A systematic review of machine learning methods for eye illness detection

Velasquez R.M.A. · Proceedings of the 2022 IEEE 29th International Conference on Electronics, Electrical Engineering and Computing, INTERCON 2022 (2022) · DOI: 10.1109/intercon55795.2022.9870114

Resumen

The objective of this paper is the evaluation of different methodologies for the eye illness, especially the glaucoma detection through the application of deep learning, in order to obtain an early diagnosis of this disease based on images. For the training process, the comparison considers people older than forty years old. Therefore, the last research articles have quantitative analysis with an evaluation of the specific criteria such as proportionality of precision, sensitivity and specificity measurable at the patient level. Our findings are the best process and methodology to apply for future analysis techniques with tools dedicated to image processing, as is the case of U-net, SEUNte, Kaggle, among others; all associated to deep learning though a classification application and structured database, which allows better learning and making predictions of the patient's diseases and having a greater scope, reducing diagnosis time to give a better life quality.

Instituciones: Universidad Tecnológica del Perú