Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
156 tokens/sec
GPT-4o
7 tokens/sec
Gemini 2.5 Pro Pro
45 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
38 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Point2SSM: Learning Morphological Variations of Anatomies from Point Cloud (2305.14486v2)

Published 23 May 2023 in cs.CV and cs.LG

Abstract: We present Point2SSM, a novel unsupervised learning approach for constructing correspondence-based statistical shape models (SSMs) directly from raw point clouds. SSM is crucial in clinical research, enabling population-level analysis of morphological variation in bones and organs. Traditional methods of SSM construction have limitations, including the requirement of noise-free surface meshes or binary volumes, reliance on assumptions or templates, and prolonged inference times due to simultaneous optimization of the entire cohort. Point2SSM overcomes these barriers by providing a data-driven solution that infers SSMs directly from raw point clouds, reducing inference burdens and increasing applicability as point clouds are more easily acquired. While deep learning on 3D point clouds has seen success in unsupervised representation learning and shape correspondence, its application to anatomical SSM construction is largely unexplored. We conduct a benchmark of state-of-the-art point cloud deep networks on the SSM task, revealing their limited robustness to clinical challenges such as noisy, sparse, or incomplete input and limited training data. Point2SSM addresses these issues through an attention-based module, providing effective correspondence mappings from learned point features. Our results demonstrate that the proposed method significantly outperforms existing networks in terms of accurate surface sampling and correspondence, better capturing population-level statistics.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (70)
  1. Learning representations and generative models for 3d point clouds. In International conference on machine learning, pp.  40–49. PMLR, 2018.
  2. From images to probabilistic anatomical shapes: A deep variational bottleneck approach. In Medical Image Computing and Computer Assisted Intervention–MICCAI 2022: 25th International Conference, Singapore, September 18–22, 2022, Proceedings, Part II, pp.  474–484. Springer, 2022.
  3. Fully bayesian vib-deepssm. In International Conference on Medical Image Computing and Computer-Assisted Intervention, pp.  346–356. Springer, 2023a.
  4. Can point cloud networks learn statistical shape models of anatomies? In International Conference on Medical Image Computing and Computer-Assisted Intervention, pp.  486–496. Springer, 2023b.
  5. Uncertain-deepssm: From images to probabilistic shape models. Shape in Medical Imaging : International Workshop, ShapeMI 2020, Held in Conjunction with MICCAI 2020, Lima, Peru, October 4, 2020, Proceedings, 12474:57–72, 2020.
  6. Computer vision with 3d point cloud data: Methods, datasets and challenges. In 2022 XXVIII International Conference on Information, Communication and Automation Technologies (ICAT), pp.  1–8. IEEE, 2022.
  7. Quantitative comparison of cortical bone thickness using correspondence-based shape modeling in patients with cam femoroacetabular impingement. Journal of Orthopaedic Research, 35(8):1743–1753, 2017.
  8. S3m: scalable statistical shape modeling through unsupervised correspondences. In International Conference on Medical Image Computing and Computer-Assisted Intervention, pp.  459–469. Springer, 2023.
  9. Method for registration of 3-d shapes. In Sensor fusion IV: control paradigms and data structures, volume 1611, pp.  586–606. Spie, 1992.
  10. Deepssm: A deep learning framework for statistical shape modeling from raw images. In Shape In Medical Imaging at MICCAI, volume 11167 of Lecture Notes in Computer Science, pp.  244–257. Springer, 2018.
  11. Quantifying the severity of metopic craniosynostosis: A pilot study application of machine learning in craniofacial surgery. Journal of Craniofacial Surgery, 2020.
  12. Deepssm: A blueprint for image-to-shape deep learning models. arXiv preprint arXiv:2110.07152, 2021.
  13. Combining implicit function learning and parametric models for 3d human reconstruction. In Computer Vision–ECCV 2020: 16th European Conference, Glasgow, UK, August 23–28, 2020, Proceedings, Part II 16, pp.  311–329. Springer, 2020.
  14. Incorporating population-level variability in orthopedic biomechanical analysis: a review. Journal of biomechanical engineering, 136(2):021004, 2014.
