RETINA FUNDUS IMAGE MASK GENERATION USING PSEUDO PARAMETRIC MODELING TECHNIQUE

Abstract

The use of vascular intersection as one of the symptoms for monitoring and diagnosis of diabetic retinopathy from fundus images have been widely reported in literatures. In this work, a new hybrid approach that makes use of three different methods of vascular intersection detection namely Modified cross-point number (MCN), Combine Cross Points (CNN) and Artificial Neural Network (ANN) is hereby proposed. Result obtained from the application of this technique to both simulated and experimental shows a very high accuracy and precision value in detecting both bifurcation and cross over points. Thus an improvement in bifurcation and vascular point detection and a good tool in the monitoring and diagnosis of diabetic retinopathy.