摘要
目前检测糖尿病视网膜病变的方法主要依赖于医生的个人判断,其诊断时间较长,占用较多的医疗资源。为解决以上问题,提出了一种基于改进的AlexNet模型的深度学习方法,对彩色眼底图像进行特征提取,自动判断糖尿病视网膜病变程度,达到了96.64%的诊断精度。
At present, the detection methods of diabetic retinopathy mainly depend on doctors’ personal judgment, which takes a long time to diagnose and occupies more medical resources. In order to solve the above problems, a deep learning method based on improved AlexNet model is proposed to extract the features of color funds images and automatically judge the degree of diabetic retinopathy, which achieves 96.64% diagnostic accuracy.
出处
《信息技术与信息化》
2019年第7期162-164,共3页
Information Technology and Informatization
