Translating Clinical Delineation of Diabetic Foot Ulcers into Machine Interpretable Segmentation
This work addresses the need for better automated monitoring of diabetic foot ulcers, but it is incremental as it focuses on dataset creation and benchmarking existing methods.
The authors tackled the problem of automatically segmenting diabetic foot ulcers by creating the largest clinical dataset (DFUC2022) and evaluating deep learning methods, finding that using image-processed refined contours as ground truth improved machine agreement, with the best model achieving a Dice score of 0.6277.
Diabetic foot ulcer is a severe condition that requires close monitoring and management. For training machine learning methods to auto-delineate the ulcer, clinical staff must provide ground truth annotations. In this paper, we propose a new diabetic foot ulcers dataset, namely DFUC2022, the largest segmentation dataset where ulcer regions were manually delineated by clinicians. We assess whether the clinical delineations are machine interpretable by deep learning networks or if image processing refined contour should be used. By providing benchmark results using a selection of popular deep learning algorithms, we draw new insights into the limitations of DFU wound delineation and report on the associated issues. This paper provides some observations on baseline models to facilitate DFUC2022 Challenge in conjunction with MICCAI 2022. The leaderboard will be ranked by Dice score, where the best FCN-based method is 0.5708 and DeepLabv3+ achieved the best score of 0.6277. This paper demonstrates that image processing using refined contour as ground truth can provide better agreement with machine predicted results. DFUC2022 will be released on the 27th April 2022.