| Name | medisynth JSON |
| Version |
0.0.4
JSON |
| download |
| home_page | |
| Summary | Python-based library for Synthesis of Ultrasound Fetal Imaging |
| upload_time | 2023-07-15 18:41:36 |
| maintainer | |
| docs_url | None |
| author | |
| requires_python | >=3.11 |
| license | MIT |
| keywords |
artificial intelligence
diffusion models
|
| VCS |
 |
| bugtrack_url |
|
| requirements |
No requirements were recorded.
|
| Travis-CI |
No Travis.
|
| coveralls test coverage |
No coveralls.
|
# medisynth - library for synthesis of ultrasound fetal imaging (:baby: :brain: :robot:) :warning: WIP :warning:
[](https://badge.fury.io/py/medisynth)
medisynth is a python-based library to syntheses fetal ultrasound images using GAN, transformers and diffusion models.
It also includes methods to quantify the quality of synthesis (FID, PSNR, SSIM, and visual touring tests).
## Installation
```
$ pip install medisynth
```
You can develop locally:
* Generate your SSH keys as suggested [here](https://docs.github.com/en/github/authenticating-to-github/generating-a-new-ssh-key-and-adding-it-to-the-ssh-agent) (or [here](https://github.com/mxochicale/tools/blob/main/github/SSH.md))
* Clone the repository by typing (or copying) the following line in a terminal at your selected path in your machine:
```
cd && mkdir -p $HOME/repositories/budai4medtech && cd $HOME/repositories/budai4medtech
git clone git@github.com:budai4medtech/medisynth.git
```
## References
### Presentation
Good practices in AI/ML for Ultrasound Fetal Brain Imaging Synthesis
Harvey Mannering, Sofia Miñano, and Miguel Xochicale
University College London
The deep learning and computer vision Journal Club
UCL Centre for Advance Research Computing
1st of June 2023, 15:00 GMT
Abstract
Medical image datasets for AI and ML methods must be diverse to generalise well unseen data (i.e. diagnoses, diseases, pathologies, scanners, demographics, etc).
However there are few public ultrasound fetal imaging datasets due to insufficient amounts of clinical data, patient privacy, rare occurrence of abnormalities, and limited experts for data collection and validation.
To address such challenges in Ultrasound Medical Imaging, Miguel will discuss two proposed generative adversarial networks (GAN)-based models: diffusion-super-resolution-GAN and transformer-based-GAN, to synthesise images of fetal Ultrasound brain image planes from one public dataset.
Similarly, Miguel will present and discuss AI and ML workflow aligned to good ML practices by FDA, and methods for quality image assessment (e.g., visual Turing test and FID scores).
Finally, a simple prototype in GitHub, google-colabs and guidelines to train it using Myriad cluster will be presented as well as applications for Medical Image Synthesis e.g., classification, augmentation, segmentation, registration and other downstream tasks, etc. will be discussed.
The resources to reproduce the work of this talk are available at https://github.com/budai4medtech/medisynth.
### Citations
BibTeX to cite
```
@misc{iskandar2023realistic,
author={
Michelle Iskandar and
Harvey Mannering and
Zhanxiang Sun and
Jacqueline Matthew and
Hamideh Kerdegari and
Laura Peralta and
Miguel Xochicale},
title={Towards Realistic Ultrasound Fetal Brain Imaging Synthesis},
year={2023},
eprint={2304.03941},
archivePrefix={arXiv},
primaryClass={eess.IV}
}
```
## Contributors
Thanks goes to all these people ([emoji key](https://allcontributors.org/docs/en/emoji-key)):
<!-- ALL-CONTRIBUTORS-LIST:START - Do not remove or modify this section -->
<!-- prettier-ignore-start -->
<!-- markdownlint-disable -->
<table>
<tr>
<!-- CONTRIBUTOR -->
<td align="center">
<!-- ADD GITHUB USERNAME AND HASH FOR GITHUB PHOTO -->
<a href="https://github.com/???"><img src="https://avatars1.githubusercontent.com/u/23114020?v=4?s=100" width="100px;" alt=""/>
<br />
<sub> <b>ADD NAME SURNAME</b> </sub>
</a>
<br />
<!-- ADD GITHUB REPOSITORY AND PROJECT, TITLE AND EMOJIS -->
<a href="https://github.com/$PROJECTNAME/$REPOSITORY_NAME/commits?author=" title="Research"> 🔬 🤔 </a>
</td>
<!-- CONTRIBUTOR -->
<td align="center">
<a href="https://github.com/sfmig"><img src="https://avatars1.githubusercontent.com/u/33267254?v=4?s=100" width="100px;" alt=""/>
<br />
<sub> <b>Sofia Miñano</b> </sub>
</a>
<br />
<a href="https://github.com/budai4medtech/medisynth/commits?author=sfmig" title="Code">💻</a>
<a href="https://github.com/budai4medtech/medisynth/commits?author=sfmig" title="Research"> 🔬 🤔 </a>
</td>
<!-- CONTRIBUTOR -->
<td align="center">
<a href="https://github.com/harveymannering"><img src="https://avatars1.githubusercontent.com/u/60523103?v=4?s=100" width="100px;" alt=""/>
<br />
<sub> <b>Harvey Mannering</b> </sub>
</a>
<br />
<a href="https://github.com/budai4medtech/medisynth/commits?author=harveymannering" title="Code">💻</a>
<a href="https://github.com/budai4medtech/medisynth/commits?author=harveymannering" title="Research"> 🔬 🤔 </a>
</td>
<!-- CONTRIBUTOR -->
<td align="center">
<a href="https://github.com/budai4medtech"><img src="https://avatars1.githubusercontent.com/u/11370681?v=4?s=100" width="100px;" alt=""/>
<br />
<sub><b>Miguel Xochicale</b></sub>
<br />
</a>
<a href="https://github.com/budai4medtech/medisynth/commits?author=mxochicale" title="Code">💻</a>
<a href="ttps://github.com/budai4medtech/medisynth/commits?author=mxochicale" title="Documentation">📖 🔧 </a>
</td>
</tr>
</table>
<!-- markdownlint-restore -->
<!-- prettier-ignore-end -->
<!-- ALL-CONTRIBUTORS-LIST:END -->
This work follows the [all-contributors](https://github.com/all-contributors/all-contributors) specification.
Contributions of any kind welcome!
Raw data
{
"_id": null,
"home_page": "",
"name": "medisynth",
"maintainer": "",
"docs_url": null,
"requires_python": ">=3.11",
"maintainer_email": "",
"keywords": "artificial intelligence,diffusion models",
"author": "",
"author_email": "Miguel Xochicale <your0@email.com>, Harvey Mannering <your1@email.com>, Sof\u00eda Mi\u00f1ano <your2@email.com>",
"download_url": "https://files.pythonhosted.org/packages/c9/16/47de617593bff1bebf53a3c566891ec7c9ea22867191203343a5e787704d/medisynth-0.0.4.tar.gz",
"platform": null,
