SAML / Git / Diff of /README.md

Models:

SCallahan/

SAML

Downloads: 1

Diff of /README.md [7b5b9f] .. [5921ce]

Switch to unified view


# SAML & A Multi-site Dataset for Prostate MRI Segmentation
by [Quande Liu](https://github.com/liuquande), [Qi Dou](http://www.cse.cuhk.edu.hk/~qdou/), [Pheng-Ann Heng](http://www.cse.cuhk.edu.hk/~pheng/). 

### Introduction

* The Tensorflow implementation for our MICCAI 2020 paper '[Shape-aware Meta-learning for Generalizing Prostate MRI Segmentation to Unseen Domains](https://arxiv.org/pdf/2007.02035.pdf)'. 

<p align="center">
  <img src="https://github.com/liuquande/SAML/blob/master/figure/saml.png?raw=true"  width="650"/>
</p>

* A well-organized multi-site dataset (from six data sources) for prostate MRI segmentation, that can support research in various problem settings with need of multi-site data, such as Domain Generalization, Multi-site Learning and Life-long Learning, etc. For more details and downloading link of the dataset, please [Find Here](https://liuquande.github.io/SAML/).
    

<p align="center">
  <img src="https://github.com/liuquande/SAML/blob/master/figure/protocol.png?raw=true"  width="650"/>
</p>
  

### Setup & Usage for the Code

1. Check dependencies:
   ```shell
   python==2.7.17
   numpy==1.16.6
   scipy==1.2.1
   tensorflow-gpu==1.12.0
   tensorboard==1.12.2
   SimpleITK==1.2.0
   ```
2. To train the model, you need to specify the training configurations (can simply use the default setting) in main.py, then run:
   ```shell
   python main.py --phase=train
   ```

2. To evaluate the model, run:
   ```shell
   python main.py --phase=test --restore_model='/path/to/test_model.cpkt'
   ```
   You will see the output results in the folder `./output/`.

### Citation
If this repository is useful for your research, please cite:

```
@article{liu2020shape,
  title={Shape-aware Meta-learning for Generalizing Prostate MRI Segmentation to Unseen Domains},
  author={Liu, Quande and Dou, Qi and Heng, Pheng-Ann},
  journal={International Conference on Medical Image Computing and Computer Assisted Intervention},
  year={2020}
}
```

### Questions

For further question about the code or dataset, please contact 'qdliu@cse.cuhk.edu.hk'

	a/README.md		b/README.md
1	# SAML & A Multi-site Dataset for Prostate MRI Segmentation	1	# SAML & A Multi-site Dataset for Prostate MRI Segmentation
2	by [Quande Liu](https://github.com/liuquande), [Qi Dou](http://www.cse.cuhk.edu.hk/~qdou/), [Pheng-Ann Heng](http://www.cse.cuhk.edu.hk/~pheng/).	2	by [Quande Liu](https://github.com/liuquande), [Qi Dou](http://www.cse.cuhk.edu.hk/~qdou/), [Pheng-Ann Heng](http://www.cse.cuhk.edu.hk/~pheng/).
3		3
4	### Introduction	4	### Introduction
5		5
6	* The Tensorflow implementation for our MICCAI 2020 paper '[Shape-aware Meta-learning for Generalizing Prostate MRI Segmentation to Unseen Domains](https://arxiv.org/pdf/2007.02035.pdf)'.	6	* The Tensorflow implementation for our MICCAI 2020 paper '[Shape-aware Meta-learning for Generalizing Prostate MRI Segmentation to Unseen Domains](https://arxiv.org/pdf/2007.02035.pdf)'.
7		7
8	<p align="center">	8	<p align="center">
9	<img src="figure/saml.png" width="650"/>	9	<img src="https://github.com/liuquande/SAML/blob/master/figure/saml.png?raw=true" width="650"/>
10	</p>	10	</p>
11		11
12	* A well-organized multi-site dataset (from six data sources) for prostate MRI segmentation, that can support research in various problem settings with need of multi-site data, such as Domain Generalization, Multi-site Learning and Life-long Learning, etc. For more details and downloading link of the dataset, please [Find Here](https://liuquande.github.io/SAML/).	12	* A well-organized multi-site dataset (from six data sources) for prostate MRI segmentation, that can support research in various problem settings with need of multi-site data, such as Domain Generalization, Multi-site Learning and Life-long Learning, etc. For more details and downloading link of the dataset, please [Find Here](https://liuquande.github.io/SAML/).
13		13
14		14
15	<p align="center">	15	<p align="center">
16	<img src="figure/protocol.png" width="650"/>	16	<img src="https://github.com/liuquande/SAML/blob/master/figure/protocol.png?raw=true" width="650"/>
17	</p>	17	</p>
18		18
19		19
20	### Setup & Usage for the Code	20	### Setup & Usage for the Code
21		21
22	1. Check dependencies:	22	1. Check dependencies:
23	```shell	23	```shell
24	python==2.7.17	24	python==2.7.17
25	numpy==1.16.6	25	numpy==1.16.6
26	scipy==1.2.1	26	scipy==1.2.1
27	tensorflow-gpu==1.12.0	27	tensorflow-gpu==1.12.0
28	tensorboard==1.12.2	28	tensorboard==1.12.2
29	SimpleITK==1.2.0	29	SimpleITK==1.2.0
30	```	30	```
31	2. To train the model, you need to specify the training configurations (can simply use the default setting) in main.py, then run:	31	2. To train the model, you need to specify the training configurations (can simply use the default setting) in main.py, then run:
32	```shell	32	```shell
33	python main.py --phase=train	33	python main.py --phase=train
34	```	34	```
35		35
36	2. To evaluate the model, run:	36	2. To evaluate the model, run:
37	```shell	37	```shell
38	python main.py --phase=test --restore_model='/path/to/test_model.cpkt'	38	python main.py --phase=test --restore_model='/path/to/test_model.cpkt'
39	```	39	```
40	You will see the output results in the folder `./output/`.	40	You will see the output results in the folder `./output/`.
41		41
42	### Citation	42	### Citation
43	If this repository is useful for your research, please cite:	43	If this repository is useful for your research, please cite:
44		44
45	```	45	```
46	@article{liu2020shape,	46	@article{liu2020shape,
47	title={Shape-aware Meta-learning for Generalizing Prostate MRI Segmentation to Unseen Domains},	47	title={Shape-aware Meta-learning for Generalizing Prostate MRI Segmentation to Unseen Domains},
48	author={Liu, Quande and Dou, Qi and Heng, Pheng-Ann},	48	author={Liu, Quande and Dou, Qi and Heng, Pheng-Ann},
49	journal={International Conference on Medical Image Computing and Computer Assisted Intervention},	49	journal={International Conference on Medical Image Computing and Computer Assisted Intervention},
50	year={2020}	50	year={2020}
51	}	51	}
52	```	52	```
53		53
54	### Questions	54	### Questions
55		55
56	For further question about the code or dataset, please contact 'qdliu@cse.cuhk.edu.hk'	56	For further question about the code or dataset, please contact 'qdliu@cse.cuhk.edu.hk'