Leveraging Artificial Intelligence in the Diagnosis and Management of Diabetic Foot Ulcers: A Review of Current Trends and Future Directions

Azab, Mai Essam; Amer, Hanan. M.; Mohammwd, Warda; ElSeddek, Mervat

doi:10.21608/ijt.2025.356528.1081

Leveraging Artificial Intelligence in the Diagnosis and Management of Diabetic Foot Ulcers: A Review of Current Trends and Future Directions"

Document Type : Original Article

Authors

¹ Department of Electronics and Communications Engineering, Faculty of Engineering, MISR Higher Institute for Engineering and Technology, Egypt

² Electronics and Communications Engineering Department - Faculty of Engineering – Mansoura University – Mansoura – Egypt

³ Department of communication and electronics engineering, Nile Higher Institute for Engineering and Technology, Egypt

⁴ Department of Electronics and Communications Engineering, Faculty of Engineering Horus University

10.21608/ijt.2025.356528.1081

Abstract

Diabetic foot ulcers (DFUs) and other related consequences of diabetes mellitus are major health challenges on a global scale. Diabetic foot ulcers (DFUs) and other severe side effects may be prevented with early detection. One serious condition that might result in a diabetic patient's lower limb being amputated is a DFU. For physicians, diagnosing DFU can be difficult because it often necessitates a variety of costly and time-consuming clinical examinations. Clinical professionals may now diagnose patients more quickly and accurately thanks to the application of machine learning, deep learning, and computer vision techniques in the age of data overload. Among the many advantages of using machine learning and deep learning for DFU detection are its ability to learn more features, versatility across several image modalities, with the ability for high task accuracy in detection and identification.
Giving academics a thorough overview of the state of automatic DFU identification tasks was the article's main goal. The utilization of both machine learning and advanced deep learning algorithms is required to assist clinicians in making quicker and more accurate diagnoses, according to several observations obtained from previous research. In conventional machine learning techniques, image features aid in precise identification and offer significant data on DFU. However, advanced deep learning techniques have shown greater promise than machine learning techniques in certain earlier studies. The problem domain has been controlled by the CNN-based solutions presented out by several authors.

Keywords