Topologically Faithful Multi-class Segmentation in Medical Images
Authors: 
Alexander H. Berger, Laurin Lux, Nico Stucki, Vincent Bürgin, Suprosanna Shit, Anna Banaszak, Daniel Rueckert, Ulrich Bauer, Johannes C. PaetzoldAuthors Info & Claims
Alexander H. Berger, Laurin Lux, Nico Stucki, Vincent Bürgin, Suprosanna Shit, Anna Banaszak, Daniel Rueckert, Ulrich Bauer, Johannes C. PaetzoldAuthors Info & ClaimsPages 721 - 731
Abstract
Topological accuracy in medical image segmentation is a highly important property for downstream applications such as network analysis and flow modeling in vessels or cell counting. Recently, significant methodological advancements have brought well-founded concepts from algebraic topology to binary segmentation. However, these approaches have been underexplored in multi-class segmentation scenarios, where topological errors are common. We propose a general loss function for topologically faithful multi-class segmentation extending the recent Betti matching concept, which is based on induced matchings of persistence barcodes. We project the N-class segmentation problem to N single-class segmentation tasks, which allows us to use 1-parameter persistent homology, making training of neural networks computationally feasible. We validate our method on a comprehensive set of four medical datasets with highly variant topological characteristics. Our loss formulation significantly enhances topological correctness in cardiac, cell, artery-vein, and Circle of Willis segmentation.
References
[1]
BenTaieb A and Hamarneh G Ourselin S, Joskowicz L, Sabuncu MR, Unal G, and Wells W Topology aware fully convolutional networks for histology gland segmentation Medical Image Computing and Computer-Assisted Intervention – MICCAI 2016 2016 Cham Springer 460-468
[2]
Bernard O et al. Deep learning techniques for automatic MRI cardiac multi-structures segmentation and diagnosis: is the problem solved? IEEE Trans. Med. Imaging 2018 37 11 2514-2525
[3]
Byrne N, Clough JR, Valverde I, Montana G, and King AP A persistent homology-based topological loss for CNN-based multiclass segmentation of CMR IEEE Trans. Med. Imaging 2022 42 1 3-14
[4]
Clough, J., Byrne, N., Oksuz, I., Zimmer, V.A., Schnabel, J.A., King, A.: A topological loss function for deep-learning based image segmentation using persistent homology. IEEE Trans. Pattern Anal. Mach. Intell. 44(12), 8766–8778 (2020)
[5]
Clough JR, Oksuz I, Byrne N, Schnabel JA, and King AP Chung ACS, Gee JC, Yushkevich PA, and Bao S Explicit topological priors for deep-learning based image segmentation using persistent homology Information Processing in Medical Imaging: 26th International Conference, IPMI 2019, Hong Kong, China, June 2–7, 2019, Proceedings 2019 Cham Springer International Publishing 16-28
[6]
Guay MD et al. Dense cellular segmentation for EM using 2D–3D neural network ensembles Sci. Rep. 2021 11 1 2561
[7]
Gupta S et al. Avidan S, Brostow G, Cissé M, Farinella GM, Hassner T, et al. Learning topological interactions for multi-class medical image segmentation Computer Vision – ECCV 2022: 17th European Conference, Tel Aviv, Israel, October 23–27, 2022, Proceedings, Part XXIX 2022 Cham Springer Nature Switzerland 701-718
[8]
Hu, X., Wang, Y., Li, F., Samaras, D., Chen, C.: Topology-aware segmentation using discrete Morse theory. In: International Conference on Learning Representations (ICLRR) (2021)
[9]
Hu, X., Chen, C.: Image segmentation with homotopy warping. arXiv preprint arXiv:2112.07812 (2021)
[10]
Hu, X., Li, F., Samaras, D., Chen, C.: Topology-preserving deep image segmentation. Adv. Neural Inf. Process. Syst. 32 (2019)
[11]
Jain, V., et al.: Boundary learning by optimization with topological constraints. In: 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, pp. 2488–2495. IEEE (2010)
[12]
Kingma, D.P., Ba, J.: Adam: a method for stochastic optimization. arXiv preprint arXiv:1412.6980 (2014)
[13]
Kreitner, L., et al.: Synthetic optical coherence tomography angiographs for detailed retinal vessel segmentation without human annotations. IEEE Trans. Med. Imag. 43(6), 2061–2073 (2024)
[14]
Li L, et al., et al. Greenspan H, et al., et al. Robust segmentation via topology violation detection and feature synthesis Medical Image Computing and Computer Assisted Intervention – MICCAI 2023: 26th International Conference, Vancouver, BC, Canada, October 8–12, 2023, Proceedings, Part IV 2023 Cham Springer Nature Switzerland 67-77
[15]
Li M et al. Octa-500: a retinal dataset for optical coherence tomography angiography study Med. Image Anal. 2024 93 103092
[16]
Lin Z, Wei D, Gupta A, Liu X, Sun D, Pfister H, et al. Greenspan H et al. Structure-preserving instance segmentation via skeleton-aware distance transform Medical Image Computing and Computer Assisted Intervention – MICCAI 2023: 26th International Conference, Vancouver, BC, Canada, October 8–12, 2023, Proceedings, Part III 2023 Cham Springer Nature Switzerland 529-539
