Skip to content

Yussef93/FewShotCellSegmentation

Repository files navigation

FewShotCellSegmentation

Automatic cell segmentation in microscopy images works well with the support of deep neural networks trained with full supervision. Collecting and annotating images, though, is not a sustainable solution for every new microscopy database and cell type. Instead, we assume that we can access a plethora of annotated image data sets from different domains (sources) and a limited number of annotated image data sets from the domain of interest (target), where each domain denotes not only different image appearance but also a different type of cell segmentation problem. We pose this problem as meta-learning where the goal is to learn a generic and adaptable few-shot learning model from the available source domain data sets and cell segmentation tasks. The model can be afterwards fine-tuned on the few annotated images of the target domain that contains different image appearance and different cell type. In our meta-learning training, we propose the combination of three objective functions to segment the cells, move the segmentation results away from the classification boundary using cross-domain tasks, and learn an invariant representation between tasks of the source domains. Our experiments on five public databases show promising results from 1- to 10-shot meta-learning using standard segmentation neural network architectures.

Link to full paper https://arxiv.org/abs/2007.01671

Algorithm

Screenshot 2020-07-06 at 13 12 29

Results

Screenshot 2020-07-06 at 13 00 37

Screenshot 2020-07-06 at 13 06 23

Screenshot 2020-07-06 at 13 06 41

Code

1- Install necessary python modules in requirements.txt

2- Run run_preprocessing.py i.e. python run_preprocessing.py to download the datasets and preprocess them, in addition to extracting and preprocessing my 10 random selections.

3- Instructions to run training and evaluation are available with examples in Learning_main.py and Evaluation_main.py

Pre-trained Models

The Pre-trained models can be downloaded from this link https://cloudstore.uni-ulm.de/s/YqD6or4DLyjF7ry

License

This project is licensed under the MIT license - see the License.md file for details

Cite

To cite this repository, please use the following citation:

@inproceedings{DBLP:conf/pkdd/DawoudHCB20,
  author    = {Youssef Dawoud and
               Julia Hornauer and
               Gustavo Carneiro and
               Vasileios Belagiannis},
  title     = {Few-Shot Microscopy Image Cell Segmentation},
  booktitle = {{ECML/PKDD} {(5)}},
  series    = {Lecture Notes in Computer Science},
  volume    = {12461},
  pages     = {139--154},
  publisher = {Springer},
  year      = {2020}
} 

About

Code of "Few-shot microscopy image cell segmentation " https://link.springer.com/chapter/10.1007/978-3-030-67670-4_9

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published