Open AccessResearch on Assistant Diagnosis of Fundus Retinopathy Based on Deep Learning
Author(s) -
Haoyu Li,
Ziwen Yuan,
Kun Zhang
Publication year - 2020
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1650/3/032197
Subject(s) - fundus (uterus) , artificial intelligence , segmentation , computer science , macular edema , deep learning , artificial neural network , field (mathematics) , key (lock) , computer vision , pattern recognition (psychology) , medicine , ophthalmology , retinal , mathematics , computer security , pure mathematics
Macular edema has three types of lesions: REA, PED and SRF. Early detection of edema areas can play a key role in the treatment of diseases. Neural network is a powerful tool for image processing in medical field. Deep learning automatically finds features that are ideal for “AI+Medical Imaging” diagnostics. This paper mainly proposes a new method of neural network that includes Object recognition and classification to improve the accuracy and speed of detection of macular edema area. The method is offered to be evaluated and compared to the traditional network method. The results indicate that the method applied reducing the over segmentation effect and getting a more accurate result of the option than traditional network method