Pith. sign in

REVIEW 3 major objections 5 minor 77 references

Wall Shear Stress Estimation in Abdominal Aortic Aneurysms: Towards Generalisable Neural Surrogate Models

T0 review · 3 major / 5 minor · reviewed 2026-08-06 · deepseek-v4-flash

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

Pith's one-line read A neural surrogate estimates wall shear stress in abdominal aortic aneurysms with 0.296 Pa mean absolute error against CFD on an external test set, and holds that accuracy across remodelling, new branches, and mesh resolution.

desk verdict Solid empirical study of an E(3)-equivariant WSS surrogate on real AAA data, but the 'boundary conditions' generalization claim rests on a single scalar inflow amplitude and should be toned down. read the letter →

arxiv 2507.22817 v1 pith:DNFKQG7S submitted 2025-07-30 cs.CV

Patryk Rygiel , Julian Suk , Christoph Brune , Kak Khee Yeung , Jelmer M. Wolterink This is my paper · ORCID
classification cs.CV
keywords neuralsurrogatesgeometricdeeplearninggeneralisationabdominalaorticaneurysmshemodynamicswallshearstresscomputationalfluiddynamicsalgebra
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 sets out to show that a geometric deep-learning model can replace hours-long computational fluid dynamics (CFD) simulations for estimating wall shear stress (WSS) in abdominal aortic aneurysms, and that it remains accurate under the variations that real clinical data contain. The authors train an $E(3)$-equivariant transformer, LaB-GATr, on CT-derived lumen geometries of 100 patients with 1,090 reference CFD simulations, and test it on 118 scans from 29 further patients. Their central result is a set of generalisation behaviours: accuracy is maintained when the inflow boundary condition changes, when an aneurysm remodels over years of follow-up, when entirely new branches are added to the vessel tree at inference time, and when the surface mesh resolution changes. If these results hold, hemodynamic markers such as time-averaged WSS and oscillatory shear index could be obtained in seconds rather than hours, making them practical for patient monitoring and large-scale studies.

What carries the argument

The load-bearing object is LaB-GATr, a transformer that operates on point clouds inside the projective geometric algebra $\mathbf{G}(3,0,1)$: every surface point carries a 16-component multivector embedding, and since all linear maps, attention, and nonlinearities are performed in the algebra, the whole model is equivariant to translations, rotations, and reflections of the geometry in 3D space. To handle meshes of one to five million elements, the architecture tokenises the point cloud by farthest-point sampling to a 10% coarse subset, applies cross-attention between coarse and full clouds, and interpolates predictions back to full resolution with inverse-distance weighting. The argument is carried as much by the input descriptors as by the architecture: normal vectors, geodesic maps to the inlet and the nearest outlet computed with the heat method, principal curvatures derived from Gaussian and mean curvature, and a flow-prior feature built by parallel-transporting the inlet velocity direction over the surface with the vector heat method. Peak inflow is encoded as a single scalar, the maximum inlet velocity $V_{\max}$ derived from $Q_{\max}$ and inlet area, and the training loss separates WSS magnitude (an L1 term) from WSS direction (a cosine-similarity term), which is what allows the authors to show that equivariance helps the direction-sensitive oscillatory shear index.

What would settle it

Run the trained model on the same geometry twice, at the same peak inflow $Q_{\max}$, but with two temporally different inflow waveform shapes that were never seen in training (for instance a biphasic or exercise waveform). The model, being conditioned only on $Q_{\max}$, will predict identical WSS fields for both; if the reference CFD shows the two waveforms produce materially different wall shear stress fields on that geometry, then the claimed boundary-condition generalisability holds only for scaled copies of the single template waveform, and the surrogate cannot be trusted for genuinely patient-specific flow conditions.

Watch

Extended reading notes

Core claim

The central claim is that the transient wall shear stress vector field on an aneurysm's lumen surface, a value at every surface point across 21 timepoints of the cardiac cycle, can be regressed directly from the geometry and the peak inflow rate, without solving the Navier-Stokes equations. On the external AAA-L test set, LaB-GATr reaches a mean absolute error of $0.296$ Pa for transient WSS, an approximation disparity of $0.333$ for TAWSS and $0.298$ for OSI, a cosine similarity of $0.684$ on vector direction, and a 1500-fold speedup over the CFD reference. The paper attributes the margin over a non-equivariant counterpart, LaB-VaTr, to the built-in Euclidean symmetry and to the geometric descriptors fed into the network: surface normals, geodesic distances to inlet and outlets, principal curvatures, and a laminar-flow prior transported across the surface. It further claims that the model has internalised the geometry-to-hemodynamics relationship well enough to adapt zero-shot to vessel trees with previously unseen branches and to be effectively indifferent to mesh discretisation, which the authors read as evidence that the model learns the underlying continuous vessel-wall manifold rather than template-specific details.

Load-bearing premise

The reference CFD solutions, computed with a single template inflow waveform, rigid vessel walls, a parabolic inlet velocity profile, and generic Windkessel outlet conditions, with no patient-specific flow measurements available, are treated as ground truth for real aneurysm hemodynamics, so if those modelling choices misrepresent an individual patient's actual blood flow, matching the CFD reference to within 0.296 Pa would not guarantee clinical accuracy.

Editorial extensions

If this is right

  • Because the model is conditioned on the peak inflow rate, the same trained network can produce WSS at any flow condition in seconds, making per-patient hemodynamic screening feasible without rerunning CFD.
  • Longitudinal tracking of low-TAWSS and high-OSI regions follows the reference CFD within 0.021 Pa and 0.020 across up to seven scans per patient, so the surrogate could monitor aneurysm progression over time.
  • Zero-shot extension to vessel trees with added branches degrades accuracy in the originally trained region by only 0.015 Pa in MAE, implying that one model can serve anatomies with different branch configurations.
  • Error distributions are statistically indistinguishable across 1.0, 1.25, and 1.5 mm meshes, so the model can consume segmentations from different scanners or preprocessing pipelines without retraining.
  • Because the model outputs the full transient WSS vector field, other clinically used markers such as relative residence time and endothelial cell activation potential can be derived from the same prediction.

