Pith. sign in

REVIEW 4 major objections 5 minor 43 references

Rethinking the Detail-Preserved Completion of Complex Tubular Structures based on Point Cloud: a Dataset and a Benchmark

T0 review · 4 major / 5 minor · reviewed 2026-08-15 · deepseek-v4-flash

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

Pith's one-line read This paper argues that reconnecting fractured coronary arteries is best posed as point cloud completion, and it introduces a clinical benchmark and a network, TSRNet, that beat voxel-based repair and prior point-cloud methods on three…

desk verdict Useful new point cloud completion benchmark for coronary vessels, but the ground truth is segmentation output and the coronaries are centerlines, so the 'detail-preserved' claim overreaches. read the letter →

arxiv 2508.17658 v2 pith:7ROVKFXD submitted 2025-08-25 cs.CV

Yaolei Qi , Yikai Yang , Wenbo Peng , Shumei Miao , Yutao Hu , Guanyu Yang This is my paper · ORCID
classification cs.CV
keywords pointcloudcompletiontubularstructurereconnectioncoronaryarteryPC-CACdatasetbenchmarktopologypreservationmedicalimageanalysisdeeplearning
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

This paper argues that fractured tubular structures, especially coronary arteries, can be reconnected more accurately and efficiently when the task is posed as point cloud completion rather than voxel repair. To support that argument, it builds PC-CAC, a point cloud benchmark derived from real clinical coronary CTAs, and introduces TSRNet, a completion network whose components target the thin, imbalanced, topologically complex geometry of vessels. The authors report that TSRNet outperforms prior point cloud completion models and two voxel-based reconnection methods on PC-CAC and on point-cloud conversions of two public datasets. If the benchmark faithfully represents clinical discontinuities, the work gives downstream diagnosis and hemodynamic simulation a reusable dataset and baseline for restoring vessel continuity.

What carries the argument

The load-bearing object is TSRNet plus the dataset-construction pipeline. The pipeline converts clinical coronary segmentations into a point cloud by extracting the aorta surface with marching cubes and farthest point sampling, extracting coronary centre lines, normalizing, and then deleting entire connected point-components at random to create fractured inputs while keeping the intact cloud as ground truth. The network's components are a Core Points Selection module that samples separately in dense and sparse regions with priority for endpoints and isolated points; a TransSA module that alternates set abstraction and transformer self-attention; a two-stage dense refinement that reconstructs a coarse cloud then a fine cloud; and a global-to-local loss that applies Chamfer Distance and Fidelity Error globally, then re-applies both only to sparse regions once global loss falls below a threshold. That loss schedule is what shifts training from overall shape alignment to fine vascular detail.

What would settle it

Take a set of coronary CTAs with expert-annotated complete vessel trees, remove vessel segments at locations of genuine stenosis or occlusion rather than random components, and ask TSRNet to reconnect them; if its completion error on those real clinical fractures is no better than generic point-cloud completion, the PC-CAC synthetic fractures are not representative of the clinical problem.

Watch

Extended reading notes

Core claim

The central claim is that vessel reconnection can be redefined as a point cloud completion problem, and that a dedicated dataset and network make that formulation effective. PC-CAC contains 3,416 point clouds from 427 patients, each with 4,096 points (3,072 on the aorta surface, 1,024 on coronary branch centre lines), with 10 to 30 percent of coronary points removed in up to eight synthetic fracture patterns per patient to simulate stenosis, occlusion, and tortuous-region breakage. TSRNet combines a core-point selection module that balances dense and sparse regions, a transformer set-abstraction feature extractor, two progressive refinement stages, and a global-to-local loss mixing Chamfer Distance and Fidelity Error. On PC-CAC, PC-ImageCAS, and PC-PTR the paper reports the lowest Chamfer Distance and Fidelity Error in every case and the highest F1 on two of the three datasets, with the largest margin on PC-PTR fidelity error (0.102 versus 3.666 for the second-best model). The authors take this as evidence that point-cloud-level completion preserves thin tubular detail better than voxel-level repair or generic point-cloud completion.

