研究目的
To design an automated grading approach for DR screening, using a publicly available database of retinal images, which can evaluate the fundus images like human experts while achieving a high sensitivity for the detection of DR.
研究成果
The proposed automated diagnosis approach for analyzing the retinopathy grade and the risk of macular edema is comparable to the human experts and seems to be useful for clinical interventions.
研究不足
The proposed approach just focused on detection of the earliest signs of DR, without a differentiation between different retinopathy grades.
1：Experimental Design and Method Selection:
The approach consists of various modules responsible for tasks such as optic disc and macula detection, red lesions detection, and bright lesions detection.
2：Sample Selection and Data Sources:
The publicly available Messidor database is used, which contains 1,200 color fundus images with three different resolutions.
3：List of Experimental Equipment and Materials:
A color video 3CCD camera on a Topcon TRC NW6 non-mydriatic retinograph with a 45-degree field of view was used for image acquisition.
4：Experimental Procedures and Operational Workflow:
The proposed approach consists of four stages: resizing the retinal images, locating the optic disc, blood vessels and macula regions, recognizing signs of DR, and evaluating the fundus images automatically.
5：Data Analysis Methods:
The performance of the proposed approach is measured using sensitivity and specificity at the image level.
独家科研数据包，助您复现前沿成果，加速创新突破
获取完整内容