Reading between the lines

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

  • The model's only inflow information is the peak value $Q_{\max}$, yet the CFD references vary only the magnitude of one template waveform; whether accuracy survives a change in waveform shape at equal $Q_{\max}$ is a test the paper does not run, and it is the natural next experiment for claiming patient-specific boundary-condition generalisation.
  • The observed over-smoothing of turbulence-driven, high-frequency WSS patterns suggests a ceiling on surface-only supervision; extrapolating the paper's own reasoning, the direction-sensitive OSI marker would be the first to degrade on more tortuous or larger geometries, so a volumetric velocity-and-pressure extension is the plausible path to lifting that ceiling.
  • The zero-shot topology result implies the learned mapping behaves more like a physics prior than a shape-memorising regressor; a testable extension is whether the same architecture transfers to other vessel beds, such as carotid or coronary arteries, or to geometries containing intraluminal thrombus, which alters the lumen surface and is excluded here.
  • The combination of 1500-fold speedup, mesh agnosticism, and external-test accuracy positions the surrogate for opportunistic screening on CTAs acquired for other indications, but that use inherits the paper's rigid-wall template-inflow CFD ground truth and would need validation against in-vivo measurements such as 4D-flow MRI.
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

3 major / 5 minor

Summary. The paper proposes an E(3)-equivariant geometric deep learning surrogate, LaB-GATr, for transient wall shear stress (WSS) estimation on abdominal aortic aneurysm (AAA) lumen surfaces. The model takes a point cloud of the vessel wall, geometric descriptors (coordinates, normals, geodesic distances, principal curvatures, and a flow-prior vector field), and a scalar inflow-conditioning feature, and predicts the transient WSS vector field. Training uses 100 CT-derived geometries with four CFD simulations each under a template inflow waveform scaled to different peak inflows. The model is evaluated with 10-fold cross-validation on AAA-100 and on an external longitudinal dataset of 29 patients (118 scans). The authors report MAE, NMAE, approximation disparity, and cosine similarity for transient WSS and derived TAWSS and OSI, and they compare against a linear-algebra variant and a 2D CNN. They then study four generalisation settings: varying boundary conditions, longitudinal geometry remodelling, zero-shot changes in vascular topology, and changes in mesh resolution. The reported results indicate the geometric-algebra model generally outperforms the baselines, and the paper claims accurate estimation and generalisation across these settings.

Significance. The study is a timely and useful addition to the literature on neural surrogates for hemodynamic simulations. Its strengths include the use of a genuinely external test set, public code release, a clean ablation of geometric descriptors, and a systematic organisation of generalisation challenges. If the central claims hold, the E(3)-equivariant architecture with the proposed descriptors offers a fast surrogate with promising cross-domain transfer, which is relevant for clinical and large-scale hemodynamic studies. The main weaknesses are that the 'boundary condition' generalisation is limited to peak-flow amplitude, the topology experiment is small and its reported effect is larger than the text suggests, and the accuracy is not calibrated against clinical tolerance. These issues do not invalidate the core method but they require a substantial revision of the claims.

major comments (3)
  1. [Abstract; Section 3.5.1; Section 2.2.3] The boundary-condition generalisation experiment (Section 3.5.1) varies only the peak inflow amplitude Qmax; the template inflow waveform [60], the parabolic spatial profile, and all Windkessel outlet parameters (P_mean, Q, C, P_d, and resistance split 0.91/0.09) are fixed as described in Section 2.1.2. The model's inflow conditioning (Section 2.2.3) uses only Vmax, a scalar derived from Qmax and inlet area. Therefore the experiment supports interpolation over the amplitude of a single waveform, not the abstract's claim of generalisation to 'changes in boundary conditions'. A model with only Vmax as input cannot, in principle, respond to a changed waveform shape or a different outlet resistance split at the same Qmax. Please either rephrase the claims to be specific about amplitude-only variation or add experiments that vary waveform shape and outlet parameters.
  2. [Section 3.5.3, Table 4] The zero-shot topology experiment uses only 10 geometries, all at a single peak inflow Qmax = 80 ml/s. The text says performance on the original region 'deteriorates significantly' yet 'differences are not large', but Table 4 shows MAE increasing from 0.179 to 0.194 Pa and Approx. disp. increasing from 0.498 to 0.660, a 33% relative increase that the reported Wilcoxon test finds significant (p<0.05). The sentence 'MAE dropping only by 0.015 [PA]' misstates the direction of change. The evidence for zero-shot extrapolation to unseen topologies is therefore weaker than the conclusion suggests; the experiment should either be expanded to a larger sample with varied inflow rates or the conclusion should be tempered.
  3. [Abstract; Section 3.2; Table 1] The paper describes the model as 'accurately estimating hemodynamics', but the external-test approximation disparity is 0.513 for transient WSS, 0.333 for TAWSS, and 0.298 for OSI, with a TAWSS MAE of 0.148 Pa (Table 1). Given that TAWSS values in AAA are commonly in the range of 0.5-2 Pa, these relative errors are not negligible for clinical decision-making. The CFD reference itself uses fixed, non-personalised boundary conditions, as acknowledged in the Discussion, so the reported accuracy is relative to that specific reference. Please provide a benchmark against a clinically acceptable tolerance (e.g., inter-observer variability in CFD-derived WSS or thresholds used in growth/rupture studies) or explicitly frame the accuracy as relative to the specific CFD reference used.
minor comments (5)
  1. [Abstract] The abstract describes the external test set as '118 AAA patients', but Section 2.1 states that AAA-L contains 29 patients with 118 unique scans; please correct this factual description.
  2. [Table 4] The title 'Qualitative comparison of zero-shot performance' is inaccurate because the table reports numerical metrics; please rename it to 'Quantitative comparison'.
  3. [Section 3.5.4] The mesh-sensitivity conclusion is based on a Wilcoxon test with p>0.05, but the report does not give the sample size or effect sizes; with 118 geometries the test may be underpowered, so please add confidence intervals or effect sizes for the differences.
  4. [Equation (3)] The angle loss is undefined for predicted vectors of zero magnitude; please specify how zero-magnitude predictions are handled.
  5. [Section 2.2.1] The novelty relative to the existing LaB-GATr architecture [55] should be clarified; as written, the architecture appears to be taken from prior work and the contribution lies in the geometric descriptors and the generalisation study rather than in the architecture itself.

Circularity Check

0 steps flagged · score 2.0 of 10

No significant circularity: WSS predictions are scored against independently generated SimVascular CFD and an external multi-institution test set; the self-citations that exist are minor and not load-bearing.

full rationale

The paper's derivation chain is self-contained and non-circular. The supervised target is the transient WSS field produced by independent SimVascular CFD solves of the incompressible Navier-Stokes equations (Sec. 2.1.2) with a literature template inflow waveform [60], rigid walls, and fixed Windkessel outlets, applied to both training and external-test geometries; no equation in Secs. 2.2-2.4 defines the prediction in terms of the labels, and the trained model receives only geometry descriptors plus the scalar Vmax, so the reported errors (external MAE 0.296 Pa, Table 1) are genuine empirical measurements. External validity is strong: the AAA-L test set (118 scans from a separate Seoul cohort [24,22]) and the MultiViewUNet [17] baseline are independent of the authors. The self-citations (LaB-GATr [55], LaB-VaTr [56], SIRE [42], AAA-100 [41]) are present but minor and not load-bearing: the architecture's state-of-the-art claim is referenced from the authors' own prior work, yet the paper's central generalisation claims are verified in this paper against independently generated CFD, including a zero-shot topology test with 10 held-out geometries deliberately excluded from training to avoid data leakage (Sec. 3.5.3). The vector-heat-method flow prior (Sec. 2.2.2) is computed from the geometry and inlet direction alone, not from the target WSS fields, and the Table 3 ablation shows it provides only a modest gain, so it is an inductive bias rather than a smuggled label. The remaining concerns are correctness and scope issues rather than circularity: the Discussion explicitly admits the lack of patient-specific flow measurements and the use of fixed Windkessel parameters, which limits clinical transferability of the CFD-based ground truth, and the 'boundary conditions' generalisation (Fig. 5) varies only the amplitude Qmax of one template waveform while the model is conditioned only on Vmax, so the abstract's wording overstates what was tested. Neither issue makes the prediction reduce to its inputs by construction, so the circularity finding is negative.

