Development of a Computer Vision System for Diabetic Foot Detection Using Deep Learning Techniques

The use of technology is very important in medical diagnosis because of its great impact on the accuracy of diagnosis. One of these diseases is diabetic foot, which is in the form of ulcers in the feet of diabetic patients. The earlier it is detected, the better it is in treating it. Therefore, this article presents a method for adopting deep learning techniques to detect these ulcers in the feet of diabetic patients. The YOLOV7 deep learning model was used in more than one version in the process of detecting these ulcers in the feet of diabetic patients. The model was trained based on a data set consisting of 1438 images divided between a part allocated 80% for training and the other part allocated for testing. A set of criteria was relied upon to measure the efficiency of the method used, which are: recall, accuracy, and F1score.