total.

Assumptions & free parameters 4 free parameters · 4 assumptions · 0 invented entities

No new physical entities are introduced; the air voids and 'unknown' objects are simulated contents, not new postulates. The central claim rests on simulation realism, PoCA reconstruction fidelity, and the choice of the 100-day simulated target as ground truth.

free parameters (4)
  • lambda_pixel (L1 weight in upsampling generator loss) = not reported
    Balances adversarial loss against pixelwise L1 loss in Equation (6); chosen by hand, not fitted, and affects the trade-off between sharpness and fidelity.
  • lambda_GP (gradient penalty weight) = not reported
    Controls the Lipschitz penalty in WGAN-GP (Equation 5); a standard hyperparameter chosen by hand.
  • learning rates and batch size = 0.01 generator, 0.001 discriminator, batch size 48
    Chosen hyperparameters with no reported tuning or sensitivity analysis; they affect convergence and output quality.
  • 100-day maximum equivalent sampling time = 100 days
    Chosen arbitrarily as the ground truth target; the paper's quality equivalences are defined relative to this target, and a different target would shift the reported curves.
assumptions (4)
  • domain assumption Geant4 with EcoMug faithfully simulates cosmic-ray muon flux, scattering, and detector response for concrete imaging.
    All training and test data are simulated; the paper's external validity rests entirely on this assumption, acknowledged in Section 4.3 as needing real-world validation.
  • domain assumption PoCA reconstruction with a Gaussian approximation to multiple Coulomb scattering is adequate for generating training images.
    Equations (1) and (2) and Section 1.1 invoke the Rossi formula and the point-of-closest-approach algorithm; systematic errors from this approximation are not quantified.
  • domain assumption The simulated geometric object labels are a valid ground truth for segmentation.
    The segmentation model is trained on labels derived from the same simulation geometry, not from independently measured truth, so label and image errors are correlated.
  • domain assumption Supervised training on paired simulated images transfers to real muography data.
    The paper states in Section 4.3 that testing on real data from Niederleithinger et al. is future work; without this transfer, the acquisition-speed claims apply only to simulation.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Muographic Image Upsampling with Machine Learning for Built Infrastructure Applications." pith.science (2026). https://pith.science/paper/MTXENHUH

@misc{pith2026250202624,
  author       = {Pith},
  title        = {Pith review of: Muographic Image Upsampling with Machine Learning for Built Infrastructure Applications},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/MTXENHUH}},
  note         = {Machine review of arXiv:2502.02624}
}
read the original abstract

The civil engineering industry faces a critical need for innovative non-destructive evaluation methods, particularly for ageing critical infrastructure, such as bridges, where current techniques fall short. Muography, a non-invasive imaging technique, constructs three-dimensional density maps by detecting interactions of naturally occurring cosmic-ray muons within the scanned volume. Cosmic-ray muons provide deep penetration and inherent safety due to their high momenta and natural source. However, the technology's reliance on this source results in constrained muon flux, leading to prolonged acquisition times, noisy reconstructions and image interpretation challenges. To address these limitations, we developed a two-model deep learning approach. First, we employed a conditional Wasserstein generative adversarial network with gradient penalty (cWGAN-GP) to perform predictive upsampling of undersampled muography images. Using the Structural Similarity Index Measure (SSIM), 1-day sampled images matched the perceptual qualities of a 21-day image, while the Peak Signal-to-Noise Ratio (PSNR) indicated noise improvement equivalent to 31 days of sampling. A second cWGAN-GP model, trained for semantic segmentation, quantitatively assessed the upsampling model's impact on concrete sample features. This model achieved segmentation of rebar grids and tendon ducts, with Dice-S{\o}rensen accuracy coefficients of 0.8174 and 0.8663. Notably, it could mitigate or remove z-plane smearing artifacts caused by muography's inverse imaging problem. Both models were trained on a comprehensive Geant4 Monte-Carlo simulation dataset reflecting realistic civil infrastructure scenarios. Our results demonstrate significant improvements in acquisition speed and image quality, marking a substantial step toward making muography more practical for reinforced concrete infrastructure monitoring applications.

Figures

Figures reproduced from arXiv: 2502.02624 by the authors.

