Revolutionizing Diabetic Foot Ulcer Care through Intelligent CNN-Based Predictions

Diabetic foot ulcers are a serious consequence of diabetes, affecting about 25% of patients with dire consequences that range from infection and amputation to a decrease in the quality of life. Diabetes mellitus is a major health risk throughout the world, as are its associated issues, including DFUs. Individuals with DFUs have a better chance of recovery if diagnosed as soon as possible. Perhaps machine learning can be used to provide clinical professionals with ideas for identifying and treating patients' conditions. Early detection is considered an important requirement for timely treatment to prevent serious complications while reducing healthcare costs. This paper concerns a new, non-invasive method for the detection of DFUs based on advanced convolutional neural networks (CNNs) and a webcam-based system. It achieves a high accuracy of 97.2%, precision of 96.8%, and recall of 97.5%, ensuring reliable and early ulcer identification. The results demonstrate the efficacy of the proposed CNN-based model in predicting diabetic foot ulcers accurately across three stages, namely healthy, early-stage, and severe, thereby having a great potential for early detection and effective medical intervention.