Load-bearing premise

The whole benchmark assumes that the authors' segmentation of clinical CTAs is an accurate complete ground truth, and that randomly deleting connected chunks of that segmentation mimics the discontinuities a clinician would see in real stenosis or occlusion.

Editorial extensions

If this is right

  • Point-cloud representation cuts the reconnection problem from over a million voxels to about four thousand points, making completion fast enough to serve as a preprocessing step in clinical pipelines.
  • The reconnected point cloud can be converted directly into a sub-voxel mesh, so downstream hemodynamic simulation and anatomical reconstruction do not need to return to voxel space.
  • Because TSRNet does not rely on explicit fracture endpoint detection, it avoids a failure mode of random-walk reconnection methods that require local endpoint matching.
  • The method remains stable as fracture severity rises from 10% to 30%, and at bifurcations it reconstructs all intersecting branches, suggesting the model is not merely filling short gaps.
  • Evaluating the same network on coronary and pulmonary datasets indicates the approach transfers across different tubular anatomies, not just one vessel tree.

Reading between the lines

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

  • The synthetic fractures are random deletions of connected point components in segmentation-derived clouds; real occlusions also corrupt image intensity, so a test on CTA regions with true stenosis may show different failure modes than the benchmark captures.
  • Because the ground truth is the authors' own segmentation output, part of the reported gap over other methods may reflect how closely TSRNet reproduces that particular segmentation's topology; an independent expert-annotated completeness standard would separate reconnection skill from dataset bias.
  • The unusually large fidelity-error margin on PC-PTR suggests the method's density-aware sampling and sparse-region loss are especially valuable when thin structures dominate; this predicts strong transfer to retinal vessels, airways, or other slender tubular networks.
  • The authors state plans to pair point clouds with original image data; that pairing could be used to resolve ambiguous long-gap reconnections by consulting image appearance, a direction their current geometry-only model cannot test.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

4 major / 5 minor

Summary. The paper introduces PC-CAC, a point cloud-based coronary artery completion dataset derived from real CT data, and proposes TSRNet, a network for reconnecting fractured tubular structures. The dataset and TSRNet are evaluated on PC-CAC and two converted public datasets (PC-ImageCAS, PC-PTR) against generic point cloud completion baselines and two voxel-based methods. The authors report that TSRNet achieves the best or second-best results on most metrics across the three datasets, and they provide ablations of the main components and hyperparameters. The dataset and code are publicly released.

Significance. If the benchmark were independently validated, this would be a useful contribution: a large-scale (427 patients, 3,416 cases) public benchmark for a clinically motivated reconnection task, with a strong baseline and broad comparisons against seven point cloud completion methods and two voxel-based methods. The paper also ships a public codebase and dataset link, which supports reproducibility. However, the significance is substantially tempered by the fact that the ground truth is the authors' own automatic segmentation output without expert correction or independent anatomical validation, and by internal inconsistencies in the reported loss threshold and ablation setup. The central clinical claim—that the benchmark and method capture 'detail-preserved' completion of true coronary anatomy—is not yet supported.

