Pith. sign in

REVIEW 4 major objections 7 minor 31 references

Multimodal Conditional MeshGAN for Personalized Aneurysm Growth Prediction

T0 review · 4 major / 7 minor · reviewed 2026-08-05 · deepseek-v4-flash

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

Pith's one-line read MCMeshGAN claims that a mesh-to-mesh GAN conditioned on age, sex, and target time interval predicts future thoracic aortic aneurysm geometry more accurately than five graph and point-cloud baselines.

desk verdict A genuinely new application and dataset; the headline numbers all run through an unvalidated non-rigid registration step, so the clinical claims need a registration sanity check. read the letter →

arxiv 2508.19862 v1 pith:6XB67S5A submitted 2025-08-27 cs.CV cs.LG

Long Chen , Ashiv Patel , Mengyun Qiao , Mohammad Yousuf Salmasi , Salah A. Hammouche , Vasilis Stavrinides , Jasleen Nagi , Soodeh Kalaie
show 3 more authors
Xiao Yun Xu Wenjia Bai Declan P. O'Regan
This is my paper
classification cs.CVcs.LG
keywords aneurysmgrowthpredictionmesh-to-meshtranslationconditionalGANgraphconvolutionalnetworkKNNconvolutionthoracicaorticlongitudinalCTpersonalizedmedicine
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 a GAN operating directly on 3D surface meshes can predict how a thoracic aortic aneurysm will deform over months to years. The proposed MCMeshGAN combines a local K-nearest-neighbour convolution branch with a global graph-convolution branch, and conditions the generated mesh on patient age, sex, and the requested time interval. The authors built a longitudinal dataset of 590 CT-derived meshes from 208 patients and report that MCMeshGAN beats five adapted baselines on per-vertex error, Chamfer distance, Hausdorff distance, and the clinically used maximum-inscribed-sphere diameter. If correct, the approach would let a clinician start from one baseline scan and generate an estimated future aneurysm surface at any chosen time interval, supporting personalized surveillance scheduling.

What carries the argument

The mechanism is a three-branch generator: a condition branch that encodes age, sex, and time interval into 100-dimensional ordinal binary vectors; a local KCN branch that finds K nearest neighbours in a CNN feature space and convolves over each vertex and its neighbours; and a global GCN branch that applies normalized graph-Laplacian message passing over four layers. A fusion module combines the features and a discriminator supplies adversarial loss. The KCN branch is the load-bearing novelty: it compensates for GCN over-smoothing by explicitly modelling fine local variation, which is where aneurysm diameter changes are expressed.

What would settle it

Run the trained model on an independent longitudinal thoracic aortic aneurysm dataset and compare predicted meshes against ground truth using a correspondence-free metric such as surface Dice or clinically reported MIS diameter; if the reported accuracy drops to baseline levels, the claimed gains are likely artifacts of the registration. Alternatively, validate the non-rigid registration by checking whether corresponding vertices track identifiable anatomical landmarks.

Watch

Extended reading notes

Core claim

The central claim is that mesh-to-mesh translation with multimodal conditioning works for aneurysm growth: given a baseline thoracic aortic aneurysm mesh, patient age and sex, and a target time interval (positive for future, negative for past), MCMeshGAN generates the patient's aneurysm mesh at that time. On the test set, it reports a mean absolute error of 1.285 ± 0.993 mm, Chamfer distance of 1.831 ± 1.251 mm, Hausdorff distance of 9.815 ± 5.131 mm, and MIS diameter error of 2.887 mm, all lower than the five baselines, including when the local branch is added to them. The paper interprets this as evidence that the global graph branch preserves overall anatomy while the KNN-based local bran

Load-bearing premise

The central premise is that the non-rigid surface registration used in preprocessing establishes a true one-to-one correspondence between vertices across scans; the paper does not validate that correspondence, so if registration misaligns vertices, the reported per-vertex errors reflect registration artifacts rather than anatomical growth.