  15. A statistical shape modelling framework to extract 3d shape biomarkers from medical imaging data: assessing arch morphology of repaired coarctation of the aorta. BMC medical imaging, 16:1–19, 2016.
  16. Apathy in parkinson’s disease is associated with nucleus accumbens atrophy: a magnetic resonance imaging shape analysis. Movement disorders, 29(7):897–903, 2014.
  17. Shape modeling and analysis with entropy-based particle systems. In IPMI, pp.  333–345. Springer, 2007.
  18. Computational shape models characterize shape change of the left atrium in atrial fibrillation. Clinical Medicine Insights: Cardiology, 8:CMC–S15710, 2014.
  19. Shapeworks: particle-based shape correspondence and visualization software. In Statistical shape and deformation analysis, pp.  257–298. Elsevier, 2017.
  20. Edgenet: Deep metric learning for 3d shapes. Computer Aided Geometric Design, 72:19–33, 2019a.
  21. Unsupervised learning of intrinsic structural representation points. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pp.  9121–9130, 2020.
  22. Bae-net: Branched autoencoder for shape co-segmentation. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pp.  8490–8499, 2019b.
  23. Learning 3d dense correspondence via canonical point autoencoder. Advances in Neural Information Processing Systems, 34:6608–6620, 2021.
  24. Fully convolutional geometric features. In Proceedings of the IEEE/CVF international conference on computer vision, pp.  8958–8966, 2019.
  25. Matching deformable objects in clutter. In 2016 Fourth international conference on 3D vision (3DV), pp.  1–10. IEEE, 2016.
  26. A minimum description length approach to statistical shape modeling. IEEE transactions on medical imaging, 21(5):525–537, 2002.
  27. Learning elementary structures for 3d shape generation and matching. Advances in Neural Information Processing Systems, 32, 2019.
  28. Deep geometric functional maps: Robust feature learning for shape correspondence. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp.  8592–8601, 2020.
  29. Deep shells: Unsupervised shape correspondence with optimal transport. Advances in Neural information processing systems, 33:10491–10502, 2020.
  30. Comprehensive review of deep learning-based 3d point cloud completion processing and analysis. IEEE Transactions on Intelligent Transportation Systems, 2022.
  31. Unsupervised learning of category-specific symmetric 3d keypoints from point sets. In Computer Vision–ECCV 2020: 16th European Conference, Glasgow, UK, August 23–28, 2020, Proceedings, Part XXV 16, pp.  546–563. Springer, 2020.
  32. Cyclic functional mapping: Self-supervised correspondence between non-isometric deformable shapes. In Computer Vision–ECCV 2020: 16th European Conference, Glasgow, UK, August 23–28, 2020, Proceedings, Part V 16, pp.  36–52. Springer, 2020.
  33. The perfect match: 3d point cloud matching with smoothed densities. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pp.  5545–5554, 2019.
  34. Benchmarking off-the-shelf statistical shape modeling tools in clinical applications. Medical Image Analysis, 76:102271, 2022.
  35. Learning local shape descriptors from part correspondences with multiview convolutional networks. ACM Transactions on Graphics (TOG), 37:1–14, 2017.
  36. Mesh2ssm: From surface meshes to statistical shape models of anatomy. In International Conference on Medical Image Computing and Computer-Assisted Intervention, pp.  615–625. Springer, 2023.
  37. Keypointdeformer: Unsupervised 3d keypoint discovery for shape control. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp.  12783–12792, 2021.
  38. David G Kendall. The diffusion of shape. Advances in applied probability, 9(3):428–430, 1977.
  39. Dpc: Unsupervised deep point correspondence via cross and self construction. In 2021 International Conference on 3D Vision (3DV), pp.  1442–1451. IEEE, 2021.
  40. Learning implicit functions for topology-varying dense 3d shape correspondence. Advances in Neural Information Processing Systems, 33:4823–4834, 2020.
  41. Landmark-free statistical shape modeling via neural flow deformations. In Medical Image Computing and Computer Assisted Intervention–MICCAI 2022: 25th International Conference, Singapore, September 18–22, 2022, Proceedings, Part II, pp.  453–463. Springer, 2022.
  42. Correspondence learning via linearly-invariant embedding. Advances in Neural Information Processing Systems, 33:1608–1620, 2020.
