## Speed up MR scanner with generative priors for image reconstruction (SPRECO)
<img src="./misc/overview.png" alt="workflow" width="350" align="right"/>
This package is to help you train generative image priors of MRI images and then use them in image reconstruction. It has the following features:
1. Distributed training
2. Interruptible training
3. Efficient dataloader for medical images
4. Customizable with a configuration file
5. Seamless deployment with [BART](https://github.com/mrirecon/bart)
**Installation:** Clone this repository and use [conda](https://www.anaconda.com/products/individual) to set up the environment.
```shell
$ git clone https://github.com/mrirecon/spreco.git
$ cd spreco
$ pip install .
```
<!--
## Quickstart with colab
1. Sample the posterior
- [Jupyter Notebook](https://github.com/mrirecon/spreco/blob/main/examples/scripts/demo_recon.ipynb)
- [](https://colab.research.google.com/github/mrirecon/spreco/blob/main/examples/scripts/demo_recon.ipynb)
2. Train an image prior
- [Jupyter Notebook](https://github.com/mrirecon/spreco/blob/main/examples/scripts/demo_train.ipynb)
- [](https://colab.research.google.com/github/mrirecon/spreco/blob/main/examples/scripts/demo_train.ipynb)
3. Using Prior with BART
- [Jupyter Notebook](https://github.com/mrirecon/bart-workshop/blob/master/ismrm2021/bart_tensorflow/bart_tf.ipynb)
- [](https://colab.research.google.com/github/mrirecon/bart-workshop/blob/master/ismrm2021/bart_tensorflow/bart_tf.ipynb)-->
## Reference
We would appreciate it if you tried our codes and cited our work.
[1] G. Luo, X. Wang, M. Blumenthal, M. Schilling, EHU. Rauf, R. Kotikalapudi, NK. Focke, M. Uecker. Generative image priors for MRI reconstruction trained from magnitude-only images. arXiv preprint arXiv:2308.02340 (2023)
[2] G. Luo, M. Blumenthal, M. Heide, M. Uecker. Bayesian MRI reconstruction with joint uncertainty estimation using diffusion models. Magn Reson Med. 2023; 1-17
[3] M. Blumenthal, G. Luo, M. Schilling, HCM. Holme, M. Uecker. Deep, deep learning with BART. Magn Reson Med. 2023; 89: 678- 693.
[4] G. Luo, N. Zhao, W. Jiang, ES. Hui, P. Cao. MRI reconstruction using deep Bayesian estimation. Magn Reson Med. 2020; 84: 2246-2261.
Raw data
{
"_id": null,
"home_page": "",
"name": "spreco",
"maintainer": "",
"docs_url": null,
"requires_python": ">=3.7",
"maintainer_email": "",
"keywords": "MRI reconstruction,generative model,image priors,sampling",
"author": "",
"author_email": "Guanxiong Luo <guanxiong.luo@med.uni-goettingen.de>",
"download_url": "https://files.pythonhosted.org/packages/3f/77/7bf234f6ab11d00cd1f08ac87746297127b9be68c0b2b51102e3e4ec9b48/spreco-0.0.4.tar.gz",
"platform": null,
"description": "## Speed up MR scanner with generative priors for image reconstruction (SPRECO)\n\n<img src=\"./misc/overview.png\" alt=\"workflow\" width=\"350\" align=\"right\"/>\nThis package is to help you train generative image priors of MRI images and then use them in image reconstruction. It has the following features:\n\n1. Distributed training\n2. Interruptible training\n3. Efficient dataloader for medical images\n4. Customizable with a configuration file\n5. Seamless deployment with [BART](https://github.com/mrirecon/bart)\n\n**Installation:** Clone this repository and use [conda](https://www.anaconda.com/products/individual) to set up the environment.\n\n```shell\n$ git clone https://github.com/mrirecon/spreco.git\n$ cd spreco\n$ pip install .\n```\n<!