Assumptions & free parameters 5 free parameters · 5 assumptions · 0 invented entities

No new physical entities are introduced. The model's input descriptors and geometric algebra multivectors are computational representations, not postulated physical objects. The free parameters are hand-selected training and architecture choices; the main domain assumptions concern the CFD ground truth and the geometric priors.

free parameters (5)
  • loss balance weight lambda = 0.1
    Used to combine angle cosine loss and magnitude L1 loss in L_total (Section 2.3); chosen by hand without reported sensitivity analysis.
  • tokenisation rate = 0.1
    Fraction of surface points kept in the coarse point cloud for cross-attention (Section 3); follows LaB-GATr setting.
  • interpolation neighborhood k = 3
    Number of nearest coarse points used for inverse-distance interpolation to recover full-resolution predictions (Section 2.2.1).
  • architecture depth and width = 10 blocks, 4 heads, hidden size 8
    Model capacity choices reported in Section 3; not fitted to data but hand-selected.
  • training schedule = learning rate 3e-4, gamma 0.9989, dropout 0.2, 5000 epochs, batch size 16
    Standard training hyperparameters (Section 3); no sensitivity analysis.
assumptions (5)
  • domain assumption CFD simulations are valid ground truth for patient hemodynamics
    SimVascular Navier-Stokes with rigid walls, no-slip, Newtonian blood (mu=0.04, rho=1.06), template inflow from cited literature, and Windkessel outlets; used as supervised labels (Section 2.1.2). Discussion admits no patient-specific flow measurements.
  • domain assumption WSS is E(3)-equivariant in this setting
    Gravity and ambient orientation are ignored in the CFD formulation, so the model is designed to be equivariant to rotation, translation, and reflection (Section 2.2). If this fails, the architecture discards relevant information.
  • domain assumption Surface point cloud adequately represents the vessel wall manifold
    Mesh connectivity is discarded; descriptors are computed per vertex and interpolated through k-nearest coarse points (Section 2.2.1). Accuracy depends on sampling and mesh quality.
  • ad hoc to paper Flow prior approximates WSS direction
    The inlet direction is parallel-transported via the vector heat method to create a tangential flow prior (Section 2.2.2). This assumes perfect laminar direction is a useful predictor of real WSS direction.
  • domain assumption Geodesic distance maps to inlet and nearest outlet capture flow topology
    Minimum aggregation over outlets is used to make the descriptor topology-agnostic (Section 2.2.2); assumes this summary carries enough hemodynamic information.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Wall Shear Stress Estimation in Abdominal Aortic Aneurysms: Towards Generalisable Neural Surrogate Models." pith.science (2026). https://pith.science/paper/DNFKQG7S

@misc{pith2026250722817,
  author       = {Pith},
  title        = {Pith review of: Wall Shear Stress Estimation in Abdominal Aortic Aneurysms: Towards Generalisable Neural Surrogate Models},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/DNFKQG7S}},
  note         = {Machine review of arXiv:2507.22817}
}
read the original abstract

Abdominal aortic aneurysms (AAAs) are pathologic dilatations of the abdominal aorta posing a high fatality risk upon rupture. Studying AAA progression and rupture risk often involves in-silico blood flow modelling with computational fluid dynamics (CFD) and extraction of hemodynamic factors like time-averaged wall shear stress (TAWSS) or oscillatory shear index (OSI). However, CFD simulations are known to be computationally demanding. Hence, in recent years, geometric deep learning methods, operating directly on 3D shapes, have been proposed as compelling surrogates, estimating hemodynamic parameters in just a few seconds. In this work, we propose a geometric deep learning approach to estimating hemodynamics in AAA patients, and study its generalisability to common factors of real-world variation. We propose an E(3)-equivariant deep learning model utilising novel robust geometrical descriptors and projective geometric algebra. Our model is trained to estimate transient WSS using a dataset of CT scans of 100 AAA patients, from which lumen geometries are extracted and reference CFD simulations with varying boundary conditions are obtained. Results show that the model generalizes well within the distribution, as well as to the external test set. Moreover, the model can accurately estimate hemodynamics across geometry remodelling and changes in boundary conditions. Furthermore, we find that a trained model can be applied to different artery tree topologies, where new and unseen branches are added during inference. Finally, we find that the model is to a large extent agnostic to mesh resolution. These results show the accuracy and generalisation of the proposed model, and highlight its potential to contribute to hemodynamic parameter estimation in clinical practice.

Figures

Figures reproduced from arXiv: 2507.22817 by the authors.

Figure 1
Figure 1. Workflow of hemodynamics estimation with CFD or deep-learning-based surrogate model. Both approaches take the CTA-obtained 3D geometry and inflow waveform as the input and yield transient WSS. Substituting the SimVascular CFD solver with a trained LaB-GATr model allows for 1500x speedup in hemodynamics estimation. We study the generalisation capabilities of our model in four settings: changing physiological conditio… view at source ↗
Figure 2
Figure 2. Input features of AAA geometry embedded through geometric objects of the geometric algebra 𝐆(3, 0, 1) [55, 11]. Geometry encoded in such a manner serves as the input to the LaB-GATr model that respects all Euclidean symmetries, i.e., rotation, translation and reflection. The time-resolved inflow waveform is encoded by taking the maximum inflow 𝑄max and computing maximum inlet velocity magnitude 𝑉max with inlet area … view at source ↗
Figure 3
Figure 3. Qualitative comparison of CFD and LaB-GATr-obtained transient WSS field - for clarity we plot only every second timepoint. Based on transient WSS, TAWSS and OSI hemodynamic markers can be computed. Two regions are zoomed in to showcase the correspondence of predicted and ground truth WSS vector directions. with the highest errors appearing in smaller branches - the iliac and renal arteries. For the estimation of OSI… view at source ↗
Figures from the paper (5 more)
Figure 4
Figure 4. Figure 4: Qualitative comparison of CFD- and LaB-GATr-derived TAWSS and OSI fields. From the left, respectively, samples approximately at the first quartile (Q1), the second quartile (Q2 - median) and the third quartile (Q3) of the MAE distribution for TAWSS and OSI [PITH_FULL_…
Figure 6
Figure 6. Figure 6: Consistency of Q1 TAWSS and Q3 OSI estimation over longitudinal patient imaging. We showcase a patient with 7 longitudinal scans over the period of almost 10 years. characteristic length of AAA (i.e., its local radius), which is proportional to the Reynolds number, gro…
Figure 5
Figure 5. Figure 5: Distribution of Approx. Disp. for TAWSS and OSI across the samples with respect to the inflow rate and AAA volume. In [PITH_FULL_IMAGE:figures/full_fig_p010_5.png]
Figure 7
Figure 7. Figure 7: Qualitative comparison of TAWSS prediction be￾tween original and extended geometry with corresponding CFD-obtained ground truth. or changes to its architecture. We utilise the SIRE frame￾work [42, 3] to extend 10 shapes from the AAA-100 to additionally include the supe…
Figure 8
Figure 8. Figure 8: Consistency of LaB-GATr predictions across different mesh resolutions. The model was trained on meshes with a resolution (global edge length) of 1.25 mm - denoted with original. Visual comparison shows a predicted TAWSS field with the underlying surface discretisations…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

77 extracted references · 61 canonical work pages

  1. [60]

    Finite Element Modeling of Three-Dimensional Pulsatile Flow in the Abdominal P

    Taylor, C.A., Hughes, T.J.R., Zarins, C.K., 1998. Finite Element Modeling of Three-Dimensional Pulsatile Flow in the Abdominal P. Rygiel et al.:Preprint submitted to Elsevier Page 16 of 17 Wall Shear Stress Estimation in Abdominal Aortic Aneurysms: Towards Generalisable Neural Surrogate Models Aorta:RelevancetoAtherosclerosis. AnnalsofBiomedicalEngineer- ...

    work page doi:10.1114/1.140 1998

  2. [1]

    (Eds.), Statistical Atlases and Computational Models of the Heart

    Alblas,D.,Brune,C.,Yeung,K.K.,Wolterink,J.M.,2022.GoingOff- Grid: Continuous Implicit Neural Representations for 3D Vascular Modeling, in: Camara, O., Puyol-Antón, E., Qin, C., Sermesant, M., Suinesiaputra, A., Wang, S., Young, A. (Eds.), Statistical Atlases and Computational Models of the Heart. Regular and CMRxMotion Challenge Papers, Springer Nature Sw...

    work page 2022

  3. [2]

    Geometric deep learning for local growth prediction on abdominal aortic aneurysm surfaces

    Alblas, D., Rygiel, P., Suk, J., Kappe, K.O., Hofman, M., Brune, C., Yeung, K.K., Wolterink, J.M., 2025a. Geometric deep learning for localgrowthpredictiononabdominalaorticaneurysmsurfaces.URL: https://arxiv.org/abs/2506.08729, arXiv:2506.08729

    work page Pith review arXiv

  4. [3]

    SIRE: Scale-invariant, rotation-equivariant estimation of artery orientations using graph neural networks

    Alblas, D., Suk, J., Brune, C., Yeung, K.K., Wolterink, J.M., 2025b. SIRE: Scale-invariant, rotation-equivariant estimation of artery orientations using graph neural networks. Medical Im- age Analysis 101, 103467. URL: https://www.sciencedirect. com/science/article/pii/S1361841525000155,doi: https://doi.org/10. 1016/j.media.2025.103467

    arXiv

  5. [4]

    Universal physics transformers: A framework for efficiently scaling neural operators, in: The Thirty-eighth Annual ConferenceonNeuralInformationProcessingSystems

    Alkin, B., Fürst, A., Schmid, S.L., Gruber, L., Holzleitner, M., Brandstetter, J., 2024. Universal physics transformers: A framework for efficiently scaling neural operators, in: The Thirty-eighth Annual ConferenceonNeuralInformationProcessingSystems. URL: https: //openreview.net/forum?id=oUXiNX5KRm

    work page 2024

  6. [5]

    Abdominal aortic aneurysm screening is safe yet lacks effectiveness

    Arnaoutakis, D.J., Upchurch, G.R., 2019. Abdominal aortic aneurysm screening is safe yet lacks effectiveness. Circulation 139, 1381–1383. URL: https://www.ahajournals.org/doi/abs/10.1161/ CIRCULATIONAHA.118.038809, doi:10.1161/CIRCULATIONAHA.118.038809

    work page doi:10.1161/circulationaha.118.038809 2019

  7. [6]

    Ma- chine Learning for Cardiovascular Biomechanics Modeling: Chal- lenges and Beyond

    Arzani, A., Wang, J.X., Sacks, M.S., Shadden, S.C., 2022. Ma- chine Learning for Cardiovascular Biomechanics Modeling: Chal- lenges and Beyond. Annals of Biomedical Engineering 50, 615–

    work page 2022

  8. [7]

    NeuralCFD: Deep Learning on High-Fidelity AutomotiveAerodynamicsSimulations.URL: http://arxiv.org/abs/ 2502.09692, doi:10.48550/arXiv.2502.09692

    Bleeker, M., Dorfer, M., Kronlachner, T., Sonnleitner, R., Alkin, B., Brandstetter, J., 2025. NeuralCFD: Deep Learning on High-Fidelity AutomotiveAerodynamicsSimulations.URL: http://arxiv.org/abs/ 2502.09692, doi:10.48550/arXiv.2502.09692. arXiv:2502.09692 [cs]

    doi:10.48550/arxiv.2502.09692 2025

Show all 77 references
  1. [8]

    Low wall shear stress predominates at sites of abdominal aor- tic aneurysm rupture

    Boyd, A.J., Kuhn, D.C., Lozowy, R.J., Kulbisky, G.P., 2016. Low wall shear stress predominates at sites of abdominal aor- tic aneurysm rupture. Journal of Vascular Surgery 63, 1613–

    2016

  2. [9]

    Geometricandphysicalquantitiesimprovee(3)equivariant messagepassing,in:InternationalConferenceonLearningRepresen- tations

    Brandstetter,J.,Hesselink,R.,vanderPol,E.,Bekkers,E.J.,Welling, M.,2022. Geometricandphysicalquantitiesimprovee(3)equivariant messagepassing,in:InternationalConferenceonLearningRepresen- tations. URL: https://openreview.net/forum?id=_xwr8gOBeV1

    2022

  3. [10]

    Does equivariance matter at scale? URL: https://arxiv.org/abs/2410

    Brehmer, J., Behrends, S., de Haan, P., Cohen, T., 2024. Does equivariance matter at scale? URL: https://arxiv.org/abs/2410. 23179, arXiv:2410.23179. P. Rygiel et al.:Preprint submitted to Elsevier Page 14 of 17 Wall Shear Stress Estimation in Abdominal Aortic Aneurysms: Towar...