[17]
Ma J et al. Loss odyssey in medical image segmentation Med. Image Anal. 2021 71 102035
[18]
Menten MJ et al. Synthetic data facilitates deep-learning-based segmentation of oct angiography images without human annotations Invest. Ophthalmol. Visual Sci. 2023 64 8 5450-5450
[19]
Mosinska, A., et al.: Beyond the pixel-wise loss for topology-aware delineation. In: CVPR, pp. 3136–3145 (2018)
[20]
Qiu Y et al. Greenspan H, Madabhushi A, Mousavi P, Salcudean S, Duncan J, Syeda-Mahmood T, Taylor R, et al. CorSegRec: a topology-preserving scheme for extracting fully-connected coronary arteries from CT angiography Medical Image Computing and Computer Assisted Intervention – MICCAI 2023: 26th International Conference, Vancouver, BC, Canada, October 8–12, 2023, Proceedings, Part III 2023 Cham Springer Nature Switzerland 670-680
[21]
Ronneberger O, Fischer P, and Brox T Navab N, Hornegger J, Wells WM, and Frangi AF U-Net: convolutional networks for biomedical image segmentation Medical Image Computing and Computer-Assisted Intervention – MICCAI 2015: 18th International Conference, Munich, Germany, October 5-9, 2015, Proceedings, Part III 2015 Cham Springer International Publishing 234-241
[22]
Shit, S., et al.: clDice-a novel topology-preserving loss function for tubular structure segmentation. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 16560–16569 (2021)
[23]
Stucki, N., Bürgin, V., Paetzold, J.C., Bauer, U.: Efficient betti matching enables topology-aware 3D segmentation via persistent homology (2024). https://arxiv.org/abs/2407.04683
[24]
Stucki, N., Paetzold, J.C., Shit, S., Menze, B., Bauer, U.: Topologically faithful image segmentation via induced matching of persistence barcodes. In: International Conference on Machine Learning, pp. 32698–32727. PMLR (2023)
[25]
Sudre CH, Li W, Vercauteren T, Ourselin S, Jorge Cardoso M, et al. Cardoso MJ et al. Generalised dice overlap as a deep learning loss function for highly unbalanced segmentations Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support 2017 Cham Springer International Publishing 240-248
[26]
Wang, X., Jiang, X.: Post-processing for retinal vessel detection. In: Tenth International Conference on Digital Image Processing (ICDIP 2018). vol. 10806, pp. 1442–1446. SPIE (2018)
[27]
Yang, K., et al.: Benchmarking the cow with the topcow challenge: topology-aware anatomical segmentation of the circle of willis for CTA and MRA. arXiv preprint arXiv:2312.17670 (2023)
[28]
Zhang X et al. Wang L, Dou Q, Fletcher PT, Speidel S, Li S, et al. Progressive deep segmentation of coronary artery via hierarchical topology learning Medical Image Computing and Computer Assisted Intervention – MICCAI 2022: 25th International Conference, Singapore, September 18–22, 2022, Proceedings, Part V 2022 Cham Springer Nature Switzerland 391-400
[29]
Zhang Z, et al., et al. Greenspan H, et al., et al. Topology-preserving automatic labeling of coronary arteries via anatomy-aware connection classifier Medical Image Computing and Computer Assisted Intervention – MICCAI 2023: 26th International Conference, Vancouver, BC, Canada, October 8–12, 2023, Proceedings, Part VII 2023 Cham Springer Nature Switzerland 759-769
Index Terms
- Topologically Faithful Multi-class Segmentation in Medical Images
Index terms have been assigned to the content through auto-classification.
Recommendations
Modeling Multi-annotator Uncertainty as Multi-class Segmentation Problem
Brainlesion: Glioma, Multiple Sclerosis, Stroke and Traumatic Brain InjuriesAbstractMedical image segmentation is a monotonous, time-consuming, and costly task performed by highly skilled medical annotators. Despite adequate training, the intra- and inter-annotator variations results in significantly differing segmentations. If ...
General Tendencies in Segmentation of Medical Ultrasound Images
ICICSE '09: Proceedings of the 2009 Fourth International Conference on Internet Computing for Science and EngineeringImage segmentation plays an important role in both qualitative and quantitative analysis of medical ultrasound images. But the performance of the classical image-segmentation techniques degrades severely when they are applied to segment medical ...
Comments
Information & Contributors
Information
Published In
Oct 2024
801 pages
ISBN:978-3-031-72110-6
DOI:10.1007/978-3-031-72111-3
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2024.
Publisher
Springer-Verlag
Berlin, Heidelberg
Publication History
Published: 07 October 2024
Author Tags
Qualifiers
- Article
Contributors
Other Metrics
Bibliometrics & Citations
Bibliometrics
Article Metrics
- 0Total Citations
- 0Total Downloads
- Downloads (Last 12 months)0
- Downloads (Last 6 weeks)0
Reflects downloads up to 19 Feb 2025