Detection of Multi Stage Diabetes Foot Ulcer using Deep Learning Techniques

The detection of diabetes foot ulcers (DFUs) is crucial for timely intervention and prevention of severe complications in diabetic patients. One common effect of diabetes is diabetic foot ulcers that occurs when a wound or sore develops on the foot of a person with diabetes. It is typically caused by a various factor, including inadequate blood circulation, nerve injury (neuropathy), and impaired immunological function. Deep learning techniques have emerged as promising tools for automating DFU detection, offering advantages in accuracy and efficiency over traditional methods. This paper presents a unique deep learning approach for DFU detection using convolutional neural networks (CNNs) and image processing techniques. The proposed system utilizes a dataset of (DFUC 2021) foot images acquired through various imaging modalities, including infrared, thermal, and color imaging, to train the CNN model. The trained model demonstrates high accuracy in identifying DFUs by classifying them and analyzing key visual features such as wound size, wound type, color, and texture. Evaluation results on multiclass (3 class) classification of a diverse dataset of diabetic foot images (DFUC 2021) provide encouraging performance measures, such as a high F1 score, accuracy, and precision, indicating how well the suggested method works for DFU detection. The proposed system indicates that the DenseNet201 attains the greatest F 1 score of 98%, 98% and 97% for Ischemia, None and infection stage, respectively. Overall, the integration of deep learning techniques into DFU detection holds significant potential for advancing diabetic care as well as lowering the chance of problems.