On the analysis of local and global features for hyperemia grading

L. Sánchez; N. Barreira; N. Sánchez; A. Mosquera; H. Pena-Verdeal; E. Yebra-Pimentel

doi:10.1117/12.2268804

17 March 2017 On the analysis of local and global features for hyperemia grading

L. Sánchez, N. Barreira, N. Sánchez, A. Mosquera, H. Pena-Verdeal, E. Yebra-Pimentel

Proceedings Volume 10341, Ninth International Conference on Machine Vision (ICMV 2016); 103411T (2017) https://doi.org/10.1117/12.2268804
Event: Ninth International Conference on Machine Vision, 2016, Nice, France

Abstract

In optometry, hyperemia is the accumulation of blood flow in the conjunctival tissue. Dry eye syndrome or allergic conjunctivitis are two of its main causes. Its main symptom is the presence of a red hue in the eye that optometrists evaluate according to a scale in a subjective manner. In this paper, we propose an automatic approach to the problem of hyperemia grading in the bulbar conjunctiva. We compute several image features on images of the patients' eyes, analyse the relations among them by using feature selection techniques and transform the feature vector of each image to the value in the adequate range by means of machine learning techniques. We analyse different areas of the conjunctiva to evaluate their importance for the diagnosis. Our results show that it is possible to mimic the experts' behaviour through the proposed approach.

Citation Download Citation

L. Sánchez, N. Barreira, N. Sánchez, A. Mosquera, H. Pena-Verdeal, and E. Yebra-Pimentel "On the analysis of local and global features for hyperemia grading", Proc. SPIE 10341, Ninth International Conference on Machine Vision (ICMV 2016), 103411T (17 March 2017); https://doi.org/10.1117/12.2268804

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available