Abstract

Optical coherence tomography is one of the key diagnostic methods used in ophthalmology and has a high potential for application in an automatic analysis. In this study, we collected, annotated, and analyzed 44 three-dimensional optical coherence tomography images obtained in 39 patients suffering from central serous chorioretinopathy. A semantic annotation of the pathological changes includes three classes: (1) retinal neuroepithelial detachment, (2) retinal pigment epithelium alteration, and (3) a leakage zone. Machine learning methods have been applied to distinguish classes 2 and 3 based on the brightness characteristics of optical coherent tomography images. Intra-group clustering of the class instances showed that the separation of two groups of changes in class 2 can be associated with differences in the volumetric characteristics, whereas the brightness characteristics in class 3 differ significantly depending on the age of the patients, which can be used to predict the course of the disease.

© 2020 Optical Society of America

PDF Article

References

You do not have subscription access to this journal. Citation lists with outbound citation links are available to subscribers only. You may subscribe either as an OSA member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access OSA Member Subscription

Cited By

You do not have subscription access to this journal. Cited by links are available to subscribers only. You may subscribe either as an OSA member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access OSA Member Subscription