major comments (4)
  1. [Sec. 4.1 and Algorithm 1] The PC-CAC ground truth is generated from automatic segmentation results (nnU-Net and the authors' own DSN, Qi et al. 2023) and then skeletonized to a centerline for the coronary arteries. No expert correction or independent anatomical verification is reported. The completion task is therefore defined as reconstructing the segmentation pipeline's output, not the patient's true coronary anatomy. This is load-bearing for the clinical framing in the Introduction and for the 'detail-preserved' claim, because the representation discards vessel caliber and the ground truth embeds any systematic bias of the segmentation. I recommend that the authors either obtain expert-verified ground truth for at least a test subset, or explicitly reframe the benchmark as 'segmentation-output reconnection' and temper claims about clinical validity.
  2. [Sec. 4.3 vs. Table 4] The implementation details state that the loss threshold epsilon is set to 6 × 10^-3, but Table 4 reports an ablation over epsilon with values 3.1, 2.9, and 2.7, and row 7 (epsilon = 2.7) is declared best. If the Table 4 values are implicitly scaled by 10^-3, then the default (6 × 10^-3) differs from the reported best (2.7 × 10^-3). If they are not scaled, the units are inconsistent. The paper does not clarify which epsilon was actually used for the main results in Table 2. This ambiguity directly affects the reproducibility of the loss function and the interpretation of the ablation study, so it must be resolved.
  3. [Sec. 5.1.1 and Table 4] The text states that the hyperparameter ablation is conducted 'on the PC-CAC dataset', but the best row (Row 7) reports CD=2.395, F1=96.83, Fidelity=2.062, which exactly match the TSRNet results for PC-ImageCAS in Table 2, not the PC-CAC results in the same table. This indicates a mismatch between the table's data and its stated evaluation dataset. As written, the ablation conclusions are not reproducible and the experimental setup is unclear. The authors should correct either the text or the table and re-run the ablation on the appropriate dataset.
  4. [Sec. 5.1.1 and Table 2] The paper's abstract claims TSRNet 'consistently outperforms state-of-the-art approaches across multiple evaluation metrics', and Sec. 5.1.1 states it achieves 'state-of-the-art performance in CD, F1, and Fidelity Error' on all three datasets. This is contradicted by Table 2 on PC-ImageCAS, where SeedFormer achieves F1=97.47 and TSRNet achieves 96.83; the text later acknowledges 'second-best F1'. The authors should either weaken the headline claim to reflect metric-wise and dataset-wise exceptions, or provide a clear statistical argument for why the method is still considered consistently superior (e.g., paired significance tests).
minor comments (5)
  1. [Sec. 5.2] The text refers to 'Table 5.2.2' while the actual table is labeled 'Table 5'; the cross-reference should be corrected.
  2. [Sec. 5.2.2] Typo: 'where fice representative cases' should be 'where five representative cases'.
  3. [Sec. 4.3] The evaluation metric 'Fidelity Error' is also used as a loss term (Sec. 3.3). This coupling should be explicitly acknowledged, since optimizing the loss directly minimizes the evaluation metric and may inflate its reported advantage.
  4. [Fig. 4 and Fig. 5] The qualitative figures show red-marked error regions, but the criteria for marking are not defined. A brief description of how error regions were determined would improve interpretability.
  5. [Sec. 4.1] The PC-ImageCAS and PC-PTR conversion details are summarized only in Table 1. It would be helpful to state explicitly whether the same fracture-simulation algorithm (Algorithm 1) is applied to these public datasets, and with which min/max break ratios beyond the point-removal percentages mentioned.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: TSRNet's superiority is tested against external baselines on held-out splits; PC-CAC's self-referential provenance is a validity caveat, not a derivationally circular step.

full rationale

The paper's central comparative claim is not circular: TSRNet is a supervised point-cloud completion network trained with Chamfer Distance and Fidelity Error losses on synthetically fractured inputs, and its reported superiority is evaluated on held-out test splits of PC-CAC, PC-ImageCAS, and PC-PTR against external baselines (GRNet, PoinTr, SeedFormer, SnowflakeNet, AnchorFormer, PointAttN, CRA-PCN, DRTT, VSR-Net) trained under the same protocol (Sec. 4.3). No fitted parameter is renamed as a prediction, and no equation defining TSRNet's output is identical by construction to the evaluation metric in a way that forces the ranking. The only self-referential element is the provenance of PC-CAC: its 'complete' ground truth is a centerline skeleton extracted from automatic segmentation results, including the authors' own Qi et al. (2023) method (Sec. 4.1-4.2, Alg. 1), and the fractured inputs are random connected-component deletions of that same skeleton. This raises a validity concern about whether the benchmark reflects independently verified anatomy, but it does not make the method's derivation circular, because the input and target are not equal by construction and the comparative results on two public datasets are external. Under the requested standard of exhibited reduction rather than vague concern, no circular step is established.