-- \n## Quickstart with colab\n\n1. Sample the posterior \n - [Jupyter Notebook](https://github.com/mrirecon/spreco/blob/main/examples/scripts/demo_recon.ipynb)\n - [](https://colab.research.google.com/github/mrirecon/spreco/blob/main/examples/scripts/demo_recon.ipynb)\n2. Train an image prior\n - [Jupyter Notebook](https://github.com/mrirecon/spreco/blob/main/examples/scripts/demo_train.ipynb)\n - [](https://colab.research.google.com/github/mrirecon/spreco/blob/main/examples/scripts/demo_train.ipynb)\n3. Using Prior with BART\n - [Jupyter Notebook](https://github.com/mrirecon/bart-workshop/blob/master/ismrm2021/bart_tensorflow/bart_tf.ipynb)\n - [](https://colab.research.google.com/github/mrirecon/bart-workshop/blob/master/ismrm2021/bart_tensorflow/bart_tf.ipynb)-->\n\n## Reference \nWe would appreciate it if you tried our codes and cited our work.\n\n[1] G. Luo, X. Wang, M. Blumenthal, M. Schilling, EHU. Rauf, R. Kotikalapudi, NK. Focke, M. Uecker. Generative image priors for MRI reconstruction trained from magnitude-only images. arXiv preprint arXiv:2308.02340 (2023)\n\n[2] G. Luo, M. Blumenthal, M. Heide, M. Uecker. Bayesian MRI reconstruction with joint uncertainty estimation using diffusion models. Magn Reson Med. 2023; 1-17\n\n[3] M. Blumenthal, G. Luo, M. Schilling, HCM. Holme, M. Uecker. Deep, deep learning with BART. Magn Reson Med. 2023; 89: 678- 693.\n\n[4] G. Luo, N. Zhao, W. Jiang, ES. Hui, P. Cao. MRI reconstruction using deep Bayesian estimation. Magn Reson Med. 2020; 84: 2246-2261.\n\n",
"bugtrack_url": null,
"license": "",
"summary": "Generative image priors for MRI image reconstruction",
"version": "0.0.4",
"project_urls": {
"Bug Tracker": "https://github.com/mrirecon/spreco/issues",
"Homepage": "https://github.com/mrirecon/spreco"
},
"split_keywords": [
"mri reconstruction",
"generative model",
"image priors",
"sampling"
],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "ac6b84890433a33fa42d8b007852572a7de64f01ab2f40c12cea274a06f6a85b",
"md5": "cdfb4b47b3253db54233aecb4505891f",
"sha256": "d75bbdc1f7bc89a783baac4b3f420875ead6c4aea890e3877ff81fbd0bab00d1"
},
"downloads": -1,
"filename": "spreco-0.0.4-py3-none-any.whl",
"has_sig": false,
"md5_digest": "cdfb4b47b3253db54233aecb4505891f",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": ">=3.7",
"size": 69842,
"upload_time": "2023-08-17T13:58:42",
"upload_time_iso_8601": "2023-08-17T13:58:42.108035Z",
"url": "https://files.pythonhosted.org/packages/ac/6b/84890433a33fa42d8b007852572a7de64f01ab2f40c12cea274a06f6a85b/spreco-0.0.4-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "3f777bf234f6ab11d00cd1f08ac87746297127b9be68c0b2b51102e3e4ec9b48",
"md5": "d7d7a0198e93293ddc4411db6fc7a688",
"sha256": "88746ff890cd10f6a3aac048fbbd62ae79a5cfdd118ca135785336c9d6500f05"
},
"downloads": -1,
"filename": "spreco-0.0.4.tar.gz",
"has_sig": false,
"md5_digest": "d7d7a0198e93293ddc4411db6fc7a688",
"packagetype": "sdist",
"python_version": "source",
"requires_python": ">=3.7",
"size": 57810,
"upload_time": "2023-08-17T13:58:43",
"upload_time_iso_8601": "2023-08-17T13:58:43.449128Z",
"url": "https://files.pythonhosted.org/packages/3f/77/7bf234f6ab11d00cd1f08ac87746297127b9be68c0b2b51102e3e4ec9b48/spreco-0.0.4.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2023-08-17 13:58:43",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "mrirecon",
"github_project": "spreco",
"travis_ci": false,
"coveralls": false,
"github_actions": false,
"lcname": "spreco"
}
JSON
