Get our free extension to see links to code for papers anywhere online!

 Add to Chrome

 Add to Firefox

CatalyzeX Code Finder - Browser extension linking code for ML papers across the web! | Product Hunt Embed

Deep Learning in Diabetic Foot Ulcers Detection: A Comprehensive Evaluation

Oct 07, 2020
Moi Hoon Yap, Ryo Hachiuma, Azadeh Alavi, Raphael Brungel, Manu Goyal, Hongtao Zhu, Bill Cassidy, Johannes Ruckert, Moshe Olshansky, Xiao Huang, Hideo Saito, Saeed Hassanpour, Christoph M. Friedrich, David Ascher, Anping Song, Hiroki Kajita, David Gillespie, Neil D. Reeves, Joseph Pappachan, Claire O'Shea, Eibe Frank

There has been a substantial amount of research on 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. With recent development and data sharing performed as part of the DFU Challenge (DFUC2020) such a comparison becomes possible: DFUC2020 provided participants with a comprehensive dataset consisting of 2,000 images for training each method and 2,000 images for testing them. The following deep learning-based algorithms are compared in this paper: Faster R-CNN; three variants of Faster R-CNN; an ensemble consisting of four models obtained using Faster R-CNN approaches; YOLOv3; YOLOv5; EfficientDet; and a new Cascade Attention Network. These deep learning methods achieved the best top 5 results in DFUC2020. For each deep learning method, this paper provides a detailed description of obtaining the model, including pre-processing, data augmentation, architecture, training and post-processing, and parameter settings. 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 is obtained by Deformable Convolution, a variant of Faster R-CNN, with a mAP of 0.6940 and an F1-Score of 0.7434. Our results show that state-of-the-art deep learning methods can detect DFU with some accuracy, but there are many challenges ahead before they can be implemented in real world settings.

* 17 pages, 17 figures, 9 tables 

Share this with someone who'll enjoy it:

   Access Paper Source

Share this with someone who'll enjoy it: