AbstractPeople having myopia are unaware about the progression of the disease until it is severe or untreatable. This can be avoided by regular checkup for the disease, but this increases the load on the professionals, and the time required for the results is also more, so an automated method using deep learning would be helpful in early detection of diseases. In order to reduce the time for detecting the disease, we propose a deep learning solution for automatic myopia detection. Our proposed model involves three models built by applying transfer learning on pretrained Xception, DenseNet201 and InceptionV3 models, respectively. Stack ensembling is used to ensemble the above three models. The ensemble model classifies the fundus images into two classes, i.e. pathological myopia and no pathological myopia, with an accuracy of 95.23%.KeywordsPathological myopiaDeep learningTransfer learningClassificationDetection of diseasesEnsemble

Full Text

Published Version
Open DOI Link

Get access to 115M+ research papers

Discover from 40M+ Open access, 2M+ Pre-prints, 9.5M Topics and 32K+ Journals.

Sign Up Now! It's FREE

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call