    2024 arXiv

  4. [11]

    Geometric algebra transformer, in: Thirty-seventh Conference on Neural Infor- mationProcessingSystems

    Brehmer, J., Haan, P.D., Behrends, S., Cohen, T., 2023. Geometric algebra transformer, in: Thirty-seventh Conference on Neural Infor- mationProcessingSystems. URL: https://openreview.net/forum?id= M7r2CO4tJC

    2023

  5. [12]

    A Review of Computational Methods to Predict the Risk of Rupture of AbdominalAorticAneurysms

    Canchi, T., Kumar, S.D., Ng, E.Y.K., Narayanan, S., 2015. A Review of Computational Methods to Predict the Risk of Rupture of AbdominalAorticAneurysms. BioMedResearchInternational2015, 1–12. URL: http://www.hindawi.com/journals/bmri/2015/861627/, doi:10.1155/2015/861627. publi...

    2015 doi

  6. [13]

    A Review of Biological Factors Impli- cated in Abdominal Aortic Aneurysm Rupture

    Choke, E., Cockerill, G., Wilson, W., Sayed, S., Dawson, J., Loftus, I., Thompson, M., 2005. A Review of Biological Factors Impli- cated in Abdominal Aortic Aneurysm Rupture. European Journal of Vascular and Endovascular Surgery 30, 227–244. URL:https: //linkinghub.elsevier.co...

    2005

  7. [14]

    The heat method for distance computation

    Crane, K., Weischedel, C., Wardetzky, M., 2017. The heat method for distance computation. Communications of the ACM 60, 90–99. URL: https://dl.acm.org/doi/10.1145/3131280, doi:10.1145/3131280. publisher: Association for Computing Machinery (ACM)

    2017 doi

  8. [15]

    Autopsy study of unoperated abdominal aortic aneurysms: The case for early resection

    Darling, R.C., Messina, C.R., Brewster, D.C., Ottinger, L.W., 1977. Autopsy study of unoperated abdominal aortic aneurysms: The case for early resection. Circulation 56, II–161–II–164. URL: https: //api.semanticscholar.org/CorpusID:43267252

    1977

  9. [16]

    Endothelial Dysfunction in the Pathogenesis of Abdominal Aortic Aneurysm

    DeRoo, E., Stranz, A., Yang, H., Hsieh, M., Se, C., Zhou, T., 2022. Endothelial Dysfunction in the Pathogenesis of Abdominal Aortic Aneurysm. Biomolecules 12, 509. URL: https://www.mdpi.com/ 2218-273X/12/4/509, doi:10.3390/biom12040509. publisher: MDPI AG

    2022 doi

  10. [17]

    Rapid wall shear stress predic- tion for aortic aneurysms using deep learning: a fast alternative to CFD

    Faisal, M.A.A., Mutlu, O., Mahmud, S., Tahir, A., Chowdhury, M.E.H., Bensaali, F., Alnabti, A., Yavuz, M.M., El-Menyar, A., Al- Thani, H., Yalcin, H.C., 2025. Rapid wall shear stress predic- tion for aortic aneurysms using deep learning: a fast alternative to CFD. Medical & Bi...

    2025

  11. [18]

    Ferdian, E., Dubowitz, D.J., Mauger, C.A., Wang, A., Young, A.A.,

  12. [19]

    A Novel Strategy to Translate the Biomechanical Rupture Risk of Abdominal Aortic Aneurysms to their Equivalent Diameter Risk: Method and Retrospective Val- idation

    Gasser, T.C., Nchimi, A., Swedenborg, J., Roy, J., Sakalihasan, N., Böckler, D., Hyhlik-Dürr, A., 2014. A Novel Strategy to Translate the Biomechanical Rupture Risk of Abdominal Aortic Aneurysms to their Equivalent Diameter Risk: Method and Retrospective Val- idation. European...

    2014 doi

  13. [20]

    Endothelial dysfunction in cardiovascular diseases: mech- anisms and in vitro models

    Grego, A., Fernandes, C., Fonseca, I., Dias-Neto, M., Costa, R., Leite-Moreira, A., Oliveira, S.M., Trindade, F., Nogueira-Ferreira, R., 2025. Endothelial dysfunction in cardiovascular diseases: mech- anisms and in vitro models. Molecular and Cellular Biochem- istry URL: https...

    2025 doi

  14. [21]

    Gauge equiv- ariant mesh cnns: Anisotropic convolutions on geometric graphs, in: International Conference on Learning Representations

    de Haan, P., Weiler, M., Cohen, T., Welling, M., 2021. Gauge equiv- ariant mesh cnns: Anisotropic convolutions on geometric graphs, in: International Conference on Learning Representations. URL:https: //openreview.net/forum?id=Jnspzp-oIZE

    2021

  15. [22]

    ADeepLearn- ing Approach to Predict Abdominal Aortic Aneurysm Expansion Using Longitudinal Data

    Jiang,Z.,Do,H.N.,Choi,J.,Lee,W.,Baek,S.,2020. ADeepLearn- ing Approach to Predict Abdominal Aortic Aneurysm Expansion Using Longitudinal Data. Frontiers in Physics 7. URL:https://www. frontiersin.org/article/10.3389/fphy.2019.00235/full,doi: 10.3389/ fphy.2019.00235. publisher...

    2020

  16. [23]

    Suggested standards for reporting on arterial aneurysms

    Johnston, K.W., Rutherford, R.B., Tilson, M.D., Shah, D.M., Hol- lier, L., Stanley, J.C., 1991. Suggested standards for reporting on arterial aneurysms. Journal of Vascular Surgery 13, 452–

    1991

  17. [24]

    Deep Learning on Multiphysical Features and Hemodynamic Modeling for Abdominal Aortic Aneurysm Growth Prediction

    Kim, S., Jiang, Z., Zambrano, B.A., Jang, Y., Baek, S., Yoo, S., Chang, H.J., 2023. Deep Learning on Multiphysical Features and Hemodynamic Modeling for Abdominal Aortic Aneurysm Growth Prediction. IEEE Transactions on Medical Imaging 42, 196–208. URL: https://ieeexplore.ieee....

    2023

  18. [25]

    Prediction of 3D Cardiovascular hemodynamics before and after coronary artery bypass surgery via deep learning

    Li, G., Wang, H., Zhang, M., Tupin, S., Qiao, A., Liu, Y., Ohta, M., Anzai, H., 2021. Prediction of 3D Cardiovascular hemodynamics before and after coronary artery bypass surgery via deep learning. Communications Biology 4, 99. URL: https://www.nature.com/ articles/s42003-020-...

