A multiscale decomposition approach to detect abnormal vasculature in the optic disc

• Autores: Agurto, Carla, Yub, Honggang, Murray, Victor, Pattichis, Marios, Nemeth, Sheila, Barriga, Simon, Soliz, Peter
• Formato: artículo
• Fecha de Publicación: 2015
• Institución: Universidad de Ingeniería y tecnología
• Repositorio: UTEC-Institucional
• Lenguaje: inglés
• OAI Identifier: oai:repositorio.utec.edu.pe:20.500.12815/37
• Enlace del recurso: https://hdl.handle.net/20.500.12815/37; https://doi.org/10.1016/j.compmedimag.2015.01.001
• Nivel de acceso: acceso abierto
• Materia: Amplitude-modulation frequency-modulation; Diabetic retinopathy; Fractal dimension; Granulometry; Neovascularization

This paper presents a multiscale method to detect neovascularization in the optic disc (NVD) using fundus images. Our method is applied to a manually selected region of interest (ROI) containing the optic disc. All the vessels in the ROI are segmented by adaptively combining contrast enhancement methods with a vessel segmentation technique. Textural features extracted using multiscale amplitude-modulation frequency-modulation, morphological granulometry, and fractal dimension are used. A linear SVM is used to perform the classification, which is tested by means of 10-fold cross-validation. The performance is evaluated using 300 images achieving an AUC of 0.93 with maximum accuracy of 88%.