Diabetic foot ulcers (DFU) are one of the most frequent and devastating complications of diabetes, representing a considerable challenge for healthcare systems and significantly affecting patients’ quality of life. This study addresses the application of advanced artificial intelligence techniques for DFU detection in order to select best performed model. State of the art deep learning models were implemented and tested, specifically YOLOv8 in its m and l variants, and Faster R-CNN in its standard and deformable convolution variants. The models were trained with the images provided by the Diabetic Foot Ulcers Grand Challenge 2020 (DFUC 2020), which is a specialized and representative annotated dataset. Also, the models were optimized using data augmentation techniques and hyperparameter tuning to maximize their performance. Additionally, an external data set compose of 80 images annotated by expert’s angiologist was used to evaluate the best models effectiveness in the local context and generalization capabilities. A mAP greater than 0.70 was obtained for all trained models during the training and validation stages, being Yolov8l model the best performed (mAP = 0.799). The external test demonstrated generalization power, obtaining accuracy values of 0.856, sensitivity of 0.789, F1 score of 0.821, and a mAP of 0.839.