Deep learning in diabetic foot ulcers detection: A comprehensive evaluation. (August 2021)

• Record Type: Journal Article
• Authors: Yap, Moi Hoon; Hachiuma, Ryo; Alavi, Azadeh; Brüngel, Raphael; Cassidy, Bill; Goyal, Manu; Zhu, Hongtao; Rückert, Johannes; Olshansky, Moshe; Huang, Xiao; Saito, Hideo; Hassanpour, Saeed; Friedrich, Christoph M.; Ascher, David B.; Song, Anping; Kajita, Hiroki; Gillespie, David; Reeves, Neil D.; Pappachan, Joseph M.; O'Shea, Claire; Frank, Eibe
• Is Part Of: Computers in biology and medicine. Volume 135(2021)
• Publication Date: 2021-08
• Subjects: Diabetic foot ulcers -- Object detection -- Machine learning -- Deep learning -- DFUC2020; Medicine -- Data processing -- Periodicals; Biology -- Data processing -- Periodicals; 610.285
• DOI: 10.1016/j.compbiomed.2021.104596
• Languages: English
• ISSNs: 0010-4825

Abstract

Abstract: There has been a substantial amount of research involving computer methods and technology for the detection and recognition of diabetic foot ulcers (DFUs), but there is a lack of systematic comparisons of state-of-the-art deep learning object detection frameworks applied to this problem. DFUC2020 provided participants with a comprehensive dataset consisting of 2, 000 images for training and 2, 000 images for testing. This paper summarizes the results of DFUC2020 by comparing the deep learning-based algorithms proposed by the winning teams: Faster R–CNN, three variants of Faster R–CNN and an ensemble method; YOLOv3; YOLOv5; EfficientDet; and a new Cascade Attention Network. For each deep learning method, we provide a detailed description of model architecture, parameter settings for training and additional stages including pre-processing, data augmentation and post-processing. We provide a comprehensive evaluation for each method. All the methods required a data augmentation stage to increase the number of images available for training and a post-processing stage to remove false positives. The best performance was obtained from Deformable Convolution, a variant of Faster R–CNN, with a mean average precision (mAP) of 0.6940 and an F1-Score of 0.7434. Finally, we demonstrate that the ensemble method based on different deep learning methods can enhance the F1-Score but not the mAP. Highlights: We evaluate popular deep learning methods on diabetic foot ulcer (DFU) detection. A new Cascade Attention Network (CA-DetNet) is proposed for DFU detection. Ensemble methods are proposed and evaluated for DFU detection. Faster R–CNN with deformable convolution achieved the best mAP on the DFUC2020 dataset. Ensemble method achieved the best F1-Score in DFU detection on the DFUC2020 dataset.