Glaucoma, a disease that damages the optic nerve, is the leading cause of irreversible blindness worldwide. The early detection of glaucoma is a challenge, which in recent years has driven the study and application of Deep Learning (DL) techniques in the automatic classification of eye fundus images. Among these intelligent systems, Convolutional Neural Networks (CNNs) stand out, although alternatives have recently appeared, such as Vision Transformers (ViTs) or hybrid systems, which are also highly efficient in image processing. The question that arises in the face of so many emerging methods is whether all these new techniques are really more efficient for the problem of glaucoma diagnosis than the CNNs that have been used so far. In this article, we present a comprehensive comparative study of all these DL models in glaucoma detection, with the aim of elucidating which strategies are significantly better. Our main conclusion is that there are no significant differences between the efficiency of both DL strategies for the medical diagnostic problem addressed.

Glaucoma, a leading cause of blindness, damages the optic nerve, making early diagnosis challenging due to no initial symptoms. Fundus eye images taken with a non-mydriatic retinograph help diagnose glaucoma by revealing structural changes, including the optic disc and cup. This research aims to thoroughly analyze saliency maps in interpreting convolutional neural network decisions for diagnosing glaucoma from fundus images. These maps highlight the most influential image regions guiding the network’s decisions. Various network architectures were trained and tested on 739 optic nerve head images, with nine saliency methods used. Some other popular datasets were also used for further validation. The results reveal disparities among saliency maps, with some consensus between the folds corresponding to the same architecture. Concerning the significance of optic disc sectors, there is generally a lack of agreement with standard medical criteria. The background, nasal, and temporal sectors emerge as particularly influential for neural network decisions, showing a likelihood of being the most relevant ranging from 14.55% to 28.16% on average across all evaluated datasets. We can conclude that saliency maps are usually difficult to interpret and even the areas indicated as the most relevant can be very unintuitive. Therefore, its usefulness as an explanatory tool may be compromised, at least in problems such as the one addressed in this study, where the features defining the model prediction are generally not consistently reflected in relevant regions of the saliency maps, and they even cannot always be related to those used as medical standards.

The automatic location of the fovea is very useful for diagnosing retinal diseases. It is a complex problem for which different solutions have been proposed based on classical image processing and Deep Learning techniques. The method presented in this paper is based on histograms that combine spatial and color information in such a way that the spatial coordinates are incorporated into conventional color histograms as an additional dimension. The binarization of these histograms retains a considerable amount of relevant information from the original image, allowing them to be processed as if they were ordinary images. This approach to the problem results in a simple, fast and effective method for locating the fovea. Different experiments have been carried out with three popular sets of images: Messidor, REFUGE1 and DIARETDB1, and a comparison was made with other state-of-the-art techniques. Our results show that the proposed method, despite its simplicity, is capable of surpassing many of these techniques.

The main goal of this paper is to show, from a technical point of view, how to implement the control of virtual industrial plants. These plants can be used in virtual laboratories for Engineering Education, specifically, in subjects related to Automation and Control. Two software tools are needed for the simulation of the plant: one for the graphical implementation of the model of the industrial plant (3D models), and the other for controlling the simulated movements of the elements of this 3D model. In this case, Autodesk Inventor and Simulink (of Matlab) have been the chosen tools. The control device can be simulated as well, or a real device could be used instead. In both cases a standard communication interface to interact with the virtual plant is needed. In case of using a real device, further connection hardware might be necessary.