Detecting Intraocular Pressure Using CNN on Frontal Eye Images
Loading...
Authors
Rahmati, Afrooz
Aloudat, Mohammad
Abuzneid, Abdelshakour
Faezipour, Miad
Issue Date
2021-04-09
Type
Other
Language
en_US
Keywords
Convolutional neural network , Glaucoma , Intraocular pressure
Alternative Title
Abstract
Glaucoma is an international disease causing vision loss for many patients around the world A gradual increase of intraocular pressure (IOP) and missing early diagnoses might cause blindness forever Observing IOP requires the patient’s presence at a healthcare facility where ophthalmologists or nurses evaluate the eye pressure through different medical tests In some cases, the healthcare professional anesthetizes the eye by dropping a numb liquid which would take at least 6 hours to totally wear off from the eyes and irritate the patient We are proposing a novel technique applied on the patient’s frontal eye images using a convolutional neural network (CNN) to extract common features of high IOP and glaucoma cases automatically This is a research work in progress built upon our previous related work in which pre determined features were extracted from eye images to distinguish healthy eye images from high IOP cases The dataset used in this work contains 473 normal and high IOP eye images However, in order to increase our data accuracy, we are working closely with few hospitals in the Middle East The result of this study has the potential to minimize the patient’s presence at healthcare facilities and offer patients’ safety by preventing glaucoma causes at the very early stages.