Deep Conditional Shape Models for 3D cardiac image segmentation
Not on the map. Why?
Abstract (arXiv)
Delineation of anatomical structures is often the first step of many medical image analysis workflows. While convolutional neural networks achieve high performance, these do not incorporate anatomical shape information. We introduce a novel segmentation algorithm that uses Deep Conditional Shape models (DCSMs) as a core component. Using deep implicit shape representations, the algorithm learns a modality-agnostic shape model that can generate the signed distance functions for any anatomy of interest. To fit the generated shape to the image, the shape model is conditioned on anatomic landmarks that can be automatically detected or provided by the user. Finally, we add a modality-dependent, lightweight refinement network to capture any fine details not represented by the implicit function. The proposed DCSM framework is evaluated on the problem of cardiac left ventricle (LV) segmentation from multiple 3D modalities (contrast-enhanced CT, non-contrasted CT, 3D echocardiography-3DE). We demonstrate that the automatic DCSM outperforms the baseline for non-contrasted CT without the local refinement, and with the refinement for contrasted CT and 3DE, especially with significant improvement in the Hausdorff distance. The semi-automatic DCSM with user-input landmarks, while only trained on contrasted CT, achieves greater than 92% Dice for all modalities. Both automatic DCSM with refinement and semi-automatic DCSM achieve equivalent or better performance compared to inter-user variability for these modalities.
Key facts. Cited by 3 papers in the map (data through 14 Oct 2026). Builds on 23 of its 36 references that are in the map. First posted on arXiv 16 Oct 2023.
Builds on
23 of 36 references.
References · 23
- 2017Prototypical Networks for Few-shot Learningcited by 3,015
- 2019DeepSDF: Learning Continuous Signed Distance Functions for Shape Representationcited by 2,689
- 2020Fourier Features Let Networks Learn High Frequency Functions in Low Dimensional Domainscited by 2,200
- 2018Occupancy Networks: Learning 3D Reconstruction in Function Spacecited by 2,041
- 2017CyCADA: Cycle-Consistent Adversarial Domain Adaptationcited by 1,082
- 2018Learning Implicit Fields for Generative Shape Modelingcited by 1,060
- 2020Convolutional Occupancy Networkscited by 761
- 2020Learning Continuous Image Representation with Local Implicit Image Functioncited by 552
- 2019DISN: Deep Implicit Surface Network for High-quality Single-view 3D Reconstructioncited by 333
- 2017Convolutional Neural Networks for Medical Image Analysis: Full Training or Fine Tuning?cited by 221
- 2020Shape-aware Semi-supervised 3D Semantic Segmentation for Medical Imagescited by 152
- 2019Deep learning for cardiac image segmentation: A reviewcited by 131
11 more
- 2017Anatomically Constrained Neural Networks (ACNN): Application to Cardiac Image Enhancement and Segmentationcited by 109
- 2019Deep Level Sets: Implicit Surface Representations for 3D Shape Inferencecited by 53
- 2020Cardiac Segmentation with Strong Anatomical Guaranteescited by 44
- 2022GenSDF: Two-Stage Learning of Generalizable Signed Distance Functionscited by 30
- 2017Deep Learning for Multi-Task Medical Image Segmentation in Multiple Modalitiescited by 25
- 2017Automatic Liver Segmentation Using an Adversarial Image-to-Image Networkcited by 17
- 2022Domain Adaptation Meets Zero-Shot Learning: An Annotation-Efficient Approach to Multi-Modality Medical Image Segmentationcited by 16
- 2019A Survey on Deep Learning of Small Sample in Biomedical Image Analysiscited by 7
- 2022Graph Convolutional Networks for Multi-modality Medical Imaging: Methods, Architectures, and Clinical Applicationscited by 7
- 2017Adversarial training and dilated convolutions for brain MRI segmentationcited by 6
- 2022PointResNet: Residual Network for 3D Point Cloud Segmentation and Classificationcited by 1
Built on by
Citing papers per year: 2024: 2; 2025: 0; 2026: 1.
3 papers.