Diabeticretinopathy detecting method and device based on deep neural network

Abstract

The embodiment of the invention provides a diabeticretinopathy detecting method and device based on a deep neural network, and belongs to the field of image processing. The method comprises the stepsof collecting data of multiple fundus photograph images; inputting the data of the fundus photograph images to a preset deep neural network model; obtaining the classification result output by the deep neural network model and confidence coefficient corresponding to the classification result; judging whether the classification result is valid of not; if yes, looking up the diabeticretinopathy degree corresponding to the classification result, and outputting a detecting result based on the diabeticretinopathy degree. Thus a user can obtain the detection result simply by uploading the fundus images, and the effects that convenience is brought to the user and the cost is low are achieved while the accuracy is effectively improved; meanwhile, under the condition of the absence or lack of professional physicians, the diagnosis efficiency can be greatly improved, and precious time is saved for a patient.

Description

technical field [0001] The present invention relates to the field of image processing, in particular, to a method and device for detecting diabetic retinopathy based on a deep neural network. Background technique [0002] Diabetic retinopathy (diabetic retinopathy, DR) is the fastest growing cause of blindness in the world, and it is one of the complications of diabetic patients. Every patient is likely to become blind due to diabetic retinopathy. According to the "Diabetes Map" data released by the International Diabetes Federation IDF, in 2015, about 415 million people aged 20-79 worldwide suffered from diabetes. The number of diabetic patients in my country ranks first in the world, and the incidence of diabetes is showing a trend of rapid increase and younger age. [0003] Diabetic retinopathy is asymptomatic until irreversible visual impairment occurs. At present, the main means of preventing blindness is regular inspection. Through the use of special equipment, fundu...

AI Technical Summary

Problems solved by technology

However, in many places there are not enough doctors to do the work, and almost half of all patients go blind before they are diagnosed

Images