Domain-Specific Deep Learning Feature Extractor for Diabetic Foot Ulcer Detection

Reza Basiri; Milos R. Popovic; Shehroz S. Khan

doi:10.1109/ICDMW58026.2022.00041

Research

Conference paper, November 2022

Domain-Specific Deep Learning Feature Extractor for Diabetic Foot Ulcer Detection

Reza Basiri, Milos R. Popovic, Shehroz S. Khan

2022 IEEE International Conference on Data Mining Workshops (ICDMW), 1-5. DOI 10.1109/ICDMW58026.2022.00041

In brief

Every wound detector is built on a feature extractor, the backbone that turns pixels into patterns. This study benchmarks 14 backbone and architecture combinations on the public DFU2020 dataset to find the best foundation for automated ulcer detection.

Key findings

14 networks compared on mAP and F1
UNet with an EfficientNet-B3 backbone performed best
The foundation for the group's later detection pipeline

Read at the publisher arXiv preprint

Themes

Clinical AI for wound care

Keywords

Cite

Basiri, R., Popovic, M. R., & Khan, S. S. (2022). Domain-Specific Deep Learning Feature Extractor for Diabetic Foot Ulcer Detection. 2022 IEEE International Conference on Data Mining Workshops (ICDMW), 1-5. https://doi.org/10.1109/ICDMW58026.2022.00041

BibTeX

@inproceedings{basiri2022domainspecific,
  title = {{Domain-Specific Deep Learning Feature Extractor for Diabetic Foot Ulcer Detection}},
  author = {Reza Basiri and Milos R. Popovic and Shehroz S. Khan},
  booktitle = {2022 IEEE International Conference on Data Mining Workshops (ICDMW)},
  year = {2022},
  pages = {1--5},
  doi = {10.1109/ICDMW58026.2022.00041}
}

Domain-Specific Deep Learning Feature Extractor for Diabetic Foot Ulcer Detection

In brief

Key findings

Related work

Temporal machine learning framework for diabetic foot ulcer healing trajectory prediction

Multimodal Artificial Intelligence System to Enhance the Classification and Analysis of Diabetic Foot Ulcer Healing Conditions

Accessible healing phase classification of diabetic foot ulcer