Shape-Aware Organ Segmentation by Predicting Signed Distance Maps

@inproceedings{Xue2020ShapeAwareOS,
  title={Shape-Aware Organ Segmentation by Predicting Signed Distance Maps},
  author={Yuan Xue and Hui Tang and Zhi Qiao and Guanzhong Gong and Yong Yin and Zhen Qian and Chao Huang and Wei Fan and Xiaolei Huang},
  booktitle={AAAI},
  year={2020}
}
In this work, we propose to resolve the issue existing in current deep learning based organ segmentation systems that they often produce results that do not capture the overall shape of the target organ and often lack smoothness. Since there is a rigorous mapping between the Signed Distance Map (SDM) calculated from object boundary contours and the binary segmentation map, we exploit the feasibility of learning the SDM directly from medical scans. By converting the segmentation task into… Expand
How Distance Transform Maps Boost Segmentation CNNs: An Empirical Study
Incorporating distance transform maps of ground truth into segmentation CNNs has been an interesting new trend in the last year. Despite many great works leading to improvements on a variety ofExpand
Residual Moment Loss for Medical Image Segmentation
TLDR
This paper proposes a novel loss function, namely residual moment (RM) loss, to explicitly embed the location information of segmentation targets during the training of deep learning networks, and demonstrates the effectiveness of this loss, which significantly boosts the accuracy of segmentsation networks. Expand
How Distance Transform Maps Boost Segmentation CNNs: An Empirical Study
TLDR
The latest developments of these methods for incorporating distance transform maps of ground truth into segmentation CNNs in the 3D medical segmentation field are summarized and unsolved problems are indicated. Expand
Dual-Term Loss Function For Shape-Aware Medical Image Segmentation
TLDR
A novel dual-term loss function with a region-based term and a shape-aware term is proposed, which forces the network to learn more boundary details by maximizing the probability difference between the two sides of the boundary. Expand
Brain Surface Reconstruction from MRI Images based on Segmentation Networks Applying Signed Distance Maps
TLDR
A new network architecture is proposed that incorporates knowledge of signed distance fields and introduces an additional Laplacian loss to ensure that the prediction results retain shape information to solve the problem confronted in current deep learning skull stripping methods. Expand
Learning Geodesic Active Contours for Embedding Object Global Information in Segmentation CNNs
TLDR
It is demonstrated that introducing global information by GAC can significantly improve segmentation performance, especially on reducing the boundary errors and outliers, which is very useful in applications such as organ transplantation surgical planning and multi-modality image registration where boundary errors can be very harmful. Expand
SimCVD: Simple Contrastive Voxel-Wise Representation Distillation for Semi-Supervised Medical Image Segmentation
TLDR
SimCVD is presented, a simple contrastive distillation framework that significantly advances state-of-the-art voxelwise representation learning and hypothesize that dropout can be viewed as a minimal form of data augmentation and makes the network robust to representation collapse. Expand
Momentum Contrastive Voxel-wise Representation Learning for Semi-supervised Volumetric Medical Image Segmentation
TLDR
This work presents a novel Contrastive Voxel-wise Representation Learning (CVRL) method with geometric constraints to learn global-local visual representations for volumetric medical image segmentation with limited annotations and introduces a voxel-to-volume contrastive algorithm to learnglobal information from 3D images. Expand
3D Liver and Tumor Segmentation with CNNs Based on Region and Distance Metrics
TLDR
An improved V-net algorithm was applied for 3D liver and tumor segmentation based on region and distance metrics to solve the problem of the highly unbalanced liver and tumors segmentation. Expand
FDA: Feature Decomposition and Aggregation for Robust Airway Segmentation
  • Minghui Zhang, Xin Yu, +5 authors Yun Gu
  • Computer Science, Engineering
  • DART/FAIR@MICCAI
  • 2021
TLDR
A new dual-stream network is proposed to address the variability between the clean domain and noisy domain, which utilizes the clean CT scans and a small amount of labeled noisy CT scans for airway segmentation. Expand
...
1
2
3
4
...

References

SHOWING 1-10 OF 29 REFERENCES
SPNet: Shape Prediction Using a Fully Convolutional Neural Network
TLDR
The proposed deep fully convolutional neural network learns to predict shapes instead of learning pixel-wise classification, and is applied to X-ray images of cervical vertebra where shape is of utmost importance. Expand
Shape-Aware Complementary-Task Learning for Multi-Organ Segmentation
TLDR
This work introduces the idea of complementary-task learning to enforce shape-prior leveraging the existing target labels and proposes two complementary-tasks namely i) distance map regression and ii) contour map detection to explicitly encode the geometric properties of each organ. Expand
A Distance Map Regularized CNN for Cardiac Cine MR Image Segmentation
TLDR
Several cardiac indices that often serve as diagnostic biomarkers, specifically blood pool volume, myocardial mass, and ejection fraction, computed using the proposed regularization method are better correlated with the indices computed from the reference, ground truth segmentation. Expand
Automatic Multi-organ Segmentation Using Learning-Based Segmentation and Level Set Optimization
TLDR
A novel generic segmentation system for the fully automatic multi-organ segmentation from CT medical images that combines the advantages of learning-based approaches on point cloud-based shape representation with those of PDE-optimization-based level set approaches, such as high accuracy and the straightforward prevention of segment overlaps. Expand
Elastic Boundary Projection for 3D Medical Image Segmentation
TLDR
Elastic Boundary Projection allows boundary-based segmentation without cutting a 3D volume into slices or patches, which stands out from conventional 2D and 3D approaches. Expand
Integrating Deformable Modeling with 3D Deep Neural Network Segmentation
TLDR
A new loss function is presented that incorporates the level set based smoothing loss together with multi Dice loss to avoid an additional post processing step and shows promising results on a dataset of 48 chest CT angiography images with 16 target anatomies. Expand
3D U-Net: Learning Dense Volumetric Segmentation from Sparse Annotation
TLDR
The proposed network extends the previous u-net architecture from Ronneberger et al. by replacing all 2D operations with their 3D counterparts and performs on-the-fly elastic deformations for efficient data augmentation during training. Expand
Evaluation of segmentation methods on head and neck CT: Auto‐segmentation challenge 2015
TLDR
The results demonstrate a clear tendency toward more general purpose and fewer structure‐specific segmentation algorithms in the state‐of‐the‐art in segmentation of organs at risk for radiotherapy treatment. Expand
Motion Segmentation of Truncated Signed Distance Function Based Volumetric Surfaces
TLDR
A novel solution to the motion segmentations of TSDF volumes by solving sparse multi-body motion segmentation and computing likelihoods for each motion label in the RGB-D image space, and, a novel pairwise term based on gradients of the TSDF volume. Expand
Efficient multi‐scale 3D CNN with fully connected CRF for accurate brain lesion segmentation
TLDR
An efficient and effective dense training scheme which joins the processing of adjacent image patches into one pass through the network while automatically adapting to the inherent class imbalance present in the data, and improves on the state-of-the‐art for all three applications. Expand
...
1
2
3
...