Deepdraw: Semantic segmentation library for medical imaging
Posted on Sun 01 January 2023 in software
Deepdraw is an open-source software package to train and evaluate a range of neural network architectures for semantic image segmentation tasks. It is built using PyTorch. It used to be called bob.ip.binseg when it was part of the Signal Processing and Machine Learning Library bob.
Example 1: Semantic segmentation output of deepdraw, where a fully-convolutional model was trained to delineate vessels on a color fundus image (ophthalmological data).
Associated Publications
If you use this work, cite at least use the BibTeX references below:
@inproceedings{renzo-2021,
title = {Development of a lung segmentation algorithm for analog imaged chest X-Ray: preliminary results},
author = {Matheus A. Renzo and Nat\'{a}lia Fernandez and Andr\'e Baceti and Natanael Nunes de Moura Junior and Andr\'e Anjos},
month = {10},
booktitle = {XV Brazilian Congress on Computational Intelligence},
year = {2021},
doi = {10.21528/CBIC2021-123},
}
@misc{laibacher-2019,
title = {On the Evaluation and Real-World Usage Scenarios of Deep Vessel Segmentation for Retinography},
author = {Tim Laibacher and Andr\'e Anjos},
year = {2019},
eprint = {1909.03856},
archivePrefix = {arXiv},
primaryClass = {cs.CV},
url = {https://arxiv.org/abs/1909.03856},
}