Continuous Eye Disease Severity Evaluation System using Siamese Neural Networks

[摘要] Evaluating the severity of eye diseases using medical images is a very essential and routine taskperformed in medical diagnosis and treatment. Current grading systems which are largely basedon discrete classification are unreliable and do reflect not the entire spectrum of eye disease severity.The unreliability of discrete classification systems for eye diseases is clear, as classification is subjective and done based on the personal opinion of various medical experts, which may vary. In abid to solve these issues, this study proposes a system for determining the severity of eye diseaseson a continuous range using a twin-convoluted neural network approach known as Siamese NeuralNetworks. This system is demonstrated in the domain of diabetic retinopathy. Samples of retinalfundus images from an eye clinic in India are taken as test cases to evaluate the performance of aSiamese Triplet network which attempts to find the distance between their image embedding. Theoutputs of the Siamese network when a reference image is juxtaposed with a collection of imageswith distant severity categories (negative images), as well as when two reference images are compared to each other, are found to have a positive correlation (95%) with originally assigned severityclasses. Hence, these outputs indicate a continuous range of the severity and change in eye diseases.

[学科分类] 环境工程

[关键词] Eye Diseases;Siamese;Neural Networks;Convoluted Neural Network;Classification