Figure 1
Figure 1. Average Structural Similarity Index Measure (SSIM) and Peak Signal-to-Noise Ratio (PSNR) metrics of the test dataset as a function of equivalent sampling time. The original dataset performance is indicated by blue circular points, where the upsampled outputs of the model are indicated by red triangular points. Demonstrating these drastic improvements, [PITH_FULL_IMAGE:figures/full_fig_p010_1.png] view at source ↗
Figure 2
Figure 2. Five examples of 1-day input images from the test dataset (left column) and their corre￾sponding output from the upsampling model (middle column). The 100-day ground truth image for each example is shown for comparison (right column) [PITH_FULL_IMAGE:figures/full_fig_p011_2.png] view at source ↗
Figure 3
Figure 3. Average Dice coefficient of the test dataset, calculated at each equivalent sampling time, for each of the four segmentation object labels (rebar grid, tendon duct, void, unknown). Performance of the original input dataset is shown by blue circular points, where upsampled outputs are displayed with red triangular points. The relative improvement of the upsampling model for a particular label is indicated by Dice dif… view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: A single X–Y plane image slice for different equivalent sampling times: (a) one day, (b) five days, (c) 10 days, (d) 20 days, (e) 40 days, (f) 60 days, (g) 80 days, (h) ground truth (100-day image for top panels, geometry truth for bottom panels). These eight image ver…
Figure 5
Figure 5. Figure 5: A single X–Z plane vertical image slice for different equivalent sampling times: (a) one day, (b) five days, (c) 10 days, (d) 20 days, (e) 40 days, (f) 60 days, (g) 80 days, (h) ground truth (100-day image for the top two panels, geometry truth for the bottom two panel…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

37 extracted references · 12 canonical work pages

  1. [1]

    First-of-a-kind muography for nuclear waste characterization

    Mahon, D.; Clarkson, A.; Gardner, S.; Ireland, D.; Jebali, R.; Kaiser, R.; Ryan, M.; Shearer, C.; Yang, G. First-of-a-kind muography for nuclear waste characterization. Philos. Trans. R. Soc. A Math. Phys. Eng. Sci. 2019, 377, 20180048. https://doi.org/10.1098/ rsta.2018.0048. Particles 2025, 1, 0 20 of 21

    arXiv 2019

  2. [2]

    Cosmic-Ray Tomography for Border Security

    Barnes, S.; Georgadze, A.; Giammanco, A.; Kiisk, M.; Kudryavtsev, V .A.; Lagrange, M.; Pinto, O.L. Cosmic-Ray Tomography for Border Security. Instruments 2023, 7, 13. https://doi.org/10.3390/instruments7010013

    doi:10.3390/instruments7010013 2023

  3. [3]

    Applications of Muography to the Industrial Sector.J

    Martinez Ruiz Del Arbol, P .; Orio Alonso, A.; Diez, C.; Garcia, P . Applications of Muography to the Industrial Sector.J. Adv. Instrum. Sci. 2022, 2022. https://doi.org/10.31526/JAIS.2022.267

    doi:10.31526/jais.2022.267 2022

  4. [4]

    Cosmic ray tracking to monitor the stability of historical buildings: A feasibility study

    Bonomi, G.; Caccia, M.; Donzella, A.; Pagano, D.; Villa, V .; Zenoni, A. Cosmic ray tracking to monitor the stability of historical buildings: A feasibility study. Meas. Sci. Technol. 2019, 30, 045901. https://doi.org/10.1088/1361-6501/ab00d7

    work page doi:10.1088/1361-6501/ab00d7 2019

  5. [5]

    Muon Tomography of the Interior of a Reinforced Concrete Block: First Experimental Proof of Concept

    Niederleithinger, E.; Gardner, S.; Kind, T.; Kaiser, R.; Grunwald, M.; Yang, G.; Redmer, B.; Waske, A.; Mielentz, F.; Effner, U.; et al. Muon Tomography of the Interior of a Reinforced Concrete Block: First Experimental Proof of Concept. J. Nondestruct. Eval. 2021, 40, 65. https://doi.org/10.1007/s10921-021-00797-3

    work page doi:10.1007/s10921-021-00797-3 2021

  6. [6]

