Choroidal Neovascularisation Classification on Fundus Retinal Images Using Local Linear Estimator

9th Annual Basic Science International Conference 2019

自然科学（总论）

Puspitawati, A.^1 ; Chamidah, N.^1^2

Stud. of Study Program of Statistics, Department of Mathematics, Airlangga University, Surabaya, Indonesia^1

Department of Mathematics, Faculty of Sciences and Technology, Airlangga University, Surabaya, Indonesia^2

关键词: Abnormal growth;    Accuracy of classifications;    Choroidal neovascularization;    Important features;    Local linear estimator;    Retinal image;    Retinal pigment epithelium;    Statistical modeling;   

Others  :  https://iopscience.iop.org/article/10.1088/1757-899X/546/5/052056/pdf
DOI  :  10.1088/1757-899X/546/5/052056

学科分类：自然科学（综合）

来源: IOP

【 摘 要 】

Choroidal neovascularization describes the abnormal growth of vessels from the choriocapillaris through the Bruch membrane into the space beneath the retinal pigment epithelium or the space beneath the retina. Choroidal neovascularization detection is considered to be the most important feature in the pathogenesis and treatment of a number of chorioretinal disorders, one of which to detection in digital fundus retinal images. In this study, we classify choroidal neovascularization by using statistical modeling approach based on local linear estimator. Based on 40 in samples and 10 out samples data images, we obtain the same accuracy of classification of 90 percent and their sensitivity are 90 percent and 83.33 percent, respectively. So, we conclude that the local linear is a good estimator to classify choroidal neovascularization on fundus retinal image.