Spaces:
Sleeping
Sleeping
Commit
·
70788a3
1
Parent(s):
2aa6515
Update README.md
Browse files
README.md
CHANGED
|
@@ -10,4 +10,52 @@ pinned: false
|
|
| 10 |
license: mit
|
| 11 |
---
|
| 12 |
|
| 13 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 10 |
license: mit
|
| 11 |
---
|
| 12 |
|
| 13 |
+
# RetinaGAN
|
| 14 |
+
|
| 15 |
+
Code Repository for: [**High-Fidelity Diabetic Retina Fundus Image Synthesis from Freestyle Lesion Maps**](https://opg.optica.org/abstract.cfm?uri=boe-14-2-533)
|
| 16 |
+
|
| 17 |
+
## About
|
| 18 |
+
|
| 19 |
+
RetinaGAN a two-step process for generating photo-realistic retinal Fundus images based on artificially generated or free-hand drawn semantic lesion maps.
|
| 20 |
+
|
| 21 |
+
|
| 22 |
+
|
| 23 |
+
StyleGAN is modified to be conditional in to synthesize pathological lesion maps based on a specified DR grade (i.e., grades 0 to 4). The DR Grades are defined by the International Clinical Diabetic Retinopathy (ICDR) disease severity scale; no apparent retinopathy, {mild, moderate, severe} Non-Proliferative Diabetic Retinopathy (NPDR), and Proliferative Diabetic Retinopathy (PDR). The output of the network is a binary image with seven channels instead of class colors to avoid ambiguity.
|
| 24 |
+
|
| 25 |
+
|
| 26 |
+
|
| 27 |
+
The generated label maps are then passed through SPADE, an image-to-image translation network, to turn them into photo-realistic retina fundus images. The input to the network are one-hot encoded labels.
|
| 28 |
+
|
| 29 |
+
|
| 30 |
+
|
| 31 |
+
## Usage
|
| 32 |
+
|
| 33 |
+
Download model checkpoints (see [here](checkpoints/README.md) for details) and run the model via Streamlit. Start the app via `streamlit run web_demo.py`.
|
| 34 |
+
|
| 35 |
+
## Example Images
|
| 36 |
+
|
| 37 |
+
Example retina Fundus images synthesised from Conditional StyleGAN generated lesion maps. Top row: synthetically generated lesion maps based on DR grade by Conditional StyleGAN. Other rows: synthetic Fundus images generated by SPADE. Images are generated sequentially with random seed and are **not** cherry picked.
|
| 38 |
+
|
| 39 |
+
| grade 0 | grade 1 | grade 2 | grade 3 | grade 4 |
|
| 40 |
+
|--------------------------------------------------------------|--------------------------------------------------------------|--------------------------------------------------------------|--------------------------------------------------------------|--------------------------------------------------------------|
|
| 41 |
+
|  |  |  |  |  |
|
| 42 |
+
|  |  |  |  |  |
|
| 43 |
+
|  |  |  |  |  |
|
| 44 |
+
|
| 45 |
+
## Cite this work
|
| 46 |
+
|
| 47 |
+
If you find this work useful for your research, give us a kudos by citing:
|
| 48 |
+
|
| 49 |
+
```
|
| 50 |
+
@article{hou2023high,
|
| 51 |
+
title={High-fidelity diabetic retina fundus image synthesis from freestyle lesion maps},
|
| 52 |
+
author={Hou, Benjamin},
|
| 53 |
+
journal={Biomedical Optics Express},
|
| 54 |
+
volume={14},
|
| 55 |
+
number={2},
|
| 56 |
+
pages={533--549},
|
| 57 |
+
year={2023},
|
| 58 |
+
publisher={Optica Publishing Group}
|
| 59 |
+
}
|
| 60 |
+
```
|
| 61 |
+
|