A deep transfer learning framework for the automated assessment of corneal inflammation on in vivo confocal microscopy images

Abstract

The deep transfer learning models provide a completely automated analysis of corneal inflammatory cellular components with high accuracy. The implementation of such models would greatly benefit the management of corneal diseases and reduce workloads for ophthalmologists.