Computer-aided Diagnosis of Glaucoma Using Fundus Images

摘要

Glaucoma is a chronic eye disease which cannot be cured, so that detecting the disease in time is important. Machine learning for glaucoma diagnosis has achieved great development in recent years. In this paper, we present an algorithm for glaucoma diagnosis from optic disc and optic cup boundary lines in fundus images based on doctors' knowledge. We do meticulous division, scaling transformation and principal component analysis on the optic disc and optic cup boundary lines to extract features. The extracted features correspond well with doctors' knowledge. Therefore, we can make an intuitive explanation for the diagnosis results to doctors, rather than just as a black-box prediction. On a real sample set, the proposed feature extraction and diagnosis algorithms achieve high prediction accuracy.