Editorial extensions

If this is right

  • If the reported accuracy holds, a single baseline CT mesh plus age, sex, and a requested interval yields a predicted 3D aneurysm surface, enabling individualized surveillance intervals instead of uniform annual imaging.
  • The time-interval conditioning supports both prospective prediction (positive intervals) and retrospective reconstruction (negative intervals), so the same model can be trained and evaluated on all longitudinal pairs in the dataset.
  • The ablation results indicate that the KCN branch improves every graph-convolution baseline, suggesting the local-detail module is a reusable component for mesh generation beyond aneurysm prediction.
  • The lowest MIS diameter error (2.887 mm) is the result most tied to clinical decisions, since maximum diameter thresholds guide surgical referral.

Reading between the lines

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

  • The paper only evaluates on its own private dataset, so its own stated limitation is that generalization to other scanners, protocols, and populations is unestablished; an external dataset test is the natural next check.
  • All geometric metrics are computed on corresponding vertices, so a registration-free metric (e.g., surface Dice or clinically reported diameters) would test whether the gains reflect anatomy rather than the alignment procedure.
  • The 100-dimensional ordinal time encoding implies a bounded time horizon and discrete interval granularity; a continuous time embedding might extend the model to arbitrary follow-up lengths.
  • No comparison is made to inter-observer or scan-rescan variability of MIS diameter; whether 2.887 mm error is clinically acceptable depends on how it compares to measurement noise in routine practice.
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

4 major / 7 minor

Summary. The paper introduces MCMeshGAN, a multimodal conditional mesh-to-mesh GAN for predicting thoracic aortic aneurysm growth from 3D surface meshes and clinical metadata (age, sex, time interval). The method combines a global GCN branch and a local KNN-based convolutional branch, conditioned on ordinal binary encodings of patient attributes. The authors also present TAAMesh, a private longitudinal dataset of 590 CT-derived meshes from 208 patients. On a held-out test set, MCMeshGAN reports the lowest MAE, CD, HD, and MIS diameter error among five adapted baselines, and ablations study the contributions of backbone, conditions, and losses. Source code is publicly released.

Significance. If the results hold, this is a useful step toward 3D mesh-to-mesh generative modeling of aneurysm progression, an area that is underexplored compared with 2D image or feature-based approaches. The public release of code and the introduction of a longitudinal TAA mesh dataset are concrete contributions. The dual-branch architecture that balances global structure and local detail is reasonable and well motivated. However, the strength of the empirical claims is currently conditional: the evaluation depends on an unvalidated non-rigid registration, one key baseline result is missing, and no statistical significance testing is provided. With these issues addressed, the paper would be a solid contribution to the medical image computing community.