  43. Personalized assessment of craniosynostosis via statistical shape modeling. Medical Image Analysis, 18(4):635–646, 2014.
  44. How many different types of femora are there in primary hip osteoarthritis? an active shape modeling study. Journal of Orthopaedic Research, 32(3):413–422, 2014.
  45. Evaluating shape correspondence for statistical shape analysis: A benchmark study. IEEE Transactions on Pattern Analysis and Machine Intelligence, 30(11):2023–2039, 2008.
  46. Shape unicode: A unified shape representation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), June 2019.
  47. Entropy-based particle correspondence for shape populations. International journal of computer assisted radiology and surgery, 11:1221–1232, 2016.
  48. Functional maps: a flexible representation of maps between shapes. ACM Transactions on Graphics (ToG), 31(4):1–11, 2012.
  49. Pointnet: Deep learning on point sets for 3d classification and segmentation. In Proceedings of the IEEE conference on computer vision and pattern recognition, pp.  652–660, 2017a.
  50. Pointnet++: Deep hierarchical feature learning on point sets in a metric space. Advances in neural information processing systems, 30, 2017b.
  51. Deep learning for medical image processing: Overview, challenges and the future. Classification in BioApps: Automation of Decision Making, pp.  323–350, 2018.
  52. Verse: a vertebrae labelling and segmentation benchmark for multi-detector ct images. Medical image analysis, 73:102166, 2021.
  53. A large annotated medical image dataset for the development and evaluation of segmentation algorithms. arXiv preprint arXiv:1902.09063, 2019.
  54. Framework for the statistical shape analysis of brain structures using spharm-pdm. The insight journal, pp.  242, 2006.
  55. Discovery of latent 3d keypoints via end-to-end geometric reasoning. Advances in neural information processing systems, 31, 2018.
  56. Effect of regional analysis methods on assessing the association between wall shear stress and coronary artery disease progression in the clinical setting. In Biomechanics of Coronary Atherosclerotic Plaque, pp.  203–223. Elsevier, 2021.
  57. Probabilistic 3d surface reconstruction from sparse mri information. In Medical Image Computing and Computer Assisted Intervention–MICCAI 2020: 23rd International Conference, Lima, Peru, October 4–8, 2020, Proceedings, Part I 23, pp.  813–823. Springer, 2020.
  58. 3d body scanning and healthcare applications. Computer, 40(7):28–34, 2007. doi: 10.1109/MC.2007.225.
  59. Localization-aware deep learning framework for statistical shape modeling directly from images. In Medical Imaging with Deep Learning, 2023.
  60. Pointattn: You only need attention for point cloud completion. arXiv preprint arXiv:2203.08485, 2022.
  61. Estimating patient-specific and anatomically correct reference model for craniomaxillofacial deformity via sparse representation. Medical physics, 42(10):5809–5816, 2015.
  62. Dynamic graph cnn for learning on point clouds. Acm Transactions On Graphics (tog), 38(5):1–12, 2019.
  63. Unsupervised point cloud representation learning with deep neural networks: A survey. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2023.
  64. Point-bert: Pre-training 3d point cloud transformers with masked point modeling. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp.  19313–19322, 2022.
  65. Pcn: Point completion network. In 2018 international conference on 3D vision (3DV), pp.  728–737. IEEE, 2018.
  66. Stefan Zachow. Computational planning in facial surgery. Facial Plastic Surgery, 31(05):446–462, 2015.
  67. Patient-specific bone modeling and analysis: the role of integration and automation in clinical adoption. Journal of biomechanics, 48(5):750–760, 2015.
  68. Corrnet3d: Unsupervised end-to-end learning of dense correspondence for 3d point clouds. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp.  6052–6061, 2021.
  69. Performance analysis of active shape reconstruction of fractured, incomplete skulls. In Computer Analysis of Images and Patterns: 16th International Conference, CAIP 2015, Valletta, Malta, September 2-4, 2015 Proceedings, Part I 16, pp.  312–324. Springer, 2015.
  70. Point transformer. In Proceedings of the IEEE/CVF international conference on computer vision, pp.  16259–16268, 2021.
Citations (6)
View on Semantic Scholar

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now