Assumptions & free parameters 5 free parameters · 4 assumptions · 0 invented entities

The central comparison rests on the dataset's ground truth, which is generated from the authors' own segmentation method, and on synthetic fracture simulation. These are domain assumptions rather than mathematical axioms; no new physical entities are introduced.

free parameters (5)
  • Loss threshold epsilon = 2.7 (reported as 6e-3 in Sec. 4.3)
    Controls the switch between global and local loss in Eq. (4); selected on the validation set in Table 4, and the best value differs from the stated configuration in Sec. 4.3.
  • Loss trade-off gamma = 0.5
    Weights the local loss in Eq. (4); chosen via hyperparameter ablation on PC-CAC.
  • Number of SA+Transformer blocks = 2
    Depth of the feature extractor; ablated in Table 4 and set to 2 for the final model.
  • Core point ratio N0 = 1/4 of complete point cloud
    Hand-set in Sec. 3.1.1; determines the sparsity of the core point cloud and affects refinement quality.
  • Fracture simulation parameters = 10% to 30% removal, random break count
    Hand-chosen in Algorithm 1 and Sec. 4.1 to simulate clinical discontinuity; directly controls task difficulty for all methods.
assumptions (4)
  • domain assumption Segmentation output is a valid proxy for true tubular anatomy.
    PC-CAC ground truth is generated from segmentation results (Isensee et al., 2021; Qi et al., 2023) without independent radiological validation; the completion target is the segmentation's main trunk, not verified anatomy.
  • domain assumption Synthetic random point deletion produces realistic clinical discontinuities.
    Fractures are simulated by removing connected components of points (10%-30%) rather than from real occlusions or stenoses, so the distribution of gaps may differ from genuine clinical cases.
  • domain assumption Point cloud representation preserves clinically relevant topology.
    Aorta is represented as a surface mesh while coronary arteries are centerlines; the paper assumes this mixed representation is sufficient for reconnection and downstream simulation without validation against full volumetric anatomy.
  • domain assumption Chamfer Distance, F1, and Fidelity Error adequately capture reconnection quality.
    The metrics are computed per case and averaged, but the F1 threshold is not specified, and topological correctness is not measured directly, so metric improvements may not correspond to clinically correct reconnection.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Rethinking the Detail-Preserved Completion of Complex Tubular Structures based on Point Cloud: a Dataset and a Benchmark." pith.science (2026). https://pith.science/paper/7ROVKFXD

@misc{pith2026250817658,
  author       = {Pith},
  title        = {Pith review of: Rethinking the Detail-Preserved Completion of Complex Tubular Structures based on Point Cloud: a Dataset and a Benchmark},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/7ROVKFXD}},
  note         = {Machine review of arXiv:2508.17658}
}
read the original abstract

Complex tubular structures are essential in medical imaging and computer-assisted diagnosis, where their integrity enhances anatomical visualization and lesion detection. However, existing segmentation algorithms struggle with structural discontinuities, particularly in severe clinical cases such as coronary artery stenosis and vessel occlusions, which leads to undesired discontinuity and compromising downstream diagnostic accuracy. Therefore, it is imperative to reconnect discontinuous structures to ensure their completeness. In this study, we explore the tubular structure completion based on point cloud for the first time and establish a Point Cloud-based Coronary Artery Completion (PC-CAC) dataset, which is derived from real clinical data. This dataset provides a novel benchmark for tubular structure completion. Additionally, we propose TSRNet, a Tubular Structure Reconnection Network that integrates a detail-preservated feature extractor, a multiple dense refinement strategy, and a global-to-local loss function to ensure accurate reconnection while maintaining structural integrity. Comprehensive experiments on our PC-CAC and two additional public datasets (PC-ImageCAS and PC-PTR) demonstrate that our method consistently outperforms state-of-the-art approaches across multiple evaluation metrics, setting a new benchmark for point cloud-based tubular structure reconstruction. Our benchmark is available at https://github.com/YaoleiQi/PCCAC.

Figures

Figures reproduced from arXiv: 2508.17658 by the authors.

