Detection of hard exudates in fundus images using convolutional neural networks

• Main Authors: Ittided Poonkasem, Nipon Theera-Umpon, Sansanee Auephanwiriyakul, Direk Patikulsila
• Format: Conference Proceeding
• Published: 2020
• Subjects: Computer Science; Decision Sciences; Energy; Physics and Astronomy
• Online Access: https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85074277142&origin=inward; http://cmuir.cmu.ac.th/jspui/handle/6653943832/67756
• Institution: Chiang Mai University

© 2019 IEEE. The patients with diabetes have a chance to have blindness. An impairment of metabolism can cause a high glucose level in blood vessel leading to an abnormality called hard exudates. Hard exudates are often arranged in clumps or circinate rings and located in the outer layer of the retina. The aim of this research is to detect hard exudates by applying image processing techniques and classify them by using convolutional neuron network (CNN). DIARETDB1 dataset is used in the experiments. The proposed method achieves the area under the curve (AUC) of 0.97 and 0.95 on the training and validation sets, respectively, of 10-fold cross validation experiment. These show that the combination of image processing techniques, three channels of fundus images, and CNN can perform as a promising classification tool in hard exudates detection system.