Deep Learning in Diabetic Foot Ulcers Detection: A Comprehensive Evaluation

TL;DR

This paper systematically compares various deep learning object detection models for diabetic foot ulcers, highlighting the best performing method and the benefits of ensemble approaches in improving detection metrics.

Contribution

It provides a comprehensive evaluation of state-of-the-art deep learning models for DFU detection, including detailed architecture descriptions and performance analysis on a large dataset.

Findings

Deformable Convolution (Faster R-CNN variant) achieved the highest mAP of 0.6940.

Data augmentation and post-processing are essential for improving model performance.

Ensemble methods improved F1-Score but not mAP.

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 summarises 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 enhanced the F1-Score but not the mAP.