Data

Tools

Indexes

Applications

Experiments

Open Source Science

evonms

https://github.com/neuluna/evonms

Science Score: 31.0%

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

  • ✓
    CITATION.cff file
    Found CITATION.cff file
  • ○
    codemeta.json file
  • ○
    .zenodo.json file
  • ✓
    DOI references
    Found 2 DOI reference(s) in README
  • ○
    Academic publication links
  • ○
    Committers with academic emails
  • ○
    Institutional organization owner
  • ○
    JOSS paper metadata
  • ○
    Scientific vocabulary similarity
    Low similarity (11.0%) to scientific vocabulary

Keywords

evolutionary-algorithm normalization semantic-segmentation
Last synced: 6 months ago · JSON representation ·

Repository

Basic Info
Statistics
  • Stars: 0
  • Watchers: 1
  • Forks: 0
  • Open Issues: 0
  • Releases: 0
Topics
evolutionary-algorithm normalization semantic-segmentation
Created over 2 years ago · Last pushed over 2 years ago
Metadata Files
Readme Changelog License Citation

README.md

Evolutionary Normalization Optimization Boosts Semantic Segmentation Network Performance

This repository contains the source code for the project to determine how layer-specific normalization methods can influence the segmentation result of an U-Net by using an evolutionary algorithm approach.

Structure of the Repository

data_generator.py

Loads the dataset and normalizes the images/masks to 0 to 1 if it is a binary task. If it is a multi-class segmentation, the masks contain the class labels from 0 to num_classes. Returns the dataset split into the tensors train, val and test for further processing.

main.py

  • main(): runs the U-Net and the evaluation of the different individuals of a generation over multiple generations and sorting the models for the first and the last generation
  • select(): Runs the selection and breeding of the new population

metrics.py

  • dice_coefficient(): calculates the Dice Similarity Coefficient (DSC)
  • dice_coef_loss(): calculates the loss based on the DSC
  • get_flat(): flattens the predicted masks
  • draw_bb(): draws a minimal rectangle based on the predictions
  • get_bb(): calls the draw_bb() function
  • bb_IoU(): calculates the IoU score of the predicted bounding boxes
  • IoU(): calculates the Intersection over Union score
  • hd_95_monai(): calculates the Hausdorff Distance 95

unet.py

  • conv_layer(): builds one layer of the U-Net with the given settings and can be variable including normalization, activation and filter size.
  • unet(): builds the U-Net architecture by using 4 Layers for encoding and Decoding by using a normal up-sampling.

Util Files

To install this project, follow these steps:

  • constants.json: defines all genes which can be chosen to build one U-Net individual
  • datasets.json: gives an overview of the datasets and its structure
  • env.yml: yaml-file to create an environment to run the code

License

This repository is licensed under the terms of the MIT License.

Citation

Please site the usage of the software as follows:

Neubig, L., Kist, A.M. (2023). Evolutionary Normalization Optimization Boosts Semantic Segmentation Network Performance. In: Greenspan, H., et al. Medical Image Computing and Computer Assisted Intervention – MICCAI 2023. MICCAI 2023. Lecture Notes in Computer Science, vol 14223. Springer, Cham. <https://doi.org/10.1007/978-3-031-43901-8_67>

Owner

  • Login: neuluna
  • Kind: user

Citation (CITATION.bib) 

@incollection{Neubig2023,
  doi = {10.1007/978-3-031-43901-8_67},
  url = {https://doi.org/10.1007/978-3-031-43901-8_67},
  year = {2023},
  publisher = {Springer Nature Switzerland},
  pages = {703--712},
  author = {Luisa Neubig and Andreas M. Kist},
  title = {Evolutionary Normalization Optimization Boosts Semantic Segmentation Network Performance},
  booktitle = {Lecture Notes in Computer Science}
}

GitHub Events

Total
Last Year

Committers

Last synced: about 2 years ago

All Time
  • Total Commits: 1
  • Total Committers: 1
  • Avg Commits per committer: 1.0
  • Development Distribution Score (DDS): 0.0
Past Year
  • Commits: 1
  • Committers: 1
  • Avg Commits per committer: 1.0
  • Development Distribution Score (DDS): 0.0
Top Committers
Name Email Commits
neuluna 7****a 1

Issues and Pull Requests

Last synced: about 2 years ago

All Time
  • Total issues: 0
  • Total pull requests: 0
  • Average time to close issues: N/A
  • Average time to close pull requests: N/A
  • Total issue authors: 0
  • Total pull request authors: 0
  • Average comments per issue: 0
  • Average comments per pull request: 0
  • Merged pull requests: 0
  • Bot issues: 0
  • Bot pull requests: 0
Past Year
  • Issues: 0
  • Pull requests: 0
  • Average time to close issues: N/A
  • Average time to close pull requests: N/A
  • Issue authors: 0
  • Pull request authors: 0
  • Average comments per issue: 0
  • Average comments per pull request: 0
  • Merged pull requests: 0
  • Bot issues: 0
  • Bot pull requests: 0
View more stats
Top Authors
Issue Authors
Pull Request Authors
Top Labels
Issue Labels
Pull Request Labels