major comments (4)
  1. [Sec. 4.1 and Sec. 5.2.3] The entire evaluation pipeline relies on IRTK non-rigid surface registration to establish one-to-one vertex correspondence across patients and time points. All reported metrics (MAE, CD, HD, MIS error) and the L1 loss are computed on registered meshes. The manuscript provides no validation of this registration: no landmark or centerline error, no comparison of MIS diameter or anatomy before vs. after non-rigid registration, and no residual analysis. If the registration locally warps the aneurysm sac, the reported accuracy numbers, as well as the argument that L1 loss outperforms CD loss because of one-to-one correspondence (Sec. 5.2.3), could reflect registration artifacts rather than biological growth. The authors should add a registration-fidelity validation and/or a sensitivity analysis showing that the main conclusions are robust to the registration choice.
  2. [Table 1, Sec. 5.1] In Table 1, the row 'GCNConv [8] GCN+Condition+KCN' is blank (all entries are '-'), while the text states that 'the KCN branch significantly enhances all GCN frameworks, particularly GCNConv and GCN2Conv' and that 'GCNConv obtains the best improvements from KCN.' Since GCNConv is the GCN backbone used in MCMeshGAN, this result is essential support for the central design choice. Please report the result, or explicitly explain why it is missing (e.g., training failure) and adjust the claim accordingly.
  3. [Sec. 5.1, Table 1] The central claim that MCMeshGAN 'consistently outperforms' baselines is based on point estimates without statistical inference. The reported standard deviations are large relative to the differences (e.g., MAE 1.285±0.993 vs. 1.605±3.516 for GCN2Conv+KCN; CD 1.831±1.251 vs. 2.228±2.457), so the superiority is not established. No paired tests, confidence intervals, or effect sizes are provided. MIS error is reported only as a mean, without standard deviation or CI. The authors should perform per-patient paired statistical tests (e.g., Wilcoxon signed-rank) or bootstrap CIs for all metrics, including MIS error.
  4. [Sec. 4.1 and Conclusion] The evaluation is conducted on a single private dataset with no external validation. The abstract and conclusion refer to 'clinically deployable prediction' and 'generalization,' but the limitations section correctly notes that the model was only tested on this private dataset. Given the single-center data and moderate sample size, the clinical and generalization claims are overstated. At a minimum, the wording should be tempered; ideally, an additional validation experiment (nested cross-validation, leave-one-center-out, or evaluation on a public AAA dataset) would strengthen the support for generalization.
minor comments (7)
  1. [Sec. 3.2, Fig. 2] The ordinal binary encoding of the time interval is not fully specified: it is unclear how negative intervals beyond -50 or positive intervals beyond +50 are handled, and what happens exactly at the boundary index 50. Please clarify the encoding with a precise formula.
  2. [Table 1] The table header appears twice (the row 'Methods' is repeated). Please merge into a single header and use clear grouping for the '+Condition' and '+Condition+KCN' blocks.
  3. [Sec. 3.3, Eq. (4)] The subscripts 'FN ei' and 'FCenter' are unclear; the notation for the neighbor and center convolution functions should be defined consistently.
  4. [Sec. 5.1] Typo: 'essenstial' should be 'essential'.
  5. [Sec. 5.2.3] The text references 'Section IV-A' but should refer to Section 4.1.
  6. [Sec. 5.2.1] The claim that age has a greater impact than sex because of its larger dynamic range is speculative. Consider reporting feature importance or performing an interaction analysis to support this conclusion.
  7. [Sec. 4.2] The KNN neighborhood size K is set to 8 without sensitivity analysis. A small grid search over K would improve confidence in the architecture's robustness.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the central claim rests on held-out patient-level test performance, not on fitted values or self-citation.

full rationale

MCMeshGAN is an empirical mesh-to-mesh prediction model; its abstract and Section 5.1 claims are benchmarked on a patient-level held-out split of the TAAMesh dataset (Section 4.1, 'split into training, validation, and test sets at patient level, using a 7:1:2 ratio'). The target meshes and MIS diameters are ground-truth follow-up measurements, not quantities reconstructed from the model's fitted parameters. The condition branch (Section 3.2) takes age, sex, and target time interval as fixed ordinal-encoded inputs; it is not fit to the target mesh. The L1, CD, and adversarial losses (Section 5.2.3) are training objectives, and the reported MAE/CD/HD/MIS errors are evaluated on independent test patients, so there is no fitted-input-called-prediction reduction. The only self-citation is SpiralNet++ [18] in Related Works (background only), and it is not load-bearing. The paper's own limitations in Section 6—'Our model was only tested on our private dataset' and 'has not yet been tested in real clinical workflows'—are honest scope restrictions, not circular reasoning. The unvalidated non-rigid registration in Section 4.1 is a real correctness/generalization risk (all reported vertex metrics depend on one-to-one correspondence), but it is an empirical validity concern, not a circular derivation: the baseline methods are evaluated under the same registration, and no equation or fitted parameter reduces the test metric to the training input by construction. Accordingly no circular step is identified.

