Modelling a Deep Network Model for Diabetic Foot Ulcer Prediction Using Learning Approaches

In today's world, among 11 adults, one adult experiences diabetes mellitus and a complex illness known as diabetic foot ulcers (DFU).DFU needs to be treated well; otherwise, it may lead to amputation.The clinician performs the DFU treatment, where these treatments show remarkable restrictions, like costly diagnosis and lengthy care of DFU and treatment.Thus, there is a need for a novel decision-making technique.Constructing the dataset and collecting foot images from various patients are time-consuming processes.After the dataset acquisition, the skin conditions must be evaluated using computer vision algorithms.Here, novel learning techniques obtain the DFU features and the skin patches, which are healthy for understanding the difference in computer vision perspective.Further, the theoretical convolutional neural network architecture, CNN-DFUNet, is proposed to learn the feature representation to find the difference among the features and enhance the prediction accuracy.The CNN-DFUNet achieves 0.961 as the AUC score and is better than the conventional learning approaches.Furthermore, the proposed model is highly sensitive to detecting the presence of DFUs.Moreover, it is used for delivering the paradigm shift potentially among patients in diabetic foot care with less cost and reliable solutions in healthcare.