Figure 1
Figure 1. Motivation. (a) shows the undesired discontinuity caused by unclear image representation in structures affected by lesions. (b) shows vessels visualized across different modalities, where thin tubular structures are prone to being obscured, making them difficult to observe without the use of contrast agents. (c) shows the differences in structural reconnection from the perspectives of both voxel-based and point clou… view at source ↗
Figure 2
Figure 2. (a) Illustration of myocardial bridging in a real clinical case, where the lesion leads to local structural loss that challenges voxel-based models. (b) Our proposed method shows superior reconnection performance, and better handles such fracture compared to existing voxel-based approaches. damage-prone regions within the main structures are removed from the input to better reflect the complexity of structural disco… view at source ↗
Figure 3
Figure 3. Framework. Our framework comprises two main stages: a Detail-preserved Feature Extractor and a Multiple Dense Refinement strategy. These are jointly optimized using a Global-to-local Loss Function to ensure the structural consistency. (a) Core Points Selection analyzes the density to extract critical points that are easily overlooked. (b) TransSA Feature Extraction utilizes a Transformer-based set abstraction to cap… view at source ↗
Figures from the paper (5 more)
Figure 4
Figure 4. Figure 4: Visual comparison of different point cloud-based models. A-1, A-2 are two cases from our PC-CAC dataset. B-1, B-2 are two cases from PC-ImageCAS dataset. C-1, C-2 are two cases from PC-PTR dataset. overall performance with 𝜖 = 2.7 and 𝛾 = 0.5. We also explored the numb…
Figure 5
Figure 5. Figure 5: Challenging cases with highly tortuous and complex anatomies. A-3,4 are two cases from our PC-CAC dataset. unintended fractures in originally intact regions. While Pointr performs relatively well in structural restoration, it fails to focus on elongated tubular structu…
Figure 6
Figure 6. Figure 6: Visual comparison of different voxel-based models. B-1 to B-5 are 5 cases from the public PC-ImageCAS dataset. We focus on vascular regions that are particularly difficult to reconnect. In particular, cases B-2 and B-4 involve long-gap structural discontinuities that p…
Figure 7
Figure 7. Figure 7: Performance comparison under varying fracture ratios (10%, 20%, and 30%) in our PC-CAC dataset. illustrates a specific example simulating three different levels of fractures. At 10% fracture severity, a single fracture occurs in a complex and elongated region. At 20% f…
Figure 8
Figure 8. Figure 8: Visual comparison of reconnection results across different fracture severities and structural types. We compare the performance of SeedFormer (the second best) and our method under three grades of fracture severity: slight (Grade 1), moderate (Grade 2), and severe (Gra…

Review history (2 revisions) →

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

43 extracted references · 40 canonical work pages

  1. [1]

    write newline

    " write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 global.max substring 't := if while FUNCTION word.in bbl.in ":" * " " * FUNCTION f...

  2. [2]

    , author Banerjee, A

    author Beetz, M. , author Banerjee, A. , author Ossenberg-Engels, J. , author Grau, V. , year 2023 . title Multi-class point cloud completion networks for 3d cardiac anatomy reconstruction from cine magnetic resonance images . journal Medical image analysis volume 90 , pages 102975

    work page 2023

  3. [3]

    , author Salamanca, L

    author Bernard, F. , author Salamanca, L. , author Thunberg, J. , author Tack, A. , author Jentsch, D. , author Lamecker, H. , author Zachow, S. , author Hertel, F. , author Goncalves, J. , author Gemmar, P. , year 2017 . title Shape-aware surface reconstruction from sparse 3d point-clouds . journal Medical image analysis volume 38 , pages 77--89

    work page 2017

  4. [4]

    , author Ravikumar, N

    author Chen, X. , author Ravikumar, N. , author Xia, Y. , author Attar, R. , author Diaz-Pinto, A. , author Piechnik, S.K. , author Neubauer, S. , author Petersen, S.E. , author Frangi, A.F. , year 2021 . title Shape registration with learned deformations for 3d shape reconstruction from sparse and incomplete point clouds . journal Medical image analysis ...

    work page 2021

  5. [5]

    , author Long, F

    author Chen, Z. , author Long, F. , author Qiu, Z. , author Yao, T. , author Zhou, W. , author Luo, J. , author Mei, T. , year 2023 . title Anchorformer: Point cloud completion from discriminative nodes , in: booktitle Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , pp. pages 13581--13590

    work page 2023

  6. [6]