    2021 doi

  19. [26]

    A deep learning approach to estimate stress distribution: a fast and accurate surrogate offinite-elementanalysis

    Liang, L., Liu, M., Martin, C., Sun, W., 2018. A deep learning approach to estimate stress distribution: a fast and accurate surrogate offinite-elementanalysis. JournalofTheRoyalSocietyInterface15, 20170844. URL: https://royalsocietypublishing.org/doi/10.1098/ rsif.2017.0844,d...

    2018

  20. [27]

    Journal of Biomechanics 99, 109544

    Liang,L.,Mao,W.,Sun,W.,2020.Afeasibilitystudyofdeeplearning for predicting hemodynamics of human thoracic aorta. Journal of Biomechanics 99, 109544. URL: https://www.sciencedirect. com/science/article/pii/S0021929019308012,doi: https://doi.org/10. 1016/j.jbiomech.2019.109544

    2020

  21. [28]

    Global Burden of Cardiovascular Dis- eases and Risks Collaboration, 1990-2021

    Lindstrom, M., DeCleene, N., Dorsey, H., Fuster, V., Johnson, C.O., LeGrand, K.E., Mensah, G.A., Razo, C., Stark, B., Varieur Turco, J., Roth, G.A., 2022. Global Burden of Cardiovascular Dis- eases and Risks Collaboration, 1990-2021. Journal of the American College of Cardiolo...

    2022

  22. [29]

    An artificial intelligence-based platform for automatically estimating time-averaged wall shear stress in the ascending aorta

    Lv, L., Li, H., Wu, Z., Zeng, W., Hua, P., Yang, S., 2022. An artificial intelligence-based platform for automatically estimating time-averaged wall shear stress in the ascending aorta. European Heart Journal - Digital Health 3, 525–534. URL: https://doi.org/10.1093/ehjdh/ztac...

    2022 doi

  23. [30]

    A Comprehensive Review on Com- putational Analysis, Research Advances, and Major Findings on Ab- dominalAorticAneurysmsfortheYears2021to2023.AnnalsofVas- cular Surgery 110, 63–81

    Manta, A., Tzirakis, K., 2025. A Comprehensive Review on Com- putational Analysis, Research Advances, and Major Findings on Ab- dominalAorticAneurysmsfortheYears2021to2023.AnnalsofVas- cular Surgery 110, 63–81. URL:https://linkinghub.elsevier.com/ retrieve/pii/S089050962400566...

    2025 doi

  24. [31]

    Local Diameter, Wall Stress, and Thrombus Thickness Influence the Local Growth of Abdominal Aortic Aneurysms

    Martufi, G., Liljeqvist, M.L., Sakalihasan, N., Panuccio, G., Hultgren, R., Roy, J., Gasser, T.C., 2016. Local Diameter, Wall Stress, and Thrombus Thickness Influence the Local Growth of Abdominal Aortic Aneurysms. Journal of Endovascular Therapy 23, 957–966. URL: https://doi....

    2016 doi

  25. [32]

    Discrete Differential-Geometry Operators for Triangulated 2-Manifolds, in: Hege, H.C., Polthier, K

    Meyer, M., Desbrun, M., Schröder, P., Barr, A.H., 2003. Discrete Differential-Geometry Operators for Triangulated 2-Manifolds, in: Hege, H.C., Polthier, K. (Eds.), Visualization and Mathematics III, SpringerBerlinHeidelberg,Berlin,Heidelberg.pp.35–57. doi: https: //doi.org/10....

    2003 doi

  26. [33]

    Deep Learning Framework for Real-Time Estimation of in-silico Thrombotic Risk Indices in the Left Atrial Appendage

    Morales, X., Mill, J., Sørensen, K., Acebes, C., Iriart, X., Legghe, B., Cochet, H., De Backer, O., Paulsen, R., Camara, O., 2021. Deep Learning Framework for Real-Time Estimation of in-silico Thrombotic Risk Indices in the Left Atrial Appendage. Frontiers in Physiology 12, 69...

    2021

  27. [34]

    Mutlu,O.,Salman,H.E.,Al-Thani,H.,El-Menyar,A.,Qidwai,U.A., Yalcin, H.C., 2023. How does hemodynamics affect rupture tissue mechanics in abdominal aortic aneurysm: Focus on wall shear stress derivedparameters,time-averagedwallshearstress,oscillatoryshear index, endothelial cell...

    2023

  28. [35]

    Cardiovascular Disease

    Nabel, E.G., 2003. Cardiovascular Disease. New England Journal of Medicine 349, 60–72. URL: https://www.nejm.org/ doi/full/10.1056/NEJMra035098, doi: 10.1056/NEJMra035098. _eprint: https://www.nejm.org/doi/pdf/10.1056/NEJMra035098

    2003 doi

  29. [36]

    Learning hemodynamic scalar fields on coronary artery meshes: A benchmark of geometric deep learning models

    Nannini, G., Suk, J., Rygiel, P., Saitta, S., Mariani, L., Maranga, R., Baggiano, A., Pontone, G., Wolterink, J.M., Redaelli, A., 2025. Learning hemodynamic scalar fields on coronary artery meshes: A benchmark of geometric deep learning models. Computers in Biol- ogy and Medic...

    2025

  30. [37]

    Parkinson, F., Ferguson, S., Lewis, P., Williams, I.M., Twine, C.P.,

  31. [38]

    Qi, C.R., Yi, L., Su, H., Guibas, L.J., 2017. PointNet++: deep hierarchical feature learning on point sets in a metric space, in: Pro- ceedings of the 31st International Conference on Neural Information Processing Systems, Curran Associates Inc., Red Hook, NY, USA. pp. 5105–51...

    2017 doi arXiv

  32. [39]

    British Journal of Surgery 100, 1405–1413

    Reimerink,J.J., vanderLaan,M.J., Koelemay,M.J.,Balm, R.,Lege- mate,D.A.,2013.Systematicreviewandmeta-analysisofpopulation- based mortality from ruptured abdominal aortic aneurysm. British Journal of Surgery 100, 1405–1413. URL:https://doi.org/10.1002/ bjs.9235, doi:10.1002/bjs.9235

    2013 doi

  33. [40]

    Wall Shear Stress – an Important Determinant of Endothelial Cell Function and Structure – in the Arterial System in vivo: Discrepancies with Theory

    Reneman, R.S., Arts, T., Hoeks, A.P., 2006. Wall Shear Stress – an Important Determinant of Endothelial Cell Function and Structure – in the Arterial System in vivo: Discrepancies with Theory. Journal of Vascular Research 43, 251–269. URL: https://doi.org/10.1159/000091648, do...