Assumptions & free parameters 3 free parameters · 4 assumptions · 0 invented entities

The central result is an empirical architecture trained on a private dataset. The hand-set hyperparameters (K, GCN depth, encoding length) are design choices, and the main domain assumptions concern segmentation and registration accuracy. No new physical entities are introduced.

free parameters (3)
  • KNN neighbor count K = 8
    K is set to 8 empirically (Section 4.2); the local KCN branch and its performance depend on this neighborhood size.
  • Number of stacked GCN blocks = 4
    Four graph convolutional blocks were chosen by the authors; depth affects the balance between global context and over-smoothing.
  • Time interval encoding length = 100 with split at index 50
    The ordinal binary vector length of 100 and the index-50 split point are design choices introduced for this dataset, not derived from data or clinical theory.
assumptions (4)
  • domain assumption Non-rigid surface registration establishes one-to-one vertex correspondence across longitudinal scans and across patients.
    Section 4.1 uses IRTK non-rigid registration to warp meshes for vertex correspondence; L1 loss and per-vertex MAE depend on this correspondence.
  • domain assumption TotalSegmentator plus manual correction yields pixel-level accurate aortic segmentation masks.
    Section 4.1: segmentation errors propagate to meshes and diameter estimates, and manual corrections are not quantified or validated.
  • ad hoc to paper The ordinal binary encoding of age, sex, and time interval is an adequate representation for conditioning.
    Section 3.2 and Figure 2: vector length 100 and split index 50 are arbitrary design choices with no evidence that they optimally encode temporal or clinical variation.
  • standard math Standard GCN message passing is valid for mesh data.
    Section 3.4 uses the normalized adjacency graph convolution from Kipf and Welling (Eq. 5).

how reviews work

0 comments
Cite this review

Pith. "Pith review of Multimodal Conditional MeshGAN for Personalized Aneurysm Growth Prediction." pith.science (2026). https://pith.science/paper/6XB67S5A

@misc{pith2026250819862,
  author       = {Pith},
  title        = {Pith review of: Multimodal Conditional MeshGAN for Personalized Aneurysm Growth Prediction},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/6XB67S5A}},
  note         = {Machine review of arXiv:2508.19862}
}
read the original abstract

Personalized, accurate prediction of aortic aneurysm progression is essential for timely intervention but remains challenging due to the need to model both subtle local deformations and global anatomical changes within complex 3D geometries. We propose MCMeshGAN, the first multimodal conditional mesh-to-mesh generative adversarial network for 3D aneurysm growth prediction. MCMeshGAN introduces a dual-branch architecture combining a novel local KNN-based convolutional network (KCN) to preserve fine-grained geometric details and a global graph convolutional network (GCN) to capture long-range structural context, overcoming the over-smoothing limitations of deep GCNs. A dedicated condition branch encodes clinical attributes (age, sex) and the target time interval to generate anatomically plausible, temporally controlled predictions, enabling retrospective and prospective modeling. We curated TAAMesh, a new longitudinal thoracic aortic aneurysm mesh dataset consisting of 590 multimodal records (CT scans, 3D meshes, and clinical data) from 208 patients. Extensive experiments demonstrate that MCMeshGAN consistently outperforms state-of-the-art baselines in both geometric accuracy and clinically important diameter estimation. This framework offers a robust step toward clinically deployable, personalized 3D disease trajectory modeling. The source code for MCMeshGAN and the baseline methods is publicly available at https://github.com/ImperialCollegeLondon/MCMeshGAN.

Figures

Figures reproduced from arXiv: 2508.19862 by the authors.

