diff --git a/README.md b/README.md index df80b7e..dc2cfd8 100644 --- a/README.md +++ b/README.md @@ -45,7 +45,7 @@ Contribute your own generative model to `medigan` to increase its visibility, re | Breast Mass | mammography | dcgan | 128x128 | [BCDR](https://bcdr.eu/information/about) | ![sample](docs/source/_static/samples/00005.png) | [`00005_DCGAN_MMG_MASS_ROI`](https://medigan.readthedocs.io/en/latest/models.html#id1) | [Zenodo (6555188)](https://doi.org/10.5281/zenodo.6555188) | [Szafranowska et al (2022)](https://doi.org/10.48550/arXiv.2203.04961) | | Breast Mass | mammography | wgan-gp | 128x128 | [BCDR](https://bcdr.eu/information/about) | ![sample](docs/source/_static/samples/00006.png) | [`00006_WGANGP_MMG_MASS_ROI`](https://medigan.readthedocs.io/en/latest/models.html#wgangp-mmg-mass-roi) | [Zenodo (6554713)](https://doi.org/10.5281/zenodo.6554713) | [Szafranowska et al (2022)](https://doi.org/10.48550/arXiv.2203.04961) | | Brain Tumors on Flair, T1, T1c, T2 with Masks | brain MRI | inpaint GAN | 256x256 | [BRATS 2018](https://wiki.cancerimagingarchive.net/pages/viewpage.action?pageId=37224922) | ![sample](docs/source/_static/samples/00007_F.png)
![sample](docs/source/_static/samples/00007_T1.png)
![sample](docs/source/_static/samples/00007_T1c.png)
![sample](docs/source/_static/samples/00007_T2.png)
![sample](docs/source/_static/samples/00007_mask.png)
![sample](docs/source/_static/samples/00007_grade_mask.png) | [`00007_INPAINT_BRAIN_MRI`](https://medigan.readthedocs.io/en/latest/models.html#inpaint-brain-mri) | [Zenodo (7041737)](https://doi.org/10.5281/zenodo.7041737) | [Kim et al (2020)](https://doi.org/10.1002/mp.14701) | -| Breast Mass (Mal/Benign) | mammography | c-dcgan | 128x128 | [CBIS-DDSM](https://wiki.cancerimagingarchive.net/display/Public/CBIS-DDSM) | ![sample](docs/source/_static/samples/00008.png) | [`00008_C-DCGAN_MMG_MASSES`](https://medigan.readthedocs.io/en/latest/models.html#c-dcgan-mmg-masses) | [Zenodo (6647349)](https://doi.org/10.5281/zenodo.6647349) | | +| Breast Mass (Mal/Benign) | mammography | c-dcgan | 128x128 | [CBIS-DDSM](https://wiki.cancerimagingarchive.net/display/Public/CBIS-DDSM) | ![sample](docs/source/_static/samples/00008.png) | [`00008_C-DCGAN_MMG_MASSES`](https://medigan.readthedocs.io/en/latest/models.html#c-dcgan-mmg-masses) | [Zenodo (6647349)](https://doi.org/10.5281/zenodo.6647349) | [Osuala et al (2024)](https://doi.org/10.48550/arXiv.2407.12669) | | Polyp with Mask | endoscopy | pggan | 256x256 | [HyperKvasir](https://osf.io/mh9sj/) | ![sample](docs/source/_static/samples/00009.png)
![sample](docs/source/_static/samples/00009_mask.png) | [`00009_PGGAN_POLYP_PATCHES_W_MASKS`](https://medigan.readthedocs.io/en/latest/models.html#pggan-polyp-patches-w-masks) | [Zenodo (6653743)](https://doi.org/10.5281/zenodo.6653743) | [Thambawita et al (2022)](https://doi.org/10.1371/journal.pone.0267976) | | Polyp with Mask | endoscopy | fastgan | 256x256 | [HyperKvasir](https://osf.io/mh9sj/) | ![sample](docs/source/_static/samples/00010.png)
![sample](docs/source/_static/samples/00010_mask.png) | [`00010_FASTGAN_POLYP_PATCHES_W_MASKS`](https://medigan.readthedocs.io/en/latest/models.html#fastgan-polyp-patches-w-masks) | [Zenodo (6660711)](https://doi.org/10.5281/zenodo.6660711) | [Thambawita et al (2022)](https://doi.org/10.1371/journal.pone.0267976) | | Polyp with Mask | endoscopy | singan | ≈250x250 | [HyperKvasir](https://osf.io/mh9sj/) | ![sample](docs/source/_static/samples/00011.png)
![sample](docs/source/_static/samples/00011_mask.png) | [`00011_SINGAN_POLYP_PATCHES_W_MASKS`](https://medigan.readthedocs.io/en/latest/models.html#singan-polyp-patches-w-masks) | [Zenodo (6667944)](https://doi.org/10.5281/zenodo.6667944) | [Thambawita et al (2022)](https://doi.org/10.1371/journal.pone.0267976) | @@ -85,7 +85,7 @@ Documentation is available at [medigan.readthedocs.io](https://medigan.readthedo ### Generation example #### DCGAN -Create mammography masses with labels (malignant or benign) using a class-conditional DCGAN model. +Create mammography masses with labels (malignant or benign) using a [class-conditional DCGAN model](https://arxiv.org/abs/2407.12669). ```python # import medigan and initialize Generators from medigan import Generators @@ -98,7 +98,7 @@ generators.generate(model_id=8, num_samples=8, install_dependencies=True) ![sample](docs/source/_static/samples/c-dcgan/model8_samples.png) The synthetic images in the top row show malignant masses (breast cancer) while the images in the bottom row show benign masses. -Given such images with class information, image classification models can be (pre-)trained. +Given such images with class information, [image classification models](https://arxiv.org/abs/2407.12669) can be (pre-)trained. #### CYCLEGAN