    2006 doi

  34. [41]

    AAA-100: A curated dataset of 3d watertight abdominal aortic aneurysm models

    Rygiel, P., Alblas, D., Brune, C., Smorenburg, S., Yeung, K.K., Wolterink, J.M., 2024. AAA-100: A curated dataset of 3d watertight abdominal aortic aneurysm models. URL:https://doi.org/10.5281/ zenodo.10932957, doi:10.5281/zenodo.10932957

    2024 doi

  35. [42]

    Rygiel, P., Alblas, D., Brune, C., Yeung, K.K., Wolterink, J.M., 2025a. Global Control for Local SO(3)-Equivariant Scale-Invariant Vessel Segmentation, in: Camara, O., Puyol-Antón, E., Sermesant, M., Suinesiaputra, A., Zhao, J., Wang, C., Tao, Q., Young, A. (Eds.), Statistical...

  36. [43]

    Rygiel, P., Płuszka, P., Zięba, M., Konopczyński, T., 2023. Center- linePointNet++:ANewPointCloudBasedArchitectureforCoronary ArteryPressure DropandvFFREstimation, in:Greenspan,H., Mad- abhushi,A.,Mousavi,P.,Salcudean,S.,Duncan,J.,Syeda-Mahmood, T., Taylor, R. (Eds.), Medical ...

    2023 doi

  37. [44]

    Active learning for deep learning-based hemodynamic parameter es- timation

    Rygiel, P., Suk, J., Yeung, K.K., Brune, C., Wolterink, J.M., 2025b. Active learning for deep learning-based hemodynamic parameter es- timation. URL:https://arxiv.org/abs/2503.03453, arXiv:2503.03453

    arXiv

  38. [45]

    The Vector Heat Method

    Sharp, N., Soliman, Y., Crane, K., 2019. The Vector Heat Method. ACM Trans. Graph. 38. URL: https://doi.org/10.1145/3243651, doi:10.1145/3243651. place: New York, NY, USA Publisher: Associa- tion for Computing Machinery

    2019 doi

  39. [46]

    Tetgen, a delaunay-based quality tetrahedral mesh generator

    Si, H., 2015. Tetgen, a delaunay-based quality tetrahedral mesh generator. ACM Transactions on Mathematical Software 41. URL: https://doi.org/10.1145/2629697, doi:10.1145/2629697

    2015 doi

  40. [47]

    Song, P., He, Y., Adeloye, D., Zhu, Y., Ye, X., Yi, Q., Rahimi, K., Rudan, I., Global Health Epidemiology Research Group (GHERG),

  41. [48]

    CFD Modelling of Abdominal Aortic Aneurysm on Hemodynamic Loads Using a Realistic Geometry with CT

    Soudah, E., Ng, E.Y.K., Loong, T.H., Bordone, M., Pua, U., Narayanan, S., 2013. CFD Modelling of Abdominal Aortic Aneurysm on Hemodynamic Loads Using a Realistic Geometry with CT. Computational and Mathematical Methods in Medicine 2013, 472564. URL: https://onlinelibrary.wiley...

    2013 doi

  42. [49]

    Biomechanical changes during abdominal aortic aneurysm growth

    Stevens, R.R., Grytsan, A., Biasetti, J., Roy, J., Lindquist Liljeqvist, M., Gasser, T.C., 2017. Biomechanical changes during abdominal aortic aneurysm growth. PLoS One 12, e0187421. doi: 10.1371/ journal.pone.0187421

    2017

  43. [50]

    Computers in Biology and Medicine 126, 104038

    Su,B.,Zhang,J.M.,Zou,H.,Ghista,D.,Le,T.T.,Chin,C.,2020.Gen- erating wall shear stress for coronary artery in real-time using neural networks: Feasibility and initial results based on idealized models. Computers in Biology and Medicine 126, 104038. URL: https: //linkinghub.else...

    2020

  44. [51]

    Physics-informed graph neural networks for flow field estimation in carotid arteries

    Suk, J., Alblas, D., Hutten, B.A., Wiegman, A., Brune, C., van Ooij, P., Wolterink, J.M., 2024a. Physics-informed graph neural networks for flow field estimation in carotid arteries. arXiv preprint arXiv:2408.07110

  45. [52]

    SE(3) Symmetry Lets Graph Neural Networks Learn Arterial Velocity Estimation from SmallDatasets,in:Bernard,O.,Clarysse,P.,Duchateau,N.,Ohayon, J., Viallon, M

    Suk, J., Brune, C., Wolterink, J.M., 2023. SE(3) Symmetry Lets Graph Neural Networks Learn Arterial Velocity Estimation from SmallDatasets,in:Bernard,O.,Clarysse,P.,Duchateau,N.,Ohayon, J., Viallon, M. (Eds.), Functional Imaging and Modeling of the Heart. Springer Nature Switz...

    2023

  46. [53]

    Geometric algebratransformersforlarge3dmeshesviacross-attention,in:ICML 2024 Workshop on Geometry-grounded Representation Learning and Generative Modeling

    Suk, J., Haan, P.D., Imre, B., Wolterink, J.M., 2024b. Geometric algebratransformersforlarge3dmeshesviacross-attention,in:ICML 2024 Workshop on Geometry-grounded Representation Learning and Generative Modeling. URL: https://openreview.net/forum?id= T2bBUlaJTA

    2024

  47. [54]

    Mesh neural networks for SE(3)-equivariant hemodynamics estima- tion on the artery wall

    Suk, J., Haan, P.d., Lippe, P., Brune, C., Wolterink, J.M., 2024c. Mesh neural networks for SE(3)-equivariant hemodynamics estima- tion on the artery wall. Computers in Biology and Medicine 173, 108328. URL: https://www.sciencedirect.com/science/article/pii/ S0010482524004128,...

    doi

  48. [55]

    Annals of Surgery 277, 912–919

    The Global and Regional Prevalence of Abdominal Aortic Aneurysms: A Systematic Review and Modeling Analysis. Annals of Surgery 277, 912–919. doi:10.1097/SLA.0000000000005716

    doi

  49. [56]

    Deep vectorised operators for pulsatile hemodynamics estimation in coronary arteries from a steady-state prior

    Suk, J., Nannini, G., Rygiel, P., Brune, C., Pontone, G., Redaelli, A., Wolterink, J.M., 2024e. Deep vectorised operators for pulsatile hemodynamics estimation in coronary arteries from a steady-state prior. URL:https://arxiv.org/abs/2410.11920, arXiv:2410.11920

    arXiv

  50. [57]

    PyVista: 3d plotting and mesh analysis through a streamlined interface for the visualization toolkit (VTK)

    Sullivan, C.B., Kaszynski, A., 2019. PyVista: 3d plotting and mesh analysis through a streamlined interface for the visualization toolkit (VTK). Journal of Open Source Software 4, 1450. URL: https: //doi.org/10.21105/joss.01450, doi:10.21105/joss.01450

    2019 doi

  51. [58]

    Meta-analysis of individual patient data to examine factors affecting growth and rupture of small abdominal aortic aneurysms

    Sweeting, M.J., Thompson, S.G., Brown, L.C., Powell, J T, o.b.o.t.R.c., 2012. Meta-analysis of individual patient data to examine factors affecting growth and rupture of small abdominal aortic aneurysms. British Journal of Surgery 99, 655–665. URL: https://doi.org/10.1002/bjs....