Figure 1
Figure 1. Overview of the MCMeshGAN architecture: (a) The condition branch receives clinically relevant information, [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 2
Figure 2. Text2Vector transforms age, sex, and target interval into vectors using ordinal binary encoding. The [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 2
Figure 2. The time interval represents the temporal difference between the baseline and target time points (e.g., a time [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figures from the paper (9 more)
Figure 3
Figure 3. Figure 3: Overview of the KCN branch. The vertex coordinates are represented as a 2D data array and a CNN feature [PITH_FULL_IMAGE:figures/full_fig_p005_3.png]
Figure 4
Figure 4. Figure 4: The adjacency matrix A, the self-looped adjacency matrix A¯, and the degree matrix D¯ in the graph branch are derived from the edges connecting vertices in the mesh. The local KCN branch explicitly gathers information on how vertices relate to their k nearest neighbors…
Figure 5
Figure 5. Figure 5: Summary statistics for patient age and maximum inscribed sphere (MIS) diameter in the TAAMesh dataset. [PITH_FULL_IMAGE:figures/full_fig_p007_5.png]
Figure 6
Figure 6. Figure 6: Overview of the CT-to-mesh preprocessing pipeline: (a) CT segmentation. The pre-trained TotalSegmentator [PITH_FULL_IMAGE:figures/full_fig_p007_6.png]
Figure 7
Figure 7. Figure 7: Given a ground truth (GT) baseline mesh (dated 2014/03), MCMeshGAN predicts follow-up mesh sequences [PITH_FULL_IMAGE:figures/full_fig_p008_7.png]
Figure 8
Figure 8. Figure 8: Performance comparisons of the evaluated methods. Box plots summarize three error metrics: mean average [PITH_FULL_IMAGE:figures/full_fig_p009_8.png]
Figure 9
Figure 9. Figure 9: Qualitative comparison of MCMeshGAN with baseline frameworks. The color maps illustrate the Chamfer [PITH_FULL_IMAGE:figures/full_fig_p009_9.png]
Figure 10
Figure 10. Figure 10: Performance comparisons (a) and losses (b, c, d) for MCMeshGANs with various backbones and losses. [PITH_FULL_IMAGE:figures/full_fig_p011_10.png]
Figure 11
Figure 11. Figure 11: Qualitative results of MCMeshGAN with different backbones. The MIS diameter and center-line shown [PITH_FULL_IMAGE:figures/full_fig_p011_11.png]

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

31 extracted references · 29 canonical work pages

  1. [8]

    Semi-supervised classification with graph convolutional networks,

    T. N. Kipf and M. Welling, “Semi-supervised classification with graph convolutional networks,” inInternational Conference on Learning Representations (ICLR), 2017

    work page 2017

  2. [1]

    An artificial intelligence based abdominal aortic aneurysm prognosis classifier to predict patient outcomes,

    T. K. Chung, P. H. Gueldner, O. U. Aloziem, N. L. Liang, and D. A. V orp, “An artificial intelligence based abdominal aortic aneurysm prognosis classifier to predict patient outcomes,” Scientific Reports, vol. 14, no. 1, p. 3390, 2024

    work page 2024

  3. [2]

    Existing and emerging approaches to risk assessment in patients with ascending thoracic aortic dilatation,

    N. D. Anfinogenova et al., “Existing and emerging approaches to risk assessment in patients with ascending thoracic aortic dilatation,” Journal of Imaging, vol. 8, no. 10, p. 280, 2022

    work page 2022

  4. [3]

    Ascending aneurysmectomy: Should we shift to the left?

    J. A. Elefteriades, J. A. Rizzo, M. A. Zafar, and B. A. Ziganshin, “Ascending aneurysmectomy: Should we shift to the left?” The Journal of Thoracic and Cardiovascular Surgery, vol. 166, no. 2, pp. 435–441, 2023

    work page 2023

  5. [4]

    Patient-specific prediction of abdominal aortic aneurysm expansion using bayesian calibration,

    L. Zhang, Z. Jiang, J. Choi, C. Y . Lim, T. Maiti, and S. Baek, “Patient-specific prediction of abdominal aortic aneurysm expansion using bayesian calibration,” IEEE Journal of Biomedical and Health Informatics, vol. 23, no. 6, pp. 2537–2550, 2019

    work page 2019

  6. [5]

    Prediction of abdominal aortic aneurysm growth using dynamical gaussian process implicit surface,

    H. N. Do et al., “Prediction of abdominal aortic aneurysm growth using dynamical gaussian process implicit surface,” IEEE Transactions on Biomedical Engineering, vol. 66, no. 3, pp. 609–622, 2018

    work page 2018

  7. [6]

    Computer-aided shape features extraction and regression models for predicting the ascending aortic aneurysm growth rate,

    L. Geronzi et al., “Computer-aided shape features extraction and regression models for predicting the ascending aortic aneurysm growth rate,” Computers in Biology and Medicine, vol. 162, p. 107052, 2023

    work page 2023

  8. [7]