    , author Su, H

    author Fan, H. , author Su, H. , author Guibas, L.J. , year 2017 . title A point set generation network for 3d object reconstruction from a single image , in: booktitle Proceedings of the IEEE conference on computer vision and pattern recognition , pp. pages 605--613

    work page 2017

  7. [7]

    , author Yang, W

    author Fei, B. , author Yang, W. , author Chen, W.M. , author Li, Z. , author Li, Y. , author Ma, T. , author Hu, X. , author Ma, L. , year 2022 . title Comprehensive review of deep learning-based 3d point cloud completion processing and analysis . journal IEEE Transactions on Intelligent Transportation Systems volume 23 , pages 22862--22883

    work page 2022

  8. [8]

    , author Adhav, Y

    author Garje, A. , author Adhav, Y. , author Bodas, D. , year 2015 . title Design and simulation of blocked blood vessel for early detection of heart diseases , in: booktitle 2015 2nd International Symposium on Physics and Technology of Sensors (ISPTS) , organization IEEE . pp. pages 204--208

    work page 2015

Show all 43 references
  1. [9]

    , author Chen, N

    author Gharleghi, R. , author Chen, N. , author Sowmya, A. , author Beier, S. , year 2022 . title Towards automated coronary artery segmentation: A systematic review . journal Computer Methods and Programs in Biomedicine volume 225 , pages 107015

    2022

  2. [10]

    , year 2006

    author Grady, L. , year 2006 . title Random walks for image segmentation . journal IEEE transactions on pattern analysis and machine intelligence volume 28 , pages 1768--1783

    2006

  3. [11]

    , author Zhang, Y

    author Gupta, S. , author Zhang, Y. , author Hu, X. , author Prasanna, P. , author Chen, C. , year 2023 . title Topology-aware uncertainty for image segmentation . journal Advances in Neural Information Processing Systems volume 36 , pages 8186--8207

    2023

  4. [12]

    , author Jeon, J

    author Han, K. , author Jeon, J. , author Jang, Y. , author Jung, S. , author Kim, S. , author Shim, H. , author Jeon, B. , author Chang, H.J. , year 2022 . title Reconnection of fragmented parts of coronary arteries using local geometric features in x-ray angiography images ....

    2022

  5. [13]

    , author Jaeger, P.F

    author Isensee, F. , author Jaeger, P.F. , author Kohl, S.A. , author Petersen, J. , author Maier-Hein, K.H. , year 2021 . title nnu-net: a self-configuring method for deep learning-based biomedical image segmentation . journal Nature methods volume 18 , pages 203--211

    2021

  6. [14]

    , author Cameron, J.D

    author Ko, B.S. , author Cameron, J.D. , author Munnur, R.K. , author Wong, D.T. , author Fujisawa, Y. , author Sakaguchi, T. , author Hirohata, K. , author Hislop-Jambrich, J. , author Fujimoto, S. , author Takamura, K. , author Crossett, M. , author Leung, M. , author Kugane...

    2017

  7. [15]

    , author Wang, X

    author Kong, B. , author Wang, X. , author Bai, J. , author Lu, Y. , author Gao, F. , author Cao, K. , author Xia, J. , author Song, Q. , author Yin, Y. , year 2020 . title Learning tree-structured representation for 3d coronary artery segmentation . journal Computerized Medic...

    2020

  8. [16]

    , author Chen, C.H

    author Lee, M.S. , author Chen, C.H. , year 2015 . title Myocardial bridging: an up-to-date review . journal The Journal of invasive cardiology volume 27 , pages 521

    2015

  9. [17]

    , author Tang, Z

    author Li, H. , author Tang, Z. , author Nan, Y. , author Yang, G. , year 2022 . title Human treelike tubular structure segmentation: A comprehensive review and future perspectives . journal Computers in Biology and Medicine volume 151 , pages 106241

    2022

  10. [18]

    , author Gong, H

    author Li, Y. , author Gong, H. , author Wu, W. , author Liu, G. , author Chen, G. , year 2015 . title An automated method using hessian matrix and random walks for retinal blood vessel segmentation , in: booktitle 2015 8th International Congress on Image and Signal Processing...

