What question did this study set out to answer?

This research aims to compare the effectiveness of CNN and transformer architectures in diagnosing diabetic foot ulcers.

February 9, 2026Open Access

Deep Learning Strategies for Diabetic Foot Ulcer Diagnosis: An Empirical Comparison of Transformer and CNN Architectures

SRSakshi RSt. Joseph's Institute of Technology STS ThirilokshaSt. Joseph's Institute of Technology STS. TephillahSt. Joseph's Institute of Technology

Puntos clave

This research aims to compare the effectiveness of CNN and transformer architectures in diagnosing diabetic foot ulcers.
Utilized ResNet18 (CNN) and DeiT-Small (Vision Transformer) models for diagnosis
Employed transfer learning on images of DFU-affected and healthy feet
Used visualization tools like Grad-CAM and transformer attention maps for analysis
DeiT-Small achieved a test accuracy of 99.17%
ResNet18 achieved an accuracy of 85.40%
A significant performance gap was observed between the two models

Resumen

Diabetic Foot Ulcers (DFU) are among the most serious complications faced by diabetic patients. What usually starts as a small injury can worsen and in many cases lead to lower limb amputation. However with modern technologies like deep learning early screening and identification of DFU has become more achievable. This helps doctors and make their job easy. This study presents a practical comparison between ResNet18 a CNN and DeiT-Small a Vision Transformer. Both these models were tuned using transfer learning on datasets containing images of both DFU affected feet and healthy feet. There was a clear performance gap between both these models. DeiT-Small reached an impressive test accuracy of 99.17% while ResNet18 achieved 85.40% accuracy. To understand their performance better visualization tools like Grad-CAM and transformer attention maps were used. For future work, the study aims to create more consistent data splits, perform deeper architectural ablation and integrate IOT based sensors to support real time DFU monitoring.

Preguntar a la IA

Me gusta

Guardar

Ver artículo completo