    Statistical Reconstruction for Cosmic Ray Muon Tomography

    Schultz, L.J.; Blanpied, G.S.; Borozdin, K.N.; Fraser, A.M.; Hengartner, N.W.; Klimenko, A.V .; Morris, C.L.; Orum, C.; Sossong, M.J. Statistical Reconstruction for Cosmic Ray Muon Tomography. IEEE Trans. Image Process. 2007, 16, 1985–1993. https: //doi.org/10.1109/TIP .2007.901239

    arXiv 2007

  7. [7]

    Cosmic rays for imaging cultural heritage objects

    Giammanco, A.; Moussawi, M.A.; Boone, M.; Kock, T.D.; Roy, J.D.; Huysmans, S.; Kumar, V .; Lagrange, M.; Tytgat, M. Cosmic rays for imaging cultural heritage objects. 2024, arXiv:2405.10417. https://doi.org/10.48550/arXiv.2405.10417

    work page Pith review arXiv doi:10.48550/arxiv.2405.10417 2024

  8. [8]

    Muography for Inspection of Civil Structures

    Das, S.; Tripathy, S.; Jagga, P .; Bhattacharya, P .; Majumdar, N.; Mukhopadhyay, S. Muography for Inspection of Civil Structures. Instruments 2022, 6, 77. https://doi.org/10.3390/instruments6040077

    work page doi:10.3390/instruments6040077 2022

Show all 37 references
  1. [9]

    MedGAN: Medical image translation using GANs

    Armanious, K.; Jiang, C.; Fischer, M.; Küstner, T.; Hepp, T.; Nikolaou, K.; Gatidis, S.; Yang, B. MedGAN: Medical image translation using GANs. Comput. Med. Imaging Graph. 2020, 79, 101684. https://doi.org/10.1016/j.compmedimag.2019.101684

    2020

  2. [10]

    A Review of Deep Learning in Medical Imaging: Imaging Traits, Technology Trends, Case Studies with Progress Highlights, and Future Promises

    Zhou, S.K.; Greenspan, H.; Davatzikos, C.; Duncan, J.S.; Van Ginneken, B.; Madabhushi, A.; Prince, J.L.; Rueckert, D.; Summers, R.M. A Review of Deep Learning in Medical Imaging: Imaging Traits, Technology Trends, Case Studies with Progress Highlights, and Future Promises. Pro...

    2021

  3. [11]

    Segment anything in medical images

    Ma, J.; He, Y.; Li, F.; Han, L.; You, C.; Wang, B. Segment anything in medical images. Nat. Commun. 2024, 15, 654. https: //doi.org/10.1038/s41467-024-44824-z

    2024 doi

  4. [12]

    Geant4—A simulation toolkit

    Agostinelli, S.; Allison, J.; Amako, K.; Apostolakis, J.; Araujo, H.; Arce, P .; Asai, M.; Axen, D.; Banerjee, S.; Barrand, G.; et al. Geant4—A simulation toolkit. Nucl. Instruments Methods Phys. Res. Sect. A Accel. Spectrometers Detect. Assoc. Equip. 2003, 506, 250–303. https...

    2003 doi

  5. [13]

    Geant4 developments and applications

    Allison, J.; Amako, K.; Apostolakis, J.; Araujo, H.; Arce Dubois, P .; Asai, M.; Barrand, G.; Capra, R.; Chauvie, S.; Chytracek, R.; et al. Geant4 developments and applications. IEEE Trans. Nucl. Sci. 2006, 53, 270–278. https://doi.org/10.1109/TNS.2006.869826

    2006

  6. [14]

    Recent developments in Geant4

    Allison, J.; Amako, K.; Apostolakis, J.; Arce, P .; Asai, M.; Aso, T.; Bagli, E.; Bagulya, A.; Banerjee, S.; Barrand, G.; et al. Recent developments in Geant4. Nucl. Instruments Methods Phys. Res. Sect. A Accel. Spectrometers Detect. Assoc. Equip. 2016, 835, 186–225. https://d...

    2016 doi

  7. [15]

    EcoMug: An Efficient COsmic MUon Generator for cosmic-ray muon applications

    Pagano, D.; Bonomi, G.; Donzella, A.; Zenoni, A.; Zumerle, G.; Zurlo, N. EcoMug: An Efficient COsmic MUon Generator for cosmic-ray muon applications. Nucl. Instruments Methods Phys. Res. Sect. A Accel. Spectrometers Detect. Assoc. Equip. 2021, 1014, 165732. https://doi.org/10....

    2021

  8. [16]

    Muon tomography for railway tunnel imaging

    Thompson, L.F.; Stowell, J.P .; Fargher, S.J.; Steer, C.A.; Loughney, K.L.; O’Sullivan, E.M.; Gluyas, J.G.; Blaney, S.W.; Pidcock, R.J. Muon tomography for railway tunnel imaging. Phys. Rev. Res. 2020, 2, 023017. https://doi.org/10.1103/PhysRevResearch.2.02 3017

    2020 doi

  9. [17]

    Structural health monitoring of sabo check dams with cosmic-ray muography

    Oláh, L.; Tanaka, H.K.; Mori, T.; Sakatani, Y.; Varga, D. Structural health monitoring of sabo check dams with cosmic-ray muography. iScience 2023, 26, 108019. https://doi.org/10.1016/j.isci.2023.108019

    2023

  10. [18]

    Investigation of the Unit-1 nuclear reactor of Fukushima Daiichi by cosmic muon radiography

    Fujii, H.; Hara, K.; Hayashi, K.; Kakuno, H.; Kodama, H.; Nagamine, K.; Sato, K.; Kim, S.H.; Suzuki, A.; Sumiyoshi, T.; et al. Investigation of the Unit-1 nuclear reactor of Fukushima Daiichi by cosmic muon radiography. Prog. Theor. Exp. Phys. 2020, 2020, 043C02. https://doi.o...

    2020 doi

  11. [19]

    Radiographic visualization of magma dynamics in an erupting volcano

    Tanaka, H.K.M.; Kusagaya, T.; Shinohara, H. Radiographic visualization of magma dynamics in an erupting volcano. Nat. Commun. 2014, 5, 3381. https://doi.org/10.1038/ncomms4381

    2014 doi

  12. [20]

    Discovery of a big void in Khufu’s Pyramid by observation of cosmic-ray muons

    Morishima, K.; Kuno, M.; Nishio, A.; Kitagawa, N.; Manabe, Y.; Moto, M.; Takasaki, F.; Fujii, H.; Satoh, K.; Kodama, H.; et al. Discovery of a big void in Khufu’s Pyramid by observation of cosmic-ray muons. Nature 2017, 552, 386–390. https: //doi.org/10.1038/nature24647

    2017 doi

  13. [21]

    Cosmic-Ray Theory

    Rossi, B.; Greisen, K. Cosmic-Ray Theory. Rev. Mod. Phys. 1941, 13, 240–309. https://doi.org/10.1103/RevModPhys.13.240

    1941 doi

  14. [22]

    Novel muon imaging techniques

    Yang, G.; Clarkson, T.; Gardner, S.; Ireland, D.; Kaiser, R.; Mahon, D.; Jebali, R.A.; Shearer, C.; Ryan, M. Novel muon imaging techniques. Philos. Trans. R. Soc. A Math. Phys. Eng. Sci. 2019, 377, 20180062. https://doi.org/10.1098/rsta.2018.0062

    2019

  15. [23]

    Performance evaluation of cosmic ray muon trajectory estimation algorithms

    Ughade, R.; Bae, J.; Chatzidakis, S. Performance evaluation of cosmic ray muon trajectory estimation algorithms. AIP Adv. 2023, 13, 125301. https://doi.org/10.1063/5.0174796

    2023 doi

  16. [24]

    Cosmic Ray Muon Radiography

    Schultz, L.J. Cosmic Ray Muon Radiography. Ph.D. Thesis, Portland State University, Portland, OR, USA, 2003

    2003

  17. [25]

    & Feldman, G

    Sobel, I. & Feldman, G. An Isotropic 3 × 3 Image Gradient Operator. Presentation at Stanford A.I. Project 1968 2014. Available online: http://dx.doi.org/10.13140/RG.2.1.1912.4965 (accessed on 20 January 2025). Particles 2025, 1, 0 21 of 21

    1968

  18. [26]

    U-Net: Convolutional Networks for Biomedical Image Segmentation

    Ronneberger, O.; Fischer, P .; Brox, T. U-Net: Convolutional Networks for Biomedical Image Segmentation. In Medical Image Computing and Computer-Assisted Intervention—MICCAI 2015 ; Series Title: Lecture Notes in Computer Science; Navab, N., Hornegger, J., Wells, W.M., Frangi, ...

