Automatic Recognition of Ocular Surface Diseases on Smartphone Images Using Densely Connected Convolutional Networks

Ocular surface disorder is one of common and prevalence eye diseases and complex to be recognized accurately. This work presents automatic classification of ocular surface disorders in accordance with densely connected convolutional networks and smartphone imaging. We use various smartphone cameras to collect clinical images that contain normal and abnormal, and modify end-to-end densely connected convolutional networks that use a hybrid unit to learn more diverse features, significantly reducing the network depth, the total number of parameters and the float calculation. The validation results demonstrate that our proposed method provides a promising and effective strategy to accurately screen ocular surface disorders. In particular, our deeply learned smartphone photographs based classification method achieved an average automatic recognition accuracy of 90.6%, while it is conveniently used by patients and integrated into smartphone applications for automatic patient-self screening ocular surface diseases without seeing a doctor in person in a hospital.

I. INTRODUCTION

Ocular surface disease (OSD) is a common eye disease related to eye surface structures including the cornea, conjunctiva and eyelids [1]. The prevalence of OSDs diagnosed on symptoms ranges from 7% to 52% worldwide [2]. Most of OSDs such as abnormal conjunctiva and cornea are commonly undiagnosed due to lack of standard description of symptoms and too many subtypes of OSD in practice. More unfortunately, OSD patients of all ages can develop photophobia, intermittent blurred vision, pain, limited ability to perform daily activities, and even depression in some cases [1], [3]. Therefore, it is important to accurately recognize OSDs at the early stage and treat them in an appropriate way without OSD exacerbation.