A Deep Learning Model-Based Neural Network Framework for Diabetic Foot Ulcer Classification

Diabetic Foot Ulcer (DFU) is a severe complication of diabetes, frequently resulting in amputation and having a significant impact on patients' quality of life. Early and accurate DFU classification is essential for prompt and efficient therapy. This study introduces an innovative neural network framework that utilizes deep learning techniques to categorize DFU based on thermograms. The proposed framework employs the pre-trained EfficientNet-B0 deep learning model, well-known for its exceptional performance in image classification tasks. The framework underwent training and assessment using a dataset sourced from Kaggle, consisting of 1866 colored images. Among them, 890 images depict plantar thermograms with negative DFU, whereas 976 images depict foot thermograms with positive DFU. The proposed framework demonstrated a higher accuracy of 98.71, outperforming previous approaches by a wide margin. The findings demonstrate the ability of deep learning to categorize DFU, paving the way for the development of automated testing techniques to assist healthcare practitioners in the timely identification and treatment of this incapacitating ailment.