# 3D Medical Image Segmentation With Distance Transform Maps
## Motivation: How Distance Transform Maps Boost Segmentation CNNs [(MIDL 2020)](https://2020.midl.io/papers/ma20a.html)
Incorporating the distance Transform maps of image segmentation labels into CNNs-based segmentation tasks has received significant attention in 2019. These methods can be classified into two main classes in terms of the main usage of distance transform maps.
- Designing new loss functions
- Adding an auxiliary task, e.g. distance map regression
![Overview](https://github.com/JunMa11/SegWithDistMap/blob/master/overview.PNG)
However, with these new methods on the one hand and the diversity of the specific implementations and dataset-related challenges on the other, it's hard to figure out which design can generalize well beyond the experiments in the original papers.
In this repository, we want to re-implement these methods (published in 2019) and evaluate them on the same 3D segmentation tasks (heart and liver tumor segmentation).
## Experiments
| Task | LA Contributor | GPU | LiTS Contributor | GPU |
| -------------------------------------- | ------------- | ---------- | ------------ | ---------- |
| Boundary loss | [Yiwen Zhang](https://github.com/whisney) | 2080ti | [Mengzhang Li](https://github.com/MengzhangLI) | TITIAN RTX |
| Hausdorff loss | [Yiwen Zhang](https://github.com/whisney) | 2080ti | [Mengzhang Li](https://github.com/MengzhangLI) | TITIAN RTX |
| Signed distance map loss (AAAI 2020) | [Zhan Wei](https://github.com/zhanwei33) | 1080ti | cancel | - |
| Multi-Head: FG DTM regression-L1 | [Yiwen Zhang](https://github.com/whisney) | 2080ti | cancel | - |
| Multi-Head: FG DTM regression-L2 | [Jianan Liu]() | 2080ti | cancel | - |
| Multi-Head: FG DTM regression-L1 + L2 | [Gaoxiang Chen](https://github.com/AMSTLHX) | 2080ti | cancel | - |
| Multi-Head: SDF regression-L1 | [Feng Cheng](836155475@qq.com) | TITAN X | [Chao Peng](https://github.com/AMSTLHX) | TITAN RTX |
| Multi-Head: SDF regression-L2 | [Rongfei Lv](https://github.com/lrfdl) | TITAN RTX | [Rongfei Lv](https://github.com/lrfdl) | TITAN RTX |
| Multi-Head: SDF regression-L1+L2 | [Yixin Wang](https://github.com/Wangyixinxin) | P100 | cancel | - |
| Add-Branch: FG DTM regression-L1 | [Yaliang Zhao](441926980) | TITAN RTX | cancel | - |
| Add-Branch: FG DTM regression-L2 | [Mengzhang Li](https://github.com/MengzhangLI) | TITIAN RTX | cancel | - |
| Add-Branch: FG DTM regression-L1+L2 | [Yixin Wang](https://github.com/Wangyixinxin) | P100 | cancel | - |
| Add-Branch: SDF regression-L1 | [Feng Cheng](836155475@qq.com) | TITAN X | [Yixin Wang](https://github.com/Wangyixinxin) | TITAN RTX |
| Add-Branch: SDF regression-L2 | [Feng Cheng](836155475@qq.com) | TITAN X | [Yixin Wang](https://github.com/Wangyixinxin) | P100 |
| Add-Branch: SDF regression-L1+L2 | [Yixin Wang](https://github.com/Wangyixinxin) | P100 | [Yunpeng Wang]() | TITAN XP |
> [Here](https://github.com/JunMa11/SegWithDistMap/tree/master/code) is the code, and trained modles can be downloaded from [Baidu Disk](https://pan.baidu.com/s/1E9SlHw4DXuvsqFQRD_HHag) (pw:mgn0).