"description": "# medisynth - library for synthesis of ultrasound fetal imaging (:baby: :brain: :robot:) :warning: WIP :warning:\n[](https://badge.fury.io/py/medisynth)\n\nmedisynth is a python-based library to syntheses fetal ultrasound images using GAN, transformers and diffusion models.\nIt also includes methods to quantify the quality of synthesis (FID, PSNR, SSIM, and visual touring tests).\n\n## Installation\n```\n$ pip install medisynth\n```\n\nYou can develop locally:\n* Generate your SSH keys as suggested [here](https://docs.github.com/en/github/authenticating-to-github/generating-a-new-ssh-key-and-adding-it-to-the-ssh-agent) (or [here](https://github.com/mxochicale/tools/blob/main/github/SSH.md))\n* Clone the repository by typing (or copying) the following line in a terminal at your selected path in your machine:\n```\ncd && mkdir -p $HOME/repositories/budai4medtech && cd $HOME/repositories/budai4medtech\ngit clone git@github.com:budai4medtech/medisynth.git\n```\n\n## References\n### Presentation\nGood practices in AI/ML for Ultrasound Fetal Brain Imaging Synthesis \nHarvey Mannering, Sofia Mi\u00f1ano, and Miguel Xochicale \n\nUniversity College London \nThe deep learning and computer vision Journal Club \nUCL Centre for Advance Research Computing \n1st of June 2023, 15:00 GMT \n\nAbstract\nMedical image datasets for AI and ML methods must be diverse to generalise well unseen data (i.e. diagnoses, diseases, pathologies, scanners, demographics, etc).\nHowever there are few public ultrasound fetal imaging datasets due to insufficient amounts of clinical data, patient privacy, rare occurrence of abnormalities, and limited experts for data collection and validation.\nTo address such challenges in Ultrasound Medical Imaging, Miguel will discuss two proposed generative adversarial networks (GAN)-based models: diffusion-super-resolution-GAN and transformer-based-GAN, to synthesise images of fetal Ultrasound brain image planes from one public dataset.\nSimilarly, Miguel will present and discuss AI and ML workflow aligned to good ML practices by FDA, and methods for quality image assessment (e.g., visual Turing test and FID scores).\nFinally, a simple prototype in GitHub, google-colabs and guidelines to train it using Myriad cluster will be presented as well as applications for Medical Image Synthesis e.g., classification, augmentation, segmentation, registration and other downstream tasks, etc. will be discussed.\nThe resources to reproduce the work of this talk are available at https://github.com/budai4medtech/medisynth.\n\n### Citations\nBibTeX to cite\n```\n@misc{iskandar2023realistic,\n author={\n \tMichelle Iskandar and \n \tHarvey Mannering and \n \tZhanxiang Sun and \n \tJacqueline Matthew and \n \tHamideh Kerdegari and \n \tLaura Peralta and \n \tMiguel Xochicale},\n title={Towards Realistic Ultrasound Fetal Brain Imaging Synthesis}, \n year={2023},\n eprint={2304.03941},\n archivePrefix={arXiv},\n primaryClass={eess.IV}\n}\n``` \n\n## Contributors\nThanks goes to all these people ([emoji key](https://allcontributors.org/docs/en/emoji-key)): \n<!-- ALL-CONTRIBUTORS-LIST:START - Do not remove or modify this section -->\n<!-- prettier-ignore-start -->\n<!-- markdownlint-disable -->\n<table>\n <tr>\n\t<!-- CONTRIBUTOR -->\n\t<td align=\"center\">\n\t\t<!-- ADD GITHUB USERNAME AND HASH FOR GITHUB PHOTO -->\n\t\t<a href=\"https://github.com/???\"><img src=\"https://avatars1.githubusercontent.com/u/23114020?v=4?s=100\" width=\"100px;\" alt=\"\"/>\n\t\t<br />\n\t\t\t<sub> <b>ADD NAME SURNAME</b> </sub> \n\t\t</a>\n\t\t<br />\n\t\t\t<!-- ADD GITHUB REPOSITORY AND PROJECT, TITLE AND EMOJIS -->\n\t\t\t<a href=\"https://github.com/$PROJECTNAME/$REPOSITORY_NAME/commits?author=\" title=\"Research\"> \ud83d\udd2c \ud83e\udd14 </a>\n\t</td>\n\t<!