In this paper a contribution to diabetic damage detection in retinal images via a cellular neurofuzzy network is proposed. Fundus symptomatic pale regions are firstly highlighted by enhancing image contrast with a neurofuzzy subnet, which is synthesized using a Cellular Neural Network (CNN). After an optimal thresholding performed by a neural subsystem, obtained contrast-enhanced images with bimodal histograms are globally segmented. In output binary images, suspect diabetic areas are then isolated by a CNN-based subnet. Performances are evaluated by percentage measures of exactness in the detection of suspect damaged areas via a comparison with gold standard images provided by clinicians. Results are discussed and compared with other researcherspsila ones.
|Titolo:||Diabetic Damage Detection in Retinal Images via a Cellular Neurofuzzy Network|
|Data di pubblicazione:||2006|
|Nome del convegno:||Biomedical Circuits and Systems Conference Healthcare Technology, BioCAS 2006|
|Digital Object Identifier (DOI):||10.1109/BIOCAS.2006.4600327|
|Appare nelle tipologie:||4.1 Contributo in Atti di convegno|