    2015

  11. [19]

    , author Xia, Q

    author Li, Z. , author Xia, Q. , author Hu, Z. , author Wang, W. , author Xu, L. , author Zhang, S. , year 2021 . title A deep reinforced tree-traversal agent for coronary artery centerline extraction , in: booktitle International Conference on Medical Image Computing and Comp...

    2021

  12. [20]

    , author Tarigopula, M

    author Madhavan, M.V. , author Tarigopula, M. , author Mintz, G.S. , author Maehara, A. , author Stone, G.W. , author G \'e n \'e reux, P. , year 2014 . title Coronary artery calcification: pathogenesis and prognostic implications . journal Journal of the American College of C...

    2014

  13. [21]

    , author Chen, L

    author Mou, L. , author Chen, L. , author Cheng, J. , author Gu, Z. , author Zhao, Y. , author Liu, J. , year 2019 . title Dense dilated network with probability regularized walk for vessel detection . journal IEEE transactions on medical imaging volume 39 , pages 1392--1403

    2019

  14. [22]

    , author Sels, J.W.E

    author Pijls, N.H. , author Sels, J.W.E. , year 2012 . title Functional measurement of coronary stenosis . journal Journal of the American College of Cardiology volume 59 , pages 1045--1057

    2012

  15. [23]

    , author Su, H

    author Qi, C.R. , author Su, H. , author Mo, K. , author Guibas, L.J. , year 2017 a. title Pointnet: Deep learning on point sets for 3d classification and segmentation , in: booktitle Proceedings of the IEEE conference on computer vision and pattern recognition , pp. pages 652--660

    2017

  16. [24]

    , author Yi, L

    author Qi, C.R. , author Yi, L. , author Su, H. , author Guibas, L.J. , year 2017 b. title Pointnet++: Deep hierarchical feature learning on point sets in a metric space . journal Advances in neural information processing systems volume 30 , pages 5105--5114

    2017

  17. [25]

    , author He, Y

    author Qi, Y. , author He, Y. , author Qi, X. , author Zhang, Y. , author Yang, G. , year 2023 . title Dynamic snake convolution based on topological geometric constraints for tubular structure segmentation , in: booktitle Proceedings of the IEEE/CVF international conference o...

    2023

  18. [26]

    , author Li, Z

    author Qiu, Y. , author Li, Z. , author Wang, Y. , author Dong, P. , author Wu, D. , author Yang, X. , author Hong, Q. , author Shen, D. , year 2023 . title Corsegrec: a topology-preserving scheme for extracting fully-connected coronary arteries from ct angiography , in: bookt...

    2023

  19. [27]

    , author Zhou, H

    author Rong, Y. , author Zhou, H. , author Yuan, L. , author Mei, C. , author Wang, J. , author Lu, T. , year 2024 . title Cra-pcn: Point cloud completion with intra-and inter-level cross-resolution transformers , in: booktitle Proceedings of the AAAI Conference on Artificial ...

    2024

  20. [28]

    , author Richter, S.R

    author Tatarchenko, M. , author Richter, S.R. , author Ranftl, R. , author Li, Z. , author Koltun, V. , author Brox, T. , year 2019 . title What do single-view 3d reconstruction networks learn? , in: booktitle Proceedings of the IEEE/CVF conference on computer vision and patte...

    2019

  21. [29]

    , author Robinson, T

    author Thanvi, B. , author Robinson, T. , year 2007 . title Complete occlusion of extracranial internal carotid artery: clinical features, pathophysiology, diagnosis and management . journal Postgraduate medical journal volume 83 , pages 95--99

    2007

  22. [30]

    , author Cui, Y

    author Wang, J. , author Cui, Y. , author Guo, D. , author Li, J. , author Liu, Q. , author Shen, C. , year 2024 . title Pointattn: You only need attention for point cloud completion , in: booktitle Proceedings of the AAAI Conference on artificial intelligence , pp. pages 5472--5480

    2024

  23. [31]

    , author Xiang, P

    author Wen, X. , author Xiang, P. , author Han, Z. , author Cao, Y.P. , author Wan, P. , author Zheng, W. , author Liu, Y.S. , year 2022 . title Pmp-net++: Point cloud completion by transformer-enhanced multi-step point moving paths . journal IEEE Transactions on Pattern Analy...

