Comparative Analysis: Deep vs. Machine Learning for Early DFU Detection in Medical Imaging

Diabetes Mellitus (DM) is a metabolic disorder characterized by aberrant insulin secretion and function, resulting in elevated blood glucose levels and chronic damage to organs. Diabetic Foot Ulcers (DFU) are a significant complication of DM, frequently leading to lower limb amputation if not effectively managed. DFU imposes a substantial burden on patients, families, and healthcare systems, particularly in developing nations where treatment expenses can be overwhelming. Leveraging machine learning and deep learning techniques allows for accurate identification of foot ulcers in diabetic individuals, facilitating early detection and prompt treatment, consequently mitigating the risk of complications and improving the overall quality of life for DM patients. The study focuses on utilizing deep learning and machine learning models to classify diabetic foot ulcers. The results reveal that the RESNET18 model outperforms the SVM models, achieving an accuracy of 97.9%. The study suggests that deep learning models, notably the RESNET18, are more suitable for identifying diabetic foot ulcers.