Sie befinden Sich nicht im Netzwerk der Universität Paderborn. Der Zugriff auf elektronische Ressourcen ist gegebenenfalls nur via VPN oder Shibboleth (DFN-AAI) möglich. mehr Informationen...
Glaucoma is considered the leading cause of blindness. Because there is no cure for the disease, treatment must begin promptly to prevent disease progression, which leads to severe visual impairment and, in some cases, total blindness. In this scenario, diagnosis in the early stages is essential and could be accomplished through screening exams. Recent studies have combined fundus images and Optical Coherence Tomography (OCT) volumes as indicators of disease progression in computer vision methods. In this work, we develop a method based on deep learning, which uses fundus images and OCT volumes to aid in detecting glaucoma at early or progressive stages. In this way, it can have clinical use, being able to be used as a tool not only for the detection of the disease but also that it helps in searching for more severe cases of the disease. As a result, the proposed method achieves 0.886 kappa score.