    2022

  24. [32]

    , author Yang, J

    author Weng, Z. , author Yang, J. , author Liu, D. , author Cai, W. , year 2023 . title Topology repairing of disconnected pulmonary airways and vessels: baselines and a dataset , in: booktitle International Conference on Medical Image Computing and Computer-Assisted Intervent...

    2023

  25. [33]

    , author Wen, X

    author Xiang, P. , author Wen, X. , author Liu, Y.S. , author Cao, Y.P. , author Wan, P. , author Zheng, W. , author Han, Z. , year 2022 . title Snowflake point deconvolution for point cloud completion and generation with skip-transformer . journal IEEE Transactions on Pattern...

    2022

  26. [34]

    , author Yao, H

    author Xie, H. , author Yao, H. , author Zhou, S. , author Mao, J. , author Zhang, S. , author Sun, W. , year 2020 . title Grnet: Gridding residual network for dense point cloud completion , in: booktitle European conference on computer vision , organization Springer . pp. pag...

    2020

  27. [35]

    , author Zhang, X

    author Ye, H. , author Zhang, X. , author Hu, Y. , author Fu, H. , author Liu, J. , year 2025 . title Vsr-net: Vessel-like structure rehabilitation network with graph clustering . journal IEEE Transactions on Image Processing , pages 1090--1105

    2025

  28. [36]

    , author Rao, Y

    author Yu, X. , author Rao, Y. , author Wang, Z. , author Liu, Z. , author Lu, J. , author Zhou, J. , year 2021 . title Pointr: Diverse point cloud completion with geometry-aware transformers , in: booktitle Proceedings of the IEEE/CVF international conference on computer visi...

    2021

  29. [37]

    , author Khot, T

    author Yuan, W. , author Khot, T. , author Held, D. , author Mertz, C. , author Hebert, M. , year 2018 . title Pcn: Point completion network , in: booktitle 2018 international conference on 3D vision (3DV) , organization IEEE . pp. pages 728--737

    2018

  30. [38]

    , author Wu, C

    author Zeng, A. , author Wu, C. , author Lin, G. , author Xie, W. , author Hong, J. , author Huang, M. , author Zhuang, J. , author Bi, S. , author Pan, D. , author Ullah, N. , author Khan, K.N. , author Wang, T. , author Shi, Y. , author Li, X. , author Xu, X. , year 2023 . t...

    2023

  31. [39]

    , author Li, X

    author Zhang, S. , author Li, X. , author Zong, M. , author Zhu, X. , author Cheng, D. , year 2017 . title Learning k for knn classification . journal ACM Transactions on Intelligent Systems and Technology (TIST) volume 8 , pages 1--19

    2017

  32. [40]

    , author Sun, K

    author Zhang, X. , author Sun, K. , author Wu, D. , author Xiong, X. , author Liu, J. , author Yao, L. , author Li, S. , author Wang, Y. , author Feng, J. , author Shen, D. , year 2023 . title An anatomy-and topology-preserving framework for coronary artery segmentation . jour...

    2023

  33. [41]

    , author Zhang, J

    author Zhang, X. , author Zhang, J. , author Ma, L. , author Xue, P. , author Hu, Y. , author Wu, D. , author Zhan, Y. , author Feng, J. , author Shen, D. , year 2022 . title Progressive deep segmentation of coronary artery via hierarchical topology learning , in: booktitle In...

    2022

  34. [42]

    , author Jiang, L

    author Zhao, H. , author Jiang, L. , author Jia, J. , author Torr, P.H. , author Koltun, V. , year 2021 . title Point transformer , in: booktitle Proceedings of the IEEE/CVF international conference on computer vision , pp. pages 16259--16268

    2021

  35. [43]

    , author Cao, Y

    author Zhou, H. , author Cao, Y. , author Chu, W. , author Zhu, J. , author Lu, T. , author Tai, Y. , author Wang, C. , year 2022 . title Seedformer: Patch seeds based point cloud completion with upsample transformer , in: booktitle European conference on computer vision , org...

    2022

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.