Microaneurysm Detection in Fundus Images Using a Two-step Convolutional Neural Networks
share
Diabetic Retinopathy (DR) is the prominent cause of blindness in the world. The early treatment of DR can be conducted from detection of microaneurysms (MA) which is reddish spots in retina images. Automated microaneurysm detection can be a helpful system for ophthalmologists for detection of MA. In this paper, deep learning, in particular convolutional neural network (CNN), is used as a powerful tool to detect MA efficiently. Our method used a new technique utilising of a two-stage training process which has a better efficiency and accuracy compared to previous works, while decreasing computational complexity. To validate our proposed method efficiency, an experiment is conducted using Keras library to implement our proposed CNN on two standard publicly available datasets. Our results show a promising sensitivity value of about 0.8 which is a competitive value with the state-of-the-art approaches. results of previous studies which is 0.5.
READ FULL TEXT