    A deep learning approach to predict abdominal aortic aneurysm expansion using longitudinal data,

    Z. Jiang, H. N. Do, J. Choi, W. Lee, and S. Baek, “A deep learning approach to predict abdominal aortic aneurysm expansion using longitudinal data,” Frontiers in Physics, vol. 7, p. 235, 2020

    work page 2020

Show all 31 references
  1. [9]

    Generating 3D faces using convolutional mesh autoencoders,

    A. Ranjan, T. Bolkart, S. Sanyal, and M. J. Black, “Generating 3D faces using convolutional mesh autoencoders,” in Proceedings of the European conference on computer vision (ECCV), 2018, pp. 704–720

    2018

  2. [10]

    MESHWGAN: Mesh-to-mesh Wasserstein GAN with multi-task gradient penalty for 3D facial geometric age transformation,

    J. Zhang, K. Zhou, Y . Luximon, T.-Y . Lee, and P. Li, “MESHWGAN: Mesh-to-mesh Wasserstein GAN with multi-task gradient penalty for 3D facial geometric age transformation,” IEEE Transactions on Visualization and Computer Graphics, 2023

    2023

  3. [11]

    Scattering GCN: Overcoming oversmoothness in graph convolutional networks,

    Y . Min, F. Wenkel, and G. Wolf, “Scattering GCN: Overcoming oversmoothness in graph convolutional networks,” Advances in Neural Information Processing Systems, vol. 33, pp. 14 498–14 508, 2020

    2020

  4. [12]

    A survey on oversmoothing in graph neural networks,

    T. K. Rusch, M. M. Bronstein, and S. Mishra, “A survey on oversmoothing in graph neural networks,” arXiv preprint arXiv:2303.10993, 2023. 12

    2023 arXiv

  5. [13]

    A multi-stream network for mesh denoising via graph neural networks with Gaussian curvature,

    Z. Zhao, W. Wu, H. Liu, and Y . Gong, “A multi-stream network for mesh denoising via graph neural networks with Gaussian curvature,” in 2023 IEEE International Conference on Image Processing (ICIP). IEEE, 2023, pp. 1355–1359

    2023

  6. [14]

    PU-MFA: Point cloud up-sampling via multi-scale features attention,

    H. Lee and S. Lim, “PU-MFA: Point cloud up-sampling via multi-scale features attention,” Sensors, vol. 22, no. 23, p. 9308, 2022

    2022

  7. [15]

    Diffusion probabilistic models for 3D point cloud generation,

    S. Luo and W. Hu, “Diffusion probabilistic models for 3D point cloud generation,” inProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2021, pp. 2837–2845

    2021

  8. [16]

    Hyper-MD: Mesh denoising with customized parameters aware of noise intensity and geometric characteristics,

    X. Wang, H. Wei, X. Fan, and D. Zhao, “Hyper-MD: Mesh denoising with customized parameters aware of noise intensity and geometric characteristics,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2024, pp. 4651–4660

    2024

  9. [17]

    FaceTuneGAN: Face autoencoder for convolutional expression transfer using neural generative adversarial networks,

    N. Olivier, K. Baert, F. Danieau, F. Multon, and Q. Avril, “FaceTuneGAN: Face autoencoder for convolutional expression transfer using neural generative adversarial networks,”Computers & Graphics, vol. 110, pp. 69–85, 2023

    2023

  10. [18]

    Spiralnet++: A fast and highly efficient mesh convolution operator,

    S. Gong, L. Chen, M. Bronstein, and S. Zafeiriou, “Spiralnet++: A fast and highly efficient mesh convolution operator,” in Proceedings of the IEEE/CVF International Conference on Computer Vision Workshops, 2019, pp. 0–0

    2019

  11. [19]

    Geometric deep learning: Going beyond Euclidean data,

    M. M. Bronstein, J. Bruna, Y . LeCun, A. Szlam, and P. Vandergheynst, “Geometric deep learning: Going beyond Euclidean data,” IEEE Signal Processing Magazine, vol. 34, no. 4, pp. 18–42, 2017

