AI-Powered CNN and U-Net Model for Diabetic Foot Ulcer Detection on DFU Dataset

One serious consequence of diabetes is diabetic foot ulcers (DFUs), which can lead to infections and limb amputation. Better patient outcomes depend on early diagnosis. Convolutional Neural Networks (CNNs) are compared with CHS, SVM, KNN, RF, and DeSGEL for DFU classification in this study. The proposed CNN model outperforms existing methods with 98.5% accuracy, 96.5% F1-score, 96.2% recall, 95.2% precision, and a 1.5% error rate. Accuracy, precision, recall, F1-score, error rate, and training duration are used to gauge performance. Artificial intelligence-based techniques improve patient care efficiency by increasing diagnostic precision, reducing human error, and providing a low-cost, scalable solution for DFU diagnosis.