    2012 doi

  52. [59]

    Deep Learning for Computational Hemodynamics: A Brief Review of Recent Advances

    Taebi, A., 2022. Deep Learning for Computational Hemodynamics: A Brief Review of Recent Advances. Fluids 7. URL:https://www. mdpi.com/2311-5521/7/6/197, doi:10.3390/fluids7060197

    2022 doi

  53. [61]

    Wall shear stress and rela- tive residence time as potential risk factors for abdominal aortic aneurysms in males: a 4D flow cardiovascular magnetic resonance case-control study

    Trenti, C., Ziegler, M., Bjarnegà ¥rd, N., Ebbers, T., Linden- berger, M., Dyverfeldt, P., 2022. Wall shear stress and rela- tive residence time as potential risk factors for abdominal aortic aneurysms in males: a 4D flow cardiovascular magnetic resonance case-control study. J...

    2022 doi

  54. [62]

    SimVascular: An Open Source Pipeline for Cardiovascular Simulation

    Updegrove, A., Wilson, N.M., Merkow, J., Lan, H., Marsden, A.L., Shadden, S.C., 2017. SimVascular: An Open Source Pipeline for Cardiovascular Simulation. Annals of Biomedical Engineering 45, 525–541. URL:https://doi.org/10.1007/s10439-016-1762-8,doi: 10. 1007/s10439-016-1762-8

    2017 doi

  55. [63]

    Attention is All you Need, in: Guyon, I., Luxburg, U.V., Bengio, S., Wal- lach, H., Fergus, R., Vishwanathan, S., Garnett, R

    Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A.N., Kaiser, L., Polosukhin, I., 2017. Attention is All you Need, in: Guyon, I., Luxburg, U.V., Bengio, S., Wal- lach, H., Fergus, R., Vishwanathan, S., Garnett, R. (Eds.), Ad- vancesinNeuralInformationPro...

    2017

  56. [64]

    Suk, J., Imre, B., Wolterink, J.M., 2024d. LaB-GATr: Geometric Algebra Transformers for Large Biomedical Surface and Volume Meshes, in: Linguraru, M.G., Dou, Q., Feragen, A., Giannarou, S., Glocker, B., Lekadir, K., Schnabel, J.A. (Eds.), Medical Image Com- putingandComputerAs...

  57. [65]

    Editor’schoice–europeansocietyforvascular surgery (ESVS) 2024 clinicalpractice guidelines on the management of abdominal Aorto-Iliac artery aneurysms

    Wanhainen, A., Van Herzeele, I., Bastos Goncalves, F., Bell- muntMontoya,S.,Berard,X.,Boyle,J.R.,D’Oria,M.,Prendes,C.F., Karkos,C.D.,Kazimierczak,A.,Koelemay,M.J.W.,Kölbel,T.,Mani, K., Melissano, G., Powell, J.T., Trimarchi, S., Tsilimparis, N., ESVS GuidelinesCommittee,Antoni...

    2024

  58. [66]

    The arterial Windkessel

    Westerhof, N., Lankhaar, J.W., Westerhof, B.E., 2009. The arterial Windkessel. Medical & Biological Engineering & Computing 47, 131–141. URL:https://doi.org/10.1007/s11517-008-0359-2,doi: 10. 1007/s11517-008-0359-2

    2009 doi

  59. [67]

    Deep Learning Based Centerline-Aggregated Aortic Hemodynamics: An Efficient Alternative to Numerical Modeling of Hemodynamics

    Yevtushenko, P., Goubergrits, L., Gundelwein, L., Setio, A., Ramm, H., Lamecker, H., Heimann, T., Meyer, A., Kuehne, T., Schafstedde, M., 2022. Deep Learning Based Centerline-Aggregated Aortic Hemodynamics: An Efficient Alternative to Numerical Modeling of Hemodynamics. IEEE J...

    2022

  60. [68]

    Physics-informed neural networks(pinns)for4dhemodynamicsprediction:Aninvestigationof optimalframeworkbasedonvascularmorphology

    Zhang, X., Mao, B., Che, Y., Kang, J., Luo, M., Qiao, A., Liu, Y., Anzai, H., Ohta, M., Guo, Y., Li, G., 2023. Physics-informed neural networks(pinns)for4dhemodynamicsprediction:Aninvestigationof optimalframeworkbasedonvascularmorphology. ComputersinBi- ology and Medicine 164,...

    2023

  61. [73]

    series Title: Lecture Notes in Computer Science

  62. [74]

    Biomechanics of abdominal aortic aneurysm

    Vorp, D.A., 2007. Biomechanics of abdominal aortic aneurysm. Journal of Biomechanics 40, 1887–1902. URL: https: //www.sciencedirect.com/science/article/pii/S002192900600323X, doi:https://doi.org/10.1016/j.jbiomech.2006.09.003

    2007 doi

  63. [454]

    series Title: Lecture Notes in Computer Science

    URL: https://link.springer.com/10.1007/978-3-031-35302-4_ 46, doi:10.1007/978-3-031-35302-4_46 . series Title: Lecture Notes in Computer Science

    doi

  64. [458]

    URL: https://www.sciencedirect.com/science/article/pii/ 0741521491700518, doi:https://doi.org/10.1067/mva.1991.26737

    1991

  65. [627]

    1007/s10439-022-02967-4

    URL: https://doi.org/10.1007/s10439-022-02967-4, doi:10. 1007/s10439-022-02967-4

    doi

  66. [1619]

    URL: https://www.sciencedirect.com/science/article/pii/ S0741521415001500, doi:https://doi.org/10.1016/j.jvs.2015.01.040

    2015 doi

  67. [2015]

    Journal of Vascular Surgery 61, 1606–1612

    Rupture rates of untreated large abdominal aortic aneurysms in patients unfit for elective repair. Journal of Vascular Surgery 61, 1606–1612. URL: https://www.sciencedirect.com/science/article/ pii/S0741521414019053, doi: https://doi.org/10.1016/j.jvs.2014.10. 023

    doi

  68. [2022]

    Frontiers in Cardiovascular Medicine 8, 769927

    WSSNet: Aortic Wall Shear Stress Estimation Using Deep Learning on 4D Flow MRI. Frontiers in Cardiovascular Medicine 8, 769927. URL: https://www.frontiersin.org/articles/10.3389/fcvm. 2021.769927/full, doi:10.3389/fcvm.2021.769927

    2021

  69. [2190]

    publisher: Springer Science and Business Media LLC

    URL:https://link.springer.com/10.1007/s11517-025-03311-3, doi:10.1007/s11517-025-03311-3. publisher: Springer Science and Business Media LLC

    doi

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.