2021/11/12 by Adrián Galdrán, Galdran, Adrian, Gustavo Carneiro +3 · 1 citation
Computer Science · Dentistry · Medicine · #Computer Vision and Pattern Recognition (cs.CV) #Diabetic Foot Ulcer Assessment and Management #Digital Imaging for Blood Diseases #FOS: Computer and information sciences #FOS: Electrical engineering #Image and Video Processing (eess.IV) #Oral microbiology and periodontitis research #electronic engineering #information engineering
paper · pdf · doi:10.48550/arxiv.2111.06894
openalex publication_date 2021/11/12 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
This paper compares well-established Convolutional Neural Networks (CNNs) to\nrecently introduced Vision Transformers for the task of Diabetic Foot Ulcer\nClassification, in the context of the DFUC 2021 Grand-Challenge, in which this\nwork attained the first position. Comprehensive experiments demonstrate that\nmodern CNNs are still capable of outperforming Transformers in a low-data\nregime, likely owing to their ability for better exploiting spatial\ncorrelations. In addition, we empirically demonstrate that the recent\nSharpness-Aware Minimization (SAM) optimization algorithm considerably improves\nthe generalization capability of both kinds of models. Our results demonstrate\nthat for this task, the combination of CNNs and the SAM optimization process\nresults in superior performance than any other of the considered approaches.\n