## Related Work in 2019
### New loss functions
| Date | First author | Title | Official Code | Publication |
| ---- | ------------- | --------------------------- | -------------- | ------------------------------ |
| 2019 | Yuan Xue | Shape-Aware Organ Segmentation by Predicting Signed Distance Maps | None | [AAAI 2020](https://www.aaai.org/Papers/AAAI/2020GB/AAAI-XueY.1482.pdf) |
| 2019 | [Hoel Kervadec](https://scholar.google.com.hk/citations?user=yeFGhfgAAAAJ&hl=zh-CN&oi=sra) | Boundary loss for highly unbalanced segmentation | [pytorch](https://github.com/LIVIAETS/surface-loss) | [MIDL 2019](http://proceedings.mlr.press/v102/kervadec19a.html) |
|2019|Davood Karimi|Reducing the Hausdorff Distance in Medical Image Segmentation with Convolutional Neural Networks [(arxiv)](https://arxiv.org/abs/1904.10030) |None|[TMI 2019](https://ieeexplore.ieee.org/document/8767031)|
### Auxiliary tasks
| Date | First author | Title | Official Code | Publication |
| ---- | ------------- | --------------------------- | -------------- | ------------------------------ |
| 2019 | Yan Wang | Deep Distance Transform for Tubular Structure Segmentation in CT Scans | None | [CVPR2020](http://openaccess.thecvf.com/content_CVPR_2020/html/Wang_Deep_Distance_Transform_for_Tubular_Structure_Segmentation_in_CT_Scans_CVPR_2020_paper.html) |
| 2019 | [Shusil Dangi](https://scholar.google.com.hk/citations?user=h12ifugAAAAJ&hl=zh-CN&oi=sra) |A Distance Map Regularized CNN for Cardiac Cine MR Image Segmentation [(arxiv)](https://arxiv.org/abs/1901.01238) | None | [Medical Physics](https://aapm.onlinelibrary.wiley.com/doi/abs/10.1002/mp.13853) |
|2019|[Fernando Navarro](https://scholar.google.com.hk/citations?user=rRKrhrwAAAAJ&hl=zh-CN&oi=sra)|Shape-Aware Complementary-Task Learning for Multi-organ Segmentation [(arxiv)](https://arxiv.org/abs/1908.05099)|None| [MICCAI MLMI 2019](https://link.springer.com/chapter/10.1007/978-3-030-32692-0_71)|
|2019|[Balamurali Murugesan](https://scholar.google.com.hk/citations?user=TmuKf44AAAAJ&hl=en&oi=sra)|Psi-Net: Shape and boundary aware joint multi-task deep network for medical image segmentation [(arXiv)](https://arxiv.org/abs/1902.04099)|None|[EMBC](https://ieeexplore.ieee.org/abstract/document/8857339)|
|2019|[Balamurali Murugesan](https://scholar.google.com.hk/citations?user=TmuKf44AAAAJ&hl=en&oi=sra)|Conv-MCD: A Plug-and-Play Multi-task Module for Medical Image Segmentation [(arXiv)](https://arxiv.org/abs/1908.05311)|[Pytorch](https://github.com/Bala93/Multi-task-deep-network)|[MLMI](https://link.springer.com/chapter/10.1007/978-3-030-32692-0_34)|
## Acknowledgments
The authors would like to thank the organization team of MICCAI 2017 liver tumor segmentation challenge MICCAI 2018 and left atrial segmentation challenge for the publicly available dataset.
We also thank the reviewers for their valuable comments and suggestions.
We appreciate Cheng Chen, Feng Cheng, Mengzhang Li, Chengwei Su, Chengfeng Zhou and Yaliang Zhao to help us finish some experiments.
Last but not least, we thank Lequan Yu for his great PyTorch implementation of [V-Net](https://github.com/yulequan/UA-MT) and Fabian Isensee for his great PyTorch implementation of [nnU-Net](https://github.com/MIC-DKFZ/nnUNett).
## Including the following citation in your work would be highly appreciated.
```
@inproceedings{ma-MIDL2020-SegWithDist,
title={How Distance Transform Maps Boost Segmentation CNNs: An Empirical Study},
author={Ma, Jun and Wei, Zhan and Zhang, Yiwen and Wang, Yixin and Lv, Rongfei and Zhu, Cheng and Chen, Gaoxiang and Liu, Jianan and Peng, Chao and Wang, Lei and Wang, Yunpeng and Chen, Jianan},
booktitle={Medical Imaging with Deep Learning},
pages = {479--492},
volume = {121},
month = {06--08 Jul},
year={2020},
series = {Proceedings of Machine Learning Research},
editor = {Tal Arbel and Ismail Ben Ayed and Marleen de Bruijne and Maxime Descoteaux and Herve Lombaert and Christopher Pal},
publisher = {PMLR},
url = {http://proceedings.mlr.press/v121/ma20b.html}
}
```
没有合适的资源?快使用搜索试试~ 我知道了~
距离变换图如何促进分割CNN:一项实证研究.zip
共55个文件
py:47个
list:3个
md:3个
1.该资源内容由用户上传,如若侵权请联系客服进行举报
2.虚拟产品一经售出概不退款(资源遇到问题,请及时私信上传者)
2.虚拟产品一经售出概不退款(资源遇到问题,请及时私信上传者)
版权申诉
0 下载量 84 浏览量
2023-03-31
22:32:29
上传
评论
收藏 187KB ZIP 举报
温馨提示
距离变换图如何促进分割CNN:一项实证研究
资源推荐
资源详情
资源评论
收起资源包目录
距离变换图如何促进分割CNN:一项实证研究.zip (55个子文件)
SegWithDistMap-master
overview.PNG 60KB
data
test_unlabel.list 1KB
train.list 2KB
test.list 420B
README.md 317B
LICENSE 11KB
README.md 8KB
code
train_LITS_Rec_SDF_L1.py 9KB
train_LA_MultiHead_FGDTM_L1PlusL2.py 8KB
test_LA_AAAISDF.py 8KB
train_LA_MultiHead_FGDTM_L2.py 8KB
test_LA_Rec_FGDTM.py 8KB
train_LA_AAAISDF_L1.py 10KB
train_LA_Rec_SDF_L2.py 9KB
networks
vnet_sdf.py 9KB
vnet.py 9KB
vnet_rec.py 10KB
vnet_multi_task.py 9KB
vnet_multi_head.py 9KB
test_LITS.py 7KB
train_LA_AAAISDF.py 10KB
train_LITS_HD.py 10KB
train_LA_MultiHead_FGDTM_L1.py 8KB
test_LA.py 2KB
test_LA_Rec_SDF.py 8KB
test_LITS_Rec_SDF.py 8KB
test_LITS_MultiHead_SDF.py 8KB
utils
losses.py 10KB
util.py 4KB
ramps.py 1KB
train_LA_Rec_FGDTM_L1PlusL2.py 8KB
test_util.py 5KB
train_LA_Rec_SDF_L1.py 9KB
train_LiTS_MultiHead_SDF_L2.py 9KB
train_LA_BD.py 11KB
train_LITS.py 6KB
train_LITS_Rec_SDF_L2.py 9KB
test_LA_MultiHead_SDF.py 8KB
train_LA_Rec_SDF_L1PlusL2.py 9KB
dataloaders
utils.py 7KB
livertumor.py 8KB
la_heart.py 8KB
train_LA.py 6KB
train_LITS_BD.py 10KB
train_LITS_Rec_SDF_L1PlusL2.py 9KB
train_LA_MultiHead_SDF_L1PlusL2.py 9KB
train_LA_HD.py 10KB
train_LA_Rec_FGDTM_L2.py 8KB
train_LiTS_MultiHead_SDF_L1.py 9KB
train_LA_SDF.py 9KB
train_LA_MultiHead_SDF_L2.py 9KB
test_LA_MultiHead_FGDTM.py 8KB
README.md 3KB
train_LA_Rec_FGDTM_L1.py 8KB
train_LA_MultiHead_SDF_L1.py 9KB
共 55 条
- 1
资源评论
快撑死的鱼
- 粉丝: 1w+
- 资源: 9154
上传资源 快速赚钱
- 我的内容管理 展开
- 我的资源 快来上传第一个资源
- 我的收益 登录查看自己的收益
- 我的积分 登录查看自己的积分
- 我的C币 登录后查看C币余额
- 我的收藏
- 我的下载
- 下载帮助
安全验证
文档复制为VIP权益,开通VIP直接复制
信息提交成功