    2015 doi

  19. [27]

    U-Net and Its Variants for Medical Image Segmentation: A Review of Theory and Applications

    Siddique, N.; Paheding, S.; Elkin, C.P .; Devabhaktuni, V . U-Net and Its Variants for Medical Image Segmentation: A Review of Theory and Applications. IEEE Access 2021, 9, 82031–82057. https://doi.org/10.1109/ACCESS.2021.3086020

    2021

  20. [28]

    Measures of the Amount of Ecologic Association Between Species

    Dice, L.R. Measures of the Amount of Ecologic Association Between Species. Ecology 1945, 26, 297–302. https://doi.org/10.2307/ 1932409

    1945

  21. [29]

    A method of establishing groups of equal amplitude in plant sociology based on similarity of species and its application to analyses of the vegetation on Danish commons

    Sørensen, T. A method of establishing groups of equal amplitude in plant sociology based on similarity of species and its application to analyses of the vegetation on Danish commons. K. Dan. Vidensk. Selsk. 1948, 5, 1–34

    1948

  22. [30]

    The design and performance of a scintillating-fibre tracker for the cosmic-ray muon tomography of legacy nuclear waste containers

    Clarkson, A.; Hamilton, D.; Hoek, M.; Ireland, D.; Johnstone, J.; Kaiser, R.; Keri, T.; Lumsden, S.; Mahon, D.; McKinnon, B.; et al. The design and performance of a scintillating-fibre tracker for the cosmic-ray muon tomography of legacy nuclear waste containers. Nucl. Instrum...

    2014 doi

  23. [31]

    Air void clustering in concrete and its effect on concrete strength.Int

    Sun, W.; Wang, K.; Taylor, P .C.; Wang, X. Air void clustering in concrete and its effect on concrete strength.Int. J. Pavement Eng. 2022, 23, 5127–5141. https://doi.org/10.1080/10298436.2021.2000986

    2022

  24. [32]

    Image-to-Image Translation with Conditional Adversarial Networks

    Isola, P .; Zhu, J.Y.; Zhou, T.; Efros, A.A. Image-to-Image Translation with Conditional Adversarial Networks. InProceedings of the 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2017; pp. 5967–5976. https://doi.org/10.1109/CVPR.20 17.632

    2017 doi

  25. [33]

    Wasserstein GAN

    Arjovsky, M.; Chintala, S.; Bottou, L. Wasserstein GAN. In Proceedings of the 34th International Conference on Machine Learning, 2017, arXiv:1701.07875v3. https://doi.org/10.48550/ARXIV .1701.07875

    2017 doi arXiv

  26. [34]

    Improved Training of Wasserstein GANs

    Gulrajani, I.; Ahmed, F.; Arjovsky, M.; Dumoulin, V .; Courville, A. Improved Training of Wasserstein GANs. InProceedings of the 31st International Conference on Neural Information Processing Systems, 2017, arXiv:1704.00028v3. https://doi.org/10.48550/ARXIV . 1704.00028

    2017 doi arXiv

  27. [35]

    Image Quality Assessment: From Error Visibility to Structural Similarity

    Wang, Z.; Bovik, A.; Sheikh, H.; Simoncelli, E. Image Quality Assessment: From Error Visibility to Structural Similarity. IEEE Trans. Image Process. 2004, 13, 600–612. https://doi.org/10.1109/TIP .2003.819861

    2004

  28. [36]

    Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentation

    Chen, L.C.; Zhu, Y.; Papandreou, G.; Schroff, F.; Adam, H. Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentation. In European Conference on Computer Vision, 2018, arXiv:1802.02611. https://doi.org/10.48550/arXiv.1802.0 2611

    2018 doi arXiv

  29. [37]

    Attention Is All You Need

    Vaswani, A.; Shazeer, N.; Parmar, N.; Uszkoreit, J.; Jones, L.; Gomez, A.N.; Kaiser, L.; Polosukhin, I. Attention Is All You Need. In 31st Conference on Neural Information Processing Systems, 2017, arXiv:1706.03762v7. https://doi.org/10.48550/ARXIV .1706.03762. Disclaimer/Publ...

    2017 doi arXiv

Pith tools

Reviewed August 9, 2026 · model on record in the stance chip above.