-- CONTRIBUTOR -->\n\t<td align=\"center\">\n\t\t<a href=\"https://github.com/sfmig\"><img src=\"https://avatars1.githubusercontent.com/u/33267254?v=4?s=100\" width=\"100px;\" alt=\"\"/>\n\t\t<br />\n\t\t\t<sub> <b>Sofia Mi\u00f1ano</b> </sub> \n\t\t</a>\n\t\t<br />\n\t\t\t<a href=\"https://github.com/budai4medtech/medisynth/commits?author=sfmig\" title=\"Code\">\ud83d\udcbb</a> \n\t\t\t<a href=\"https://github.com/budai4medtech/medisynth/commits?author=sfmig\" title=\"Research\"> \ud83d\udd2c \ud83e\udd14 </a>\n\t</td>\n\t<!-- CONTRIBUTOR -->\n\t<td align=\"center\">\n\t\t<a href=\"https://github.com/harveymannering\"><img src=\"https://avatars1.githubusercontent.com/u/60523103?v=4?s=100\" width=\"100px;\" alt=\"\"/>\n\t\t<br />\n\t\t\t<sub> <b>Harvey Mannering</b> </sub> \n\t\t</a>\n\t\t<br />\n\t\t\t<a href=\"https://github.com/budai4medtech/medisynth/commits?author=harveymannering\" title=\"Code\">\ud83d\udcbb</a> \n\t\t\t<a href=\"https://github.com/budai4medtech/medisynth/commits?author=harveymannering\" title=\"Research\"> \ud83d\udd2c \ud83e\udd14 </a>\n\t</td>\n\t<!-- CONTRIBUTOR -->\n\t<td align=\"center\">\n\t\t<a href=\"https://github.com/budai4medtech\"><img src=\"https://avatars1.githubusercontent.com/u/11370681?v=4?s=100\" width=\"100px;\" alt=\"\"/>\n\t\t\t<br />\n\t\t\t<sub><b>Miguel Xochicale</b></sub> \n\t\t\t<br />\n\t\t</a>\n\t\t\t<a href=\"https://github.com/budai4medtech/medisynth/commits?author=mxochicale\" title=\"Code\">\ud83d\udcbb</a> \n\t\t\t<a href=\"ttps://github.com/budai4medtech/medisynth/commits?author=mxochicale\" title=\"Documentation\">\ud83d\udcd6 \ud83d\udd27 </a>\n\t</td>\n </tr>\n</table>\n<!-- markdownlint-restore -->\n<!-- prettier-ignore-end -->\n\n<!-- ALL-CONTRIBUTORS-LIST:END -->\n\nThis work follows the [all-contributors](https://github.com/all-contributors/all-contributors) specification. \nContributions of any kind welcome!\n",
"bugtrack_url": null,
"license": "MIT",
"summary": "Python-based library for Synthesis of Ultrasound Fetal Imaging",
"version": "0.0.4",
"project_urls": {
"homepage": "https://github.com/budai4medtech/medisynth"
},
"split_keywords": [
"artificial intelligence",
"diffusion models"
],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "16a39c6424da97148b2f2faf78c0284771fe7abb9a52ebba2057ec6be9f17bba",
"md5": "71e152d0b4a6bf04c546a9b2c1db0d6c",
"sha256": "c7bbef7a13bfd89247eed408c21de09a2dfc80b055005945938c054a53d9419c"
},
"downloads": -1,
"filename": "medisynth-0.0.4-py3-none-any.whl",
"has_sig": false,
"md5_digest": "71e152d0b4a6bf04c546a9b2c1db0d6c",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": ">=3.11",
"size": 3473,
"upload_time": "2023-07-15T18:41:35",
"upload_time_iso_8601": "2023-07-15T18:41:35.297509Z",
"url": "https://files.pythonhosted.org/packages/16/a3/9c6424da97148b2f2faf78c0284771fe7abb9a52ebba2057ec6be9f17bba/medisynth-0.0.4-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "c91647de617593bff1bebf53a3c566891ec7c9ea22867191203343a5e787704d",
"md5": "8e1b3e4568f5e822167f4fb3c55e1342",
"sha256": "4a169287a354eeb0243310e50e95136d6d90b05f4e9cf497d3b09bc5d4229fd6"
},
"downloads": -1,
"filename": "medisynth-0.0.4.tar.gz",
"has_sig": false,
"md5_digest": "8e1b3e4568f5e822167f4fb3c55e1342",
"packagetype": "sdist",
"python_version": "source",
"requires_python": ">=3.11",
"size": 3723,
"upload_time": "2023-07-15T18:41:36",
"upload_time_iso_8601": "2023-07-15T18:41:36.654084Z",
"url": "https://files.pythonhosted.org/packages/c9/16/47de617593bff1bebf53a3c566891ec7c9ea22867191203343a5e787704d/medisynth-0.0.4.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2023-07-15 18:41:36",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "budai4medtech",
"github_project": "medisynth",
"travis_ci": false,
"coveralls": false,
"github_actions": true,
"lcname": "medisynth"
}