    2017

  12. [20]

    Data-driven geometry-recovering mesh denoising,

    J. Wang, J. Huang, F. L. Wang, M. Wei, H. Xie, and J. Qin, “Data-driven geometry-recovering mesh denoising,” Computer-Aided Design, vol. 114, pp. 133–142, 2019

    2019

  13. [21]

    Mesh defiltering via cascaded geometry recovery,

    M. Wei et al., “Mesh defiltering via cascaded geometry recovery,” inComputer Graphics Forum, vol. 38, no. 7. Wiley Online Library, 2019, pp. 591–605

    2019

  14. [22]

    Simple and deep graph convolutional networks,

    M. Chen, Z. Wei, Z. Huang, B. Ding, and Y . Li, “Simple and deep graph convolutional networks,” inInternational Conference on Machine Learning. PMLR, 2020, pp. 1725–1735

    2020

  15. [23]

    Simple spectral graph convolution,

    H. Zhu and P. Koniusz, “Simple spectral graph convolution,” inInternational Conference on Learning Representa- tions, 2021

    2021

  16. [24]

    Graph neural networks are inherently good generalizers: Insights by bridging GNNs and MLPs,

    C. Yang, Q. Wu, J. Wang, and J. Yan, “Graph neural networks are inherently good generalizers: Insights by bridging GNNs and MLPs,” 2023

    2023

  17. [25]

    On growth measurements of abdominal aortic aneurysms using maximally inscribed spheres,

    H. Gharahi, B. Zambrano, C. Lim, J. Choi, W. Lee, and S. Baek, “On growth measurements of abdominal aortic aneurysms using maximally inscribed spheres,” Medical Engineering & Physics, vol. 37, no. 7, pp. 683–691, 2015

    2015

  18. [26]

    Deep learning on multiphysical features and hemodynamic modeling for abdominal aortic aneurysm growth prediction,

    S. Kim et al., “Deep learning on multiphysical features and hemodynamic modeling for abdominal aortic aneurysm growth prediction,” IEEE Transactions on Medical Imaging, vol. 42, no. 1, pp. 196–208, 2022

    2022

  19. [27]

    Lifetime risk and risk factors for abdominal aortic aneurysm in a 24-year prospective study: the aric study (atherosclerosis risk in communities),

    W. Tanget al., “Lifetime risk and risk factors for abdominal aortic aneurysm in a 24-year prospective study: the aric study (atherosclerosis risk in communities),” Arteriosclerosis, Thrombosis, and Vascular Biology, vol. 36, no. 12, pp. 2468–2477, 2016

    2016

  20. [28]

    TotalSegmentator: robust segmentation of 104 anatomic structures in CT images,

    J. Wasserthal et al., “TotalSegmentator: robust segmentation of 104 anatomic structures in CT images,”Radiology: Artificial Intelligence, vol. 5, no. 5, 2023

    2023

  21. [29]

    Pointnet++: Deep hierarchical feature learning on point sets in a metric space,

    C. R. Qi, L. Yi, H. Su, and L. J. Guibas, “Pointnet++: Deep hierarchical feature learning on point sets in a metric space,” Advances in Neural Information Processing Systems, vol. 30, 2017

    2017

  22. [30]

    Fast graph representation learning with PyTorch Geometric,

    M. Fey and J. E. Lenssen, “Fast graph representation learning with PyTorch Geometric,” inICLR Workshop on Representation Learning on Graphs and Manifolds, 2019

    2019

  23. [31]

    Association of Mortality and Acute Aortic Events With Ascending Aortic Aneurysm: A Systematic Review and Meta-analysis

    M.H. Guo et al., “Association of Mortality and Acute Aortic Events With Ascending Aortic Aneurysm: A Systematic Review and Meta-analysis.” JAMA Netw Open, vol. 1, pp. e181282, 2018. 13

    2018

Pith tools

Reviewed August 5, 2026 · model on record in the stance chip above.