Diabetic foot ulcers are caused by neuropathic and vascular complications of diabetes mellitus. In order to provide a proper diagnosis and treatment, wound care professionals need to extract accurate morphological features from the foot wounds. Using computer-aided systems is a promising approach to extract related morphological features and segment the lesions. We propose a convolution neural network called HarDNet-DFUS by enhancing the backbone and replacing the decoder of HarDNet-MSEG, which was the state-of-the-art network for colonoscopy polyp segmentation in 2021. For the MICCAI 2022 Diabetic Foot Ulcer Segmentation Challenge (DFUC2022), we train HarDNet-DFUS using the DFUC2022 dataset and increase its robustness by means of five-fold cross validation and Test Time Augmentation. In the validation phase of DFUC2022, HarDNet-DFUS achieved 0.7063 mean Dice and was ranked third among all participants. In the final testing phase of DFUC2022, it achieved 0.7287 mean Dice and was the first place winner. The code is available on https://github.com/kytimmylai/DFUC2022.
CITATION STYLE
Liao, T. Y., Yang, C. H., Lo, Y. W., Lai, K. Y., Shen, P. H., & Lin, Y. L. (2023). HarDNet-DFUS: Enhancing Backbone and Decoder of HarDNet-MSEG for Diabetic Foot Ulcer Image Segmentation. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 13797 LNCS, pp. 21–30). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-031-26354-5_2
Mendeley helps you to discover research relevant for your work.