一种新的眼底图像视神经乳头分割方法

A Novel Segmentation Method for the Optic Nerve Head in the Retinal Images

视神经检查是减小青光眼致盲风险最为重要的临床诊断方法,医学研究发现视神经乳头形态参数是诊断人眼是否患病的主要指标。为实现视乳头形态参数的计算机辅助测量,本文提出了一种新的视神经乳头自动分割方法,基于极坐标空间形态学预处理和改进的随机游走自动分割算法实现视神经乳头的精确分割。采用临床视神经图像库进行实验,分割结果与眼科专家手工分割金标准比较,取得了96.52%的分割重叠精度,较之传统算法有较大提高,验证了该方法的有效性。

Early detection of glaucoma is essential to minimizing the risk of visual loss. It has been shown that a good predictor of glaucoma is the morphological parameters of the optic nerve head. To realize the computer-aided measurement of these parameters, this paper presents a novel, automated method to segment the optic nerve head based on the improved Random walks segmentation method and morphological pre-processing in polar coordination. In contrast with the state-of-art algorithms, the overlap accuracy is 96.52% acquired in the comparing segmentation results for our clinic retinal images database with the golden standards and verifies the effectiveness of the proposed algorithm.