Data

Tools

Indexes

Applications

Experiments

Open Source Science

https://github.com/berenslab/retinal-vessel-segmentation-benchmark

https://github.com/berenslab/retinal-vessel-segmentation-benchmark

Science Score: 36.0%

This score indicates how likely this project is to be science-related based on various indicators:

  • ○
    CITATION.cff file
  • ✓
    codemeta.json file
    Found codemeta.json file
  • ○
    .zenodo.json file
  • ✓
    DOI references
    Found 2 DOI reference(s) in README
  • ✓
    Academic publication links
    Links to: arxiv.org, ieee.org
  • ○
    Academic email domains
  • ○
    Institutional organization owner
  • ○
    JOSS paper metadata
  • ○
    Scientific vocabulary similarity
    Low similarity (9.0%) to scientific vocabulary
Last synced: 9 months ago · JSON representation

Repository

Basic Info
  • Host: GitHub
  • Owner: berenslab
  • Language: Jupyter Notebook
  • Default Branch: main
  • Size: 185 KB
Statistics
  • Stars: 0
  • Watchers: 3
  • Forks: 0
  • Open Issues: 0
  • Releases: 0
Created about 2 years ago · Last pushed about 2 years ago
Metadata Files
Readme

README.md

Retinal Vessel Segmenation Model Benchmark

Code associated with paper: Benchmarking *Retinal Blood Vessel Segmentation Models for** Cross-Dataset and Cross-Disease Generalization*.

This is a benchmark with the aim to investigate the performance of various deep learning model for retinal vessel segmentation It implements 5 backbone models on 3 datasets and compare the performances for different loss functions, image qualities and pathological conditions as well as the cross dataset generalization capabilities of the models.

We are actively expanding the benchmark to include more models, UQ methods, and datasets.

Backbones

| Backbone Models | Paper | Official Repo | | ----------- | ----------- | ----------- | | UNet |link | | | FR-UNet | link | link| | MA-Net| link | | | SA-UNet | link | link | | W-Net | link | link |

Data

  • FIVES
  • CHASEDB1
  • DRIVE

Environment Setup

Our code is developed with Python 3.10, The required packages are listed in requirements.txt and the environment can be created by running: bash pip install -r requirements.txt in your created environment.

Specify Arguments using .yaml Files

Specify the arguments in any of the files in the config folder with the respective model and loss function

Training

To run the code for training the baseline models:

```bash

python train_baseline.py --config config/fives.yaml ``` This example is for fives dataset with the models specified inside.

To train the 3vs1 subgroup setting. It requires an additional argument with the train_disease.py file models on a particular dis:

```bash

python train_disease.py --config config/fives.yaml --disease D `` This example is for fives dataset with the models specified inside and to train on other group of disease except DDiabetic-Retinppathy`

Owner

  • Name: Berens Lab @ University of Tübingen
  • Login: berenslab
  • Kind: organization
  • Email: philipp.berens@uni-tuebingen.de
  • Location: Tübingen, Germany

Department of Data Science at the Hertie Institute for AI in Brain Health, University of Tübingen

GitHub Events

Total
  • Watch event: 3
  • Fork event: 2
Last Year
  • Watch event: 3
  • Fork event: 2

Dependencies

requirements.txt pypi
  • Jinja2 ==3.1.2
  • Pillow ==9.5.0
  • albumentations ==1.3.0
  • bunch ==1.0.1
  • chardet ==5.1.0
  • efficientnet-pytorch ==0.7.1
  • einops ==0.6.1
  • imageio ==2.31.0
  • ipykernel ==6.23.1
  • ipython ==8.12.2
  • ipython-genutils ==0.2.0
  • ipywidgets ==8.0.6
  • isoduration ==20.11.0
  • jedi ==0.18.2
  • joblib ==1.2.0
  • json5 ==0.9.14
  • jsonpointer ==2.3
  • jsonschema ==4.17.3
  • jupyter ==1.0.0
  • jupyter-console ==6.6.3
  • jupyter-events ==0.6.3
  • jupyter-lsp ==2.2.0
  • jupyter_client ==8.2.0
  • jupyter_core ==5.3.0
  • jupyter_server ==2.6.0
  • jupyter_server_terminals ==0.4.4
  • jupyterlab ==4.0.2
  • jupyterlab-pygments ==0.2.2
  • jupyterlab-widgets ==3.0.7
  • jupyterlab_server ==2.23.0
  • kiwisolver ==1.4.4
  • loguru ==0.7.0
  • matplotlib ==3.7.1
  • matplotlib-inline ==0.1.6
  • numpy ==1.24.3
  • opencv-python-headless ==4.7.0.72
  • ruamel.base ==1.0.0
  • ruamel.yaml ==0.17.32
  • scikit-learn ==1.2.2
  • segmentation-models-pytorch ==0.3.3
  • tensorboard ==2.13.0
  • timm ==0.9.2
  • torch ==2.0.1
  • torchstat ==0.0.7
  • torchvision ==0.15.2
  • tqdm ==4.61.1
  • ttach ==